docs: fix active memory gateway command

Merged via squash.

Prepared head SHA: dfe6a3581c
Co-authored-by: balajisiva <13068516+balajisiva@users.noreply.github.com>
Co-authored-by: scoootscooob <167050519+scoootscooob@users.noreply.github.com>
Reviewed-by: @scoootscooob

.agents/skills/openclaw-parallels-smoke/SKILL.md

@@ -30,9 +30,12 @@ Use this skill for Parallels guest workflows and smoke interpretation. Do not lo
 
 - Preferred entrypoint: `pnpm test:parallels:npm-update`
 - Flow: fresh snapshot -> install npm package baseline -> smoke -> install current main tgz on the same guest -> smoke again.
+- For beta/stable verification, resolve the tag immediately before the run (`npm view openclaw@beta version dist.tarball` or `npm view openclaw@latest ...`). Tags can move while a long VM matrix is already running; restart the matrix when the intended prerelease appears after an earlier registry 404/tag-lag check.
+- Source Peter's profile in the host shell (`set -a; source "$HOME/.profile"; set +a`) before OpenAI/Anthropic lanes. Do not print profile contents or env dumps; pass provider secrets through the guest exec environment.
 - Same-guest update verification should set the default model explicitly to `openai/gpt-5.4` before the agent turn and use a fresh explicit `--session-id` so old session model state does not leak into the check.
 - The aggregate npm-update wrapper must resolve the Linux VM with the same Ubuntu fallback policy as `parallels-linux-smoke.sh` before both fresh and update lanes. Treat any Ubuntu guest with major version `>= 24` as acceptable when the exact default VM is missing, preferring the closest version match. On Peter's current host today, missing `Ubuntu 24.04.3 ARM64` should fall back to `Ubuntu 25.10`.
 - On macOS same-guest update checks, restart the gateway after the npm upgrade before `gateway status` / `agent`; launchd can otherwise report a loaded service while the old process has exited and the fresh process is not RPC-ready yet.
+- The npm-update aggregate's macOS update leg writes the guest update script as root, then runs it as the desktop user. If `prlctl exec "$MACOS_VM" --current-user ...` cannot authenticate, retry through plain root `prlctl exec` plus `sudo -u <desktop-user> /usr/bin/env HOME=/Users/<desktop-user> USER=<desktop-user> LOGNAME=<desktop-user> PATH=/opt/homebrew/bin:/opt/homebrew/opt/node/bin:/usr/bin:/bin:/usr/sbin:/sbin ...`. That is a Parallels transport fallback; still verify `openclaw --version`, gateway RPC, and an agent turn after the update.
 - On Windows same-guest update checks, restart the gateway after the npm upgrade before `gateway status` / `agent`; in-place global npm updates can otherwise leave stale hashed `dist/*` module imports alive in the running service.
 - In those Windows same-guest update checks, do not treat one nonzero `openclaw gateway restart` as definitive failure. Current login-item restarts can report failure before the background service becomes observable again; follow with a longer RPC-ready wait and use `gateway start` only as a recovery step if readiness still never returns.
 - After that Windows restart, do not trust one `gateway status --deep --require-rpc` call after a fixed sleep. Retry the RPC-ready probe for roughly 30 seconds and log each attempt; current guests can keep port `18789` bound while the fresh RPC endpoint is still coming up.
@@ -41,6 +44,7 @@ Use this skill for Parallels guest workflows and smoke interpretation. Do not lo
 - Linux same-guest update verification should also export `HOME=/root`, pass `OPENAI_API_KEY` via `prlctl exec ... /usr/bin/env`, and use `openclaw agent --local`; the fresh Linux baseline does not rely on persisted gateway credentials.
 - The npm-update wrapper now prints per-lane progress from the nested log files. If a lane still looks stuck, inspect the nested logs in `runDir` first (`macos-fresh.log`, `windows-fresh.log`, `linux-fresh.log`, `macos-update.log`, `windows-update.log`, `linux-update.log`) instead of assuming the outer wrapper hung.
 - If the wrapper fails a lane, read the auto-dumped tail first, then the full nested lane log under `/tmp/openclaw-parallels-npm-update.*`.
+- Current known macOS update-lane transport signature when the fallback is missing or bypassed: `Unable to authenticate the user. Make sure that the specified credentials are correct and try again.` Treat that as Parallels current-user authentication before blaming npm or OpenClaw.
 
 ## CLI invocation footgun
 
@@ -64,6 +68,7 @@ Use this skill for Parallels guest workflows and smoke interpretation. Do not lo
 - If a packaged install regresses with `500` on `/`, `/healthz`, or `__openclaw/control-ui-config.json` after `fresh.install-main` or `upgrade.install-main`, suspect bundled plugin runtime deps resolving from the package root `node_modules` rather than `dist/extensions/*/node_modules`. Repro quickly with a real `npm pack`/global install lane before blaming dashboard auth or Safari.
 - `prlctl exec` is fine for deterministic repo commands, but use the guest Terminal or `prlctl enter` when installer parity or shell-sensitive behavior matters.
 - Multi-word `openclaw agent --message ...` checks should go through a guest shell wrapper (`guest_current_user_sh` / `guest_current_user_cli` or `/bin/sh -lc ...`), not raw `prlctl exec ... node openclaw.mjs ...`, or the message can be split into extra argv tokens and Commander reports `too many arguments for 'agent'`.
+- The same wrapper rule applies when bypassing `--current-user`: write a tiny `/tmp/*.sh` on the guest and execute `/bin/bash /tmp/*.sh` through the sudo desktop-user environment. Do not pass `openclaw agent --message '...'` directly as one raw `prlctl exec` command.
 - When ref-mode onboarding stores `OPENAI_API_KEY` as an env secret ref, the post-onboard agent verification should also export `OPENAI_API_KEY` for the guest command. The gateway can still reject with pairing-required and fall back to embedded execution, and that fallback needs the env-backed credential available in the shell.
 - On the fresh Tahoe snapshot, `brew` exists but `node` may be missing from PATH in noninteractive exec. Use `/opt/homebrew/bin/node` when needed.
 - Fresh host-served tgz installs should install as guest root with `HOME=/var/root`, then run onboarding as the desktop user via `prlctl exec --current-user`.

.agents/skills/openclaw-qa-testing/SKILL.md

@@ -57,31 +57,28 @@ Use `qa character-eval` for style/persona/vibe checks across multiple live model
 pnpm openclaw qa character-eval \
   --model openai/gpt-5.4,thinking=xhigh \
   --model openai/gpt-5.2,thinking=xhigh \
+  --model openai/gpt-5,thinking=xhigh \
   --model anthropic/claude-opus-4-6,thinking=high \
   --model anthropic/claude-sonnet-4-6,thinking=high \
-  --model minimax/MiniMax-M2.7,thinking=high \
   --model zai/glm-5.1,thinking=high \
   --model moonshot/kimi-k2.5,thinking=high \
-  --model qwen/qwen3.6-plus,thinking=high \
-  --model xiaomi/mimo-v2-pro,thinking=high \
   --model google/gemini-3.1-pro-preview,thinking=high \
-  --model codex-cli/<codex-model>,thinking=high \
   --judge-model openai/gpt-5.4,thinking=xhigh,fast \
   --judge-model anthropic/claude-opus-4-6,thinking=high \
-  --concurrency 8 \
-  --judge-concurrency 8 \
+  --concurrency 16 \
+  --judge-concurrency 16 \
   --output-dir .artifacts/qa-e2e/character-eval-<tag>
 ```
 
 - Runs local QA gateway child processes, not Docker.
 - Preferred model spec syntax is `provider/model,thinking=<level>[,fast|,no-fast|,fast=<bool>]` for both `--model` and `--judge-model`.
 - Do not add new examples with separate `--model-thinking`; keep that flag as legacy compatibility only.
-- Defaults to candidate models `openai/gpt-5.4`, `openai/gpt-5.2`, `anthropic/claude-opus-4-6`, `anthropic/claude-sonnet-4-6`, `minimax/MiniMax-M2.7`, `zai/glm-5.1`, `moonshot/kimi-k2.5`, `qwen/qwen3.6-plus`, `xiaomi/mimo-v2-pro`, and `google/gemini-3.1-pro-preview` when no `--model` is passed.
+- Defaults to candidate models `openai/gpt-5.4`, `openai/gpt-5.2`, `openai/gpt-5`, `anthropic/claude-opus-4-6`, `anthropic/claude-sonnet-4-6`, `zai/glm-5.1`, `moonshot/kimi-k2.5`, and `google/gemini-3.1-pro-preview` when no `--model` is passed.
 - Candidate thinking defaults to `high`, with `xhigh` for OpenAI models that support it. Prefer inline `--model provider/model,thinking=<level>`; `--thinking <level>` and `--model-thinking <provider/model=level>` remain compatibility shims.
 - OpenAI candidate refs default to fast mode so priority processing is used where supported. Use inline `,fast`, `,no-fast`, or `,fast=false` for one model; use `--fast` only to force fast mode for every candidate.
 - Judges default to `openai/gpt-5.4,thinking=xhigh,fast` and `anthropic/claude-opus-4-6,thinking=high`.
 - Report includes judge ranking, run stats, durations, and full transcripts; do not include raw judge replies. Duration is benchmark context, not a grading signal.
-- Candidate and judge concurrency default to 8. Use `--concurrency <n>` and `--judge-concurrency <n>` to override when local gateways or provider limits need a gentler lane.
+- Candidate and judge concurrency default to 16. Use `--concurrency <n>` and `--judge-concurrency <n>` to override when local gateways or provider limits need a gentler lane.
 - Scenario source should stay markdown-driven under `qa/scenarios/`.
 - For isolated character/persona evals, write the persona into `SOUL.md` and blank `IDENTITY.md` in the scenario flow. Use `SOUL.md + IDENTITY.md` only when intentionally testing how the normal OpenClaw identity combines with the character.
 - Keep prompts natural and task-shaped. The candidate model should receive character setup through `SOUL.md`, then normal user turns such as chat, workspace help, and small file tasks; do not ask "how would you react?" or tell the model it is in an eval.

.github/labeler.yml

@@ -297,6 +297,10 @@
   - changed-files:
       - any-glob-to-any-file:
           - "extensions/openai/**"
+"extensions: codex":
+  - changed-files:
+      - any-glob-to-any-file:
+          - "extensions/codex/**"
 "extensions: kimi-coding":
   - changed-files:
       - any-glob-to-any-file:

.github/workflows/ci.yml

@@ -37,9 +37,10 @@ jobs:
       run_build_artifacts: ${{ steps.manifest.outputs.run_build_artifacts }}
       run_checks_fast: ${{ steps.manifest.outputs.run_checks_fast }}
       checks_fast_core_matrix: ${{ steps.manifest.outputs.checks_fast_core_matrix }}
-      checks_fast_extensions_matrix: ${{ steps.manifest.outputs.checks_fast_extensions_matrix }}
+      checks_node_extensions_matrix: ${{ steps.manifest.outputs.checks_node_extensions_matrix }}
       run_checks: ${{ steps.manifest.outputs.run_checks }}
       checks_matrix: ${{ steps.manifest.outputs.checks_matrix }}
+      checks_node_core_test_matrix: ${{ steps.manifest.outputs.checks_node_core_test_matrix }}
       run_extension_fast: ${{ steps.manifest.outputs.run_extension_fast }}
       extension_fast_matrix: ${{ steps.manifest.outputs.extension_fast_matrix }}
       run_check: ${{ steps.manifest.outputs.run_check }}
@@ -135,6 +136,9 @@ jobs:
         run: |
           node --input-type=module <<'EOF'
           import { appendFileSync } from "node:fs";
+          import {
+            createNodeTestShards,
+          } from "./scripts/lib/ci-node-test-plan.mjs";
           import {
             createExtensionTestShards,
             DEFAULT_EXTENSION_TEST_SHARD_COUNT,
@@ -211,12 +215,11 @@ jobs:
                   ]
                 : [],
             ),
-            checks_fast_extensions_matrix: extensionShardMatrix,
+            checks_node_extensions_matrix: extensionShardMatrix,
             run_checks: runNode,
             checks_matrix: createMatrix(
               runNode
                 ? [
-                    { check_name: "checks-node-test", runtime: "node", task: "test" },
                     { check_name: "checks-node-channels", runtime: "node", task: "channels" },
                     ...(isPush
                       ? [
@@ -232,6 +235,17 @@ jobs:
                   ]
                 : [],
             ),
+            checks_node_core_test_matrix: createMatrix(
+              runNode
+                ? createNodeTestShards().map((shard) => ({
+                    check_name: shard.checkName,
+                    runtime: "node",
+                    task: "test-shard",
+                    shard_name: shard.shardName,
+                    configs: shard.configs,
+                  }))
+                : [],
+            ),
             run_extension_fast: hasChangedExtensions,
             extension_fast_matrix: createMatrix(
               hasChangedExtensions
@@ -470,7 +484,7 @@ jobs:
               ;;
           esac
 
-  checks-fast-extensions-shard:
+  checks-node-extensions-shard:
     name: ${{ matrix.check_name }}
     needs: [preflight]
     if: needs.preflight.outputs.run_checks_fast == 'true'
@@ -478,7 +492,7 @@ jobs:
     timeout-minutes: 60
     strategy:
       fail-fast: false
-      matrix: ${{ fromJson(needs.preflight.outputs.checks_fast_extensions_matrix) }}
+      matrix: ${{ fromJson(needs.preflight.outputs.checks_node_extensions_matrix) }}
     steps:
       - name: Checkout
         uses: actions/checkout@v6
@@ -497,16 +511,16 @@ jobs:
           OPENCLAW_EXTENSION_BATCH: ${{ matrix.extensions_csv }}
         run: pnpm test:extensions:batch -- "$OPENCLAW_EXTENSION_BATCH"
 
-  checks-fast-extensions:
-    name: checks-fast-extensions
-    needs: [preflight, checks-fast-extensions-shard]
+  checks-node-extensions:
+    name: checks-node-extensions
+    needs: [preflight, checks-node-extensions-shard]
     if: always() && needs.preflight.outputs.run_checks_fast == 'true'
     runs-on: blacksmith-16vcpu-ubuntu-2404
     timeout-minutes: 5
     steps:
       - name: Verify extension shards
         env:
-          SHARD_RESULT: ${{ needs.checks-fast-extensions-shard.result }}
+          SHARD_RESULT: ${{ needs.checks-node-extensions-shard.result }}
         run: |
           if [ "$SHARD_RESULT" != "success" ]; then
             echo "Extension shard checks failed: $SHARD_RESULT" >&2
@@ -599,6 +613,102 @@ jobs:
               ;;
           esac
 
+  checks-node-core-test-shard:
+    name: ${{ matrix.check_name }}
+    needs: [preflight, build-artifacts]
+    if: always() && needs.preflight.outputs.run_checks == 'true' && needs.build-artifacts.result == 'success'
+    runs-on: blacksmith-16vcpu-ubuntu-2404
+    timeout-minutes: 60
+    strategy:
+      fail-fast: false
+      matrix: ${{ fromJson(needs.preflight.outputs.checks_node_core_test_matrix) }}
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v6
+        with:
+          persist-credentials: false
+          submodules: false
+
+      - name: Setup Node environment
+        uses: ./.github/actions/setup-node-env
+        with:
+          node-version: "${{ matrix.node_version || '24.x' }}"
+          cache-key-suffix: "${{ matrix.cache_key_suffix || 'node24' }}"
+          install-bun: "false"
+          use-sticky-disk: "false"
+
+      - name: Configure Node test resources
+        run: echo "OPENCLAW_VITEST_MAX_WORKERS=2" >> "$GITHUB_ENV"
+
+      - name: Download dist artifact
+        uses: actions/download-artifact@v8
+        with:
+          name: dist-build
+          path: dist/
+
+      - name: Download A2UI bundle artifact
+        uses: actions/download-artifact@v8
+        with:
+          name: canvas-a2ui-bundle
+          path: src/canvas-host/a2ui/
+
+      - name: Run Node test shard
+        env:
+          NODE_OPTIONS: --max-old-space-size=6144
+          OPENCLAW_NODE_TEST_CONFIGS_JSON: ${{ toJson(matrix.configs) }}
+        shell: bash
+        run: |
+          set -euo pipefail
+          node --input-type=module <<'EOF'
+          import { spawnSync } from "node:child_process";
+          import { resolveVitestCliEntry, resolveVitestNodeArgs } from "./scripts/run-vitest.mjs";
+
+          const configs = JSON.parse(process.env.OPENCLAW_NODE_TEST_CONFIGS_JSON ?? "[]");
+          if (!Array.isArray(configs) || configs.length === 0) {
+            console.error("Missing node test shard configs");
+            process.exit(1);
+          }
+
+          for (const config of configs) {
+            console.error(`[test] starting ${config}`);
+            const result = spawnSync(
+              "pnpm",
+              [
+                "exec",
+                "node",
+                ...resolveVitestNodeArgs(process.env),
+                resolveVitestCliEntry(),
+                "run",
+                "--config",
+                config,
+              ],
+              {
+                env: process.env,
+                stdio: "inherit",
+              },
+            );
+            if ((result.status ?? 1) !== 0) {
+              process.exit(result.status ?? 1);
+            }
+          }
+          EOF
+
+  checks-node-core-test:
+    name: checks-node-core-test
+    needs: [preflight, checks-node-core-test-shard]
+    if: always() && needs.preflight.outputs.run_checks == 'true'
+    runs-on: blacksmith-16vcpu-ubuntu-2404
+    timeout-minutes: 5
+    steps:
+      - name: Verify node test shards
+        env:
+          SHARD_RESULT: ${{ needs.checks-node-core-test-shard.result }}
+        run: |
+          if [ "$SHARD_RESULT" != "success" ]; then
+            echo "Node test shards failed: $SHARD_RESULT" >&2
+            exit 1
+          fi
+
   extension-fast:
     name: "extension-fast"
     needs: [preflight]
@@ -779,6 +889,11 @@ jobs:
         continue-on-error: true
         run: pnpm test:gateway:watch-regression
 
+      - name: Run import cycle guard
+        id: import_cycles
+        continue-on-error: true
+        run: pnpm check:import-cycles
+
       - name: Upload gateway watch regression artifacts
         if: always()
         uses: actions/upload-artifact@v7
@@ -811,6 +926,7 @@ jobs:
           NO_RAW_WINDOW_OPEN_OUTCOME: ${{ steps.no_raw_window_open.outcome }}
           CONTROL_UI_I18N_OUTCOME: ${{ steps.control_ui_i18n.outcome == 'skipped' && 'success' || steps.control_ui_i18n.outcome }}
           GATEWAY_WATCH_REGRESSION_OUTCOME: ${{ steps.gateway_watch_regression.outcome }}
+          IMPORT_CYCLES_OUTCOME: ${{ steps.import_cycles.outcome }}
         run: |
           failures=0
           for result in \
@@ -834,7 +950,8 @@ jobs:
             "test:extensions:package-boundary|$EXTENSION_PACKAGE_BOUNDARY_TSC_OUTCOME" \
             "lint:ui:no-raw-window-open|$NO_RAW_WINDOW_OPEN_OUTCOME" \
             "ui:i18n:check|$CONTROL_UI_I18N_OUTCOME" \
-            "gateway-watch-regression|$GATEWAY_WATCH_REGRESSION_OUTCOME"; do
+            "gateway-watch-regression|$GATEWAY_WATCH_REGRESSION_OUTCOME" \
+            "check:import-cycles|$IMPORT_CYCLES_OUTCOME"; do
             name="${result%%|*}"
             outcome="${result#*|}"
             if [ "$outcome" != "success" ]; then
@@ -1042,7 +1159,9 @@ jobs:
           set -euo pipefail
           case "$TASK" in
             test)
-              pnpm test
+              # Linux owns the full repo test suite. Keep the Windows runner focused on
+              # Windows-native process/path wrappers so platform regressions fail fast.
+              pnpm test:windows:ci
               ;;
             *)
               echo "Unsupported Windows checks task: $TASK" >&2
@@ -1093,7 +1212,9 @@ jobs:
           set -euo pipefail
           case "$TASK" in
             test)
-              pnpm test
+              # Linux owns the full repo test suite. Keep macOS CI focused on
+              # launchd/Homebrew/runtime path coverage and the process-group wrapper.
+              pnpm test:macos:ci
               ;;
             *)
               echo "Unsupported macOS node task: $TASK" >&2

.github/workflows/openclaw-npm-release.yml

@@ -162,9 +162,63 @@ jobs:
           RELEASE_NPM_DIST_TAG: ${{ inputs.npm_dist_tag }}
         run: |
           set -euo pipefail
-          PACK_JSON="$(npm pack --json)"
-          echo "$PACK_JSON"
-          PACK_PATH="$(printf '%s\n' "$PACK_JSON" | node -e 'const chunks=[]; process.stdin.on("data", (chunk) => chunks.push(chunk)); process.stdin.on("end", () => { const parsed = JSON.parse(Buffer.concat(chunks).toString("utf8")); const first = Array.isArray(parsed) ? parsed[0] : null; if (!first || typeof first.filename !== "string" || !first.filename) { process.exit(1); } process.stdout.write(first.filename); });')"
+          PACK_OUTPUT="$RUNNER_TEMP/npm-pack-output.txt"
+          npm pack --json 2>&1 | tee "$PACK_OUTPUT"
+          PACK_PATH="$(node - "$PACK_OUTPUT" <<'NODE'
+          const fs = require("node:fs");
+          const input = fs.readFileSync(process.argv[2], "utf8");
+
+          function arrayEndFrom(start) {
+            let depth = 0;
+            let inString = false;
+            let escape = false;
+            for (let i = start; i < input.length; i += 1) {
+              const char = input[i];
+              if (inString) {
+                if (escape) {
+                  escape = false;
+                } else if (char === "\\") {
+                  escape = true;
+                } else if (char === "\"") {
+                  inString = false;
+                }
+                continue;
+              }
+              if (char === "\"") {
+                inString = true;
+              } else if (char === "[") {
+                depth += 1;
+              } else if (char === "]") {
+                depth -= 1;
+                if (depth === 0) {
+                  return i + 1;
+                }
+              }
+            }
+            return -1;
+          }
+
+          for (let start = input.indexOf("["); start !== -1; start = input.indexOf("[", start + 1)) {
+            const end = arrayEndFrom(start);
+            if (end === -1) {
+              continue;
+            }
+            try {
+              const parsed = JSON.parse(input.slice(start, end));
+              const first = Array.isArray(parsed) ? parsed[0] : null;
+              if (first && typeof first.filename === "string" && first.filename) {
+                process.stdout.write(first.filename);
+                process.exit(0);
+              }
+            } catch {
+              // Keep scanning; npm lifecycle output can legally precede the JSON.
+            }
+          }
+
+          console.error("Could not find npm pack --json output with a filename.");
+          process.exit(1);
+          NODE
+          )"
           if [[ -z "$PACK_PATH" || ! -f "$PACK_PATH" ]]; then
             echo "npm pack did not produce a tarball file." >&2
             exit 1

.oxfmtrc.jsonc

@@ -1,9 +1,9 @@
 {
   "$schema": "./node_modules/oxfmt/configuration_schema.json",
-  "experimentalSortImports": {
+  "sortImports": {
     "newlinesBetween": false,
   },
-  "experimentalSortPackageJson": {
+  "sortPackageJson": {
     "sortScripts": true,
   },
   "tabWidth": 2,

.oxlintrc.json

@@ -8,19 +8,23 @@
   },
   "rules": {
     "curly": "error",
-    "eslint-plugin-unicorn/prefer-array-find": "off",
+    "eslint-plugin-unicorn/prefer-array-find": "error",
     "eslint/no-await-in-loop": "off",
-    "eslint/no-new": "off",
+    "eslint/no-new": "error",
     "eslint/no-shadow": "off",
-    "eslint/no-unmodified-loop-condition": "off",
-    "oxc/no-accumulating-spread": "off",
+    "eslint/no-unmodified-loop-condition": "error",
+    "eslint-plugin-unicorn/prefer-set-size": "error",
+    "oxc/no-accumulating-spread": "error",
     "oxc/no-async-endpoint-handlers": "off",
     "oxc/no-map-spread": "off",
+    "typescript/consistent-return": "error",
     "typescript/no-explicit-any": "error",
-    "typescript/no-extraneous-class": "off",
+    "typescript/no-extraneous-class": "error",
+    "typescript/no-unnecessary-type-conversion": "error",
     "typescript/no-unsafe-type-assertion": "off",
     "unicorn/consistent-function-scoping": "off",
-    "unicorn/require-post-message-target-origin": "off"
+    "unicorn/prefer-set-size": "error",
+    "unicorn/require-post-message-target-origin": "error"
   },
   "ignorePatterns": [
     "assets/",
@@ -54,13 +58,7 @@
         "**/*test-support.ts"
       ],
       "rules": {
-        "typescript/await-thenable": "off",
-        "typescript/no-base-to-string": "off",
         "typescript/no-explicit-any": "off",
-        "typescript/no-floating-promises": "off",
-        "typescript/no-misused-spread": "off",
-        "typescript/no-redundant-type-constituents": "off",
-        "typescript/no-unnecessary-template-expression": "off",
         "typescript/unbound-method": "off",
         "eslint/no-unsafe-optional-chaining": "off"
       }

AGENTS.md

@@ -73,6 +73,8 @@
 - Extension test boundary:
   - Keep extension-owned onboarding/config/provider coverage under the owning bundled plugin package when feasible.
   - If core tests need bundled plugin behavior, consume it through public `src/plugin-sdk/<id>.ts` facades or the plugin's `api.ts`, not private extension modules.
+  - Shared helpers under `test/helpers/**` are part of that same boundary. Do not hardcode repo-relative `extensions/**` imports there, and do not keep plugin-local deep mocks in shared helpers just because multiple tests use them.
+  - When core tests or shared helpers need bundled plugin public surfaces, use `src/test-utils/bundled-plugin-public-surface.ts` for `api.ts`, `runtime-api.ts`, `contract-api.ts`, `test-api.ts`, plugin entrypoint `index.js`, and resolved module ids for dynamic import or mocking.
   - If a core test is asserting extension-specific behavior instead of a generic contract, move it to the owning extension package.
 
 ## Docs Linking (Mintlify)
@@ -149,6 +151,7 @@
 - Config schema drift uses `pnpm config:docs:gen` / `pnpm config:docs:check`.
 - Plugin SDK API drift uses `pnpm plugin-sdk:api:gen` / `pnpm plugin-sdk:api:check`.
 - If you change config schema/help or the public Plugin SDK surface, run the matching gen command and commit the updated `.sha256` hash file. Keep the two drift-check flows adjacent in scripts/workflows/docs guidance rather than inventing a third pattern.
+- When `pnpm tsgo` fails, triage by coherent surface instead of by raw error count: rerun the gate, group failures by package/module/type contract, open the source-of-truth type or export file first, fix the root mismatch, then rerun `pnpm tsgo` before widening into downstream consumers. Check `origin/main` before doing broad cleanup because some apparent type debt is already fixed upstream.
 - For narrowly scoped changes, prefer narrowly scoped tests that directly validate the touched behavior. If no meaningful scoped test exists, say so explicitly and use the next most direct validation available.
 - Verification modes for work on `main`:
   - Default mode: `main` is relatively stable. Count pre-commit hook coverage when it already verified the current tree, avoid rerunning the exact same checks just for ceremony, and prefer keeping CI/main green before landing.
@@ -296,7 +299,7 @@
 
 - When working on a GitHub Issue or PR, print the full URL at the end of the task.
 - When answering questions, respond with high-confidence answers only: verify in code; do not guess.
-- Carbon: prefer latest published beta over stable when possible; do not switch to stable casually.
+- Carbon version edits are owner-only: do not change `@buape/carbon` version pins unless you are Shadow (@thewilloftheshadow) as verified by gh.
 - Any dependency with `pnpm.patchedDependencies` must use an exact version (no `^`/`~`).
 - Patching dependencies (pnpm patches, overrides, or vendored changes) requires explicit approval; do not do this by default.
 - **Multi-agent safety:** do **not** create/apply/drop `git stash` entries unless explicitly requested (this includes `git pull --rebase --autostash`). Assume other agents may be working; keep unrelated WIP untouched and avoid cross-cutting state changes.

CHANGELOG.md

@@ -6,31 +6,179 @@ Docs: https://docs.openclaw.ai
 
 ### Changes
 
-- iOS: pin release versioning to an explicit CalVer in `apps/ios/version.json`, keep TestFlight iteration on the same short version until maintainers intentionally promote the next gateway version, and add the documented `pnpm ios:version:pin -- --from-gateway` workflow for release trains. (#63001) Thanks @ngutman.
-- Plugins/provider-auth: let provider manifests declare `providerAuthAliases` so provider variants can share env vars, auth profiles, config-backed auth, and API-key onboarding choices without core-specific wiring.
-- Memory/dreaming: add a grounded REM backfill lane with historical `rem-harness --path`, diary commit, and reset flows so old daily notes can be replayed safely into `DREAMS.md`. Thanks @mbelinky.
-- Memory/dreaming: harden grounded diary extraction so `What Happened`, `Reflections`, and durable candidates suppress operational noise and preserve more atomic lasting facts. Thanks @mbelinky.
-- Control UI/dreaming: add a structured diary view with timeline navigation, backfill/reset controls, and traceable dreaming summaries. Thanks @mbelinky.
+### Fixes
+
+## 2026.4.10
+
+### Changes
+
+- Models/Codex: add the bundled Codex provider and plugin-owned app-server harness so `codex/gpt-*` models use Codex-managed auth, native threads, model discovery, and compaction while `openai/gpt-*` stays on the normal OpenAI provider path. (#64298)
+- Memory/Active Memory: add a new optional Active Memory plugin that gives OpenClaw a dedicated memory sub-agent right before the main reply, so ongoing chats can automatically pull in relevant preferences, context, and past details without making users remember to manually say "remember this" or "search memory" first. Includes configurable message/recent/full context modes, live `/verbose` inspection, advanced prompt/thinking overrides for tuning, and opt-in transcript persistence for debugging. Docs: https://docs.openclaw.ai/concepts/active-memory. (#63286) Thanks @Takhoffman.
+- macOS/Talk: add an experimental local MLX speech provider for Talk Mode, with explicit provider selection, local utterance playback, interruption handling, and system-voice fallback. (#63539) Thanks @ImLukeF.
+- Tools/video generation: add Seedance 2.0 model refs to the bundled fal provider and submit the provider-specific duration, resolution, audio, and seed metadata fields needed for live Seedance 2.0 runs.
+- Microsoft Teams: add message actions for pin, unpin, read, react, and listing reactions. (#53432) Thanks @sudie-codes.
+- QA/Matrix: add a live `openclaw qa matrix` lane backed by a disposable Matrix homeserver, shared live-transport seams, and Matrix-specific transport coverage for threading, reactions, restart, and allowlist behavior. (#64489) Thanks @gumadeiras.
+- QA/Telegram: add a live `openclaw qa telegram` lane for private-group bot-to-bot checks, harden its artifact handling, and preserve native Telegram command reply threading for QA verification. (#64303) Thanks @obviyus.
+- QA/testing: add a `--runner multipass` lane for `openclaw qa suite` so repo-backed QA scenarios can run inside a disposable Linux VM and write back the usual report, summary, and VM logs. (#63426) Thanks @shakkernerd.
+- CLI/exec policy: add a local `openclaw exec-policy` command with `show`, `preset`, and `set` subcommands for synchronizing requested `tools.exec.*` config with the local exec approvals file, plus follow-up hardening for node-host rejection, rollback safety, and sync conflict detection. (#64050)
+- Gateway: add a `commands.list` RPC so remote gateway clients can discover runtime-native, text, skill, and plugin commands with surface-aware naming and serialized argument metadata. (#62656) Thanks @samzong.
+- Models/providers: add per-provider `models.providers.*.request.allowPrivateNetwork` for trusted self-hosted OpenAI-compatible endpoints, keep the opt-in scoped to model request surfaces, and refresh cached WebSocket managers when request transport overrides change. (#63671) Thanks @qas.
+- Feishu: standardize request user agents and register the bot as an AI agent so Feishu deployments identify OpenClaw consistently. (#63835) Thanks @evandance.
+- Matrix/partial streaming: add MSC4357 live markers to draft preview sends and edits so supporting Matrix clients can render a live/typewriter animation and stop it when the final edit lands. (#63513) Thanks @TigerInYourDream.
+- Control UI/dreaming: simplify the Scene and Diary surfaces, preserve unknown phase state for partial status payloads, and stabilize waiting-entry recency ordering so Dreaming status and review lists stay clear and deterministic. (#64035) Thanks @davemorin.
+- Agents: add an opt-in strict-agentic embedded Pi execution contract for GPT-5-family runs so plan-only or filler turns keep acting until they hit a real blocker. (#64241) Thanks @100yenadmin.
+- Agents/OpenAI: add provider-owned OpenAI/Codex tool schema compatibility and surface embedded-run replay/liveness state for long-running runs. (#64300) Thanks @100yenadmin.
+- Docs i18n: chunk raw doc translation, reject truncated tagged outputs, avoid ambiguous body-only wrapper unwrapping, and recover from terminated Pi translation sessions without changing the default `openai/gpt-5.4` path. (#62969, #63808) Thanks @hxy91819.
+
+### Fixes
+
+- Browser/security: tighten browser and sandbox navigation defenses across strict SSRF defaults, hostname allowlists, interaction-driven redirects, subframes, CDP discovery, existing sessions, tab actions, noVNC, marker-span sanitization, and Docker CDP source-range enforcement. (#61404, #63332, #63882, #63885, #63889, #64367, #64370, #64371)
+- Security/tools: harden exec preflight reads, host env denylisting, node output boundaries, outbound host-media reads, profile-mutation authorization, plugin install dependency scanning, ACPX tool hooks, Gmail watcher token redaction, and oversized realtime WebSocket frame handling. (#62333, #62661, #62662, #63277, #63551, #63553, #63886, #63890, #63891, #64459)
+- OpenAI/Codex: add required Codex OAuth scopes, classify provider/runtime failures more clearly, stop suggesting `/elevated full` when auto-approved host exec is unavailable, add OpenAI/Codex tool-schema compatibility, and preserve embedded-run replay/liveness truth across compaction retries and mutating side effects. (#64300, #64439) Thanks @100yenadmin.
+- CLI/WhatsApp media sends: route gateway-mode outbound sends with `--media` through the channel `sendMedia` path and preserve media access context, so WhatsApp document and attachment sends stop silently dropping the file while still delivering the caption. (#64478, #64492) Thanks @ShionEria.
+- Microsoft Teams: restore media downloads for personal DMs, Bot Framework `a:` conversations, OneDrive/SharePoint shared files, and Graph-backed chat IDs; accept Bot Framework audience tokens; prevent feedback-learning filename collisions; keep long tool chains alive with typing indicators; add SSO sign-in callbacks; inject parent context for thread replies; and deliver cron announcements to Teams conversation IDs. (#54932, #55383, #55386, #58001, #58249, #58774, #59731, #60956, #62219, #62674, #63063, #63942, #63945, #63949, #63951, #63953, #64087, #64088, #64089)
+- Gateway/tailscale: start Tailscale exposure and the gateway update check before awaiting channel and plugin sidecar startup so remote operators are not locked out when startup sidecars stall.
+- Gateway/startup: keep WebSocket RPC available while channels and plugin sidecars start, hold `chat.history` unavailable until startup sidecars finish so synchronous history reads cannot stall startup (reported in #63450), refresh advertised gateway methods after deferred plugin reloads, and enforce the pre-auth WebSocket upgrade budget before the no-handler 503 path so upgrade floods cannot bypass connection limits during that window. (#63480) Thanks @neeravmakwana.
+- WhatsApp: keep inbound replies, media, composing indicators, and queued outbound deliveries attached to the current socket across reconnect gaps, including fresh retry-eligible sends after the listener comes back. (#30806, #46299, #62892, #63916) Thanks @mcaxtr.
+- Gateway/thread routing: preserve Slack, Telegram, Mattermost, Matrix, ACP, restart-sentinel, and agent announce delivery targets so subagent, cron, stream-relay, session fallback, and restart messages land back in the originating thread, topic, or room casing. (#54840, #57056, #63143, #63228, #63506, #64343, #64391)
+- Models/fallback: preserve `/models` selection across transient primary-model failures and config reloads, allow timeout cooldown probes, classify OpenRouter no-endpoints responses, detect llama.cpp context overflows, and keep provider/runtime context metadata stable through reloads. (#61472, #64196, #64471)
+- Agents/BTW: keep `/btw` side questions working after tool-use turns by stripping replayed tool blocks, hidden reasoning, and malformed image payloads, omitting empty tool arrays, allowing Bedrock `auth: "aws-sdk"`, and routing Feishu `/btw` plus `/stop` through bounded out-of-band lanes. (#64218, #64219, #64225, #64324) Thanks @ngutman.
+- Control UI/BTW: render `/btw` side results as dismissible ephemeral cards in the browser, send `/btw` immediately during active runs, and clear stale BTW cards on reset flows so webchat matches the intended detached side-question behavior. (#64290) Thanks @ngutman.
+- Commands/targeting: use the selected agent or session for command output, send policy, usage/cost, context reports, model lists, bash sandbox hints, BTW/compact working directories, plugin commands, and session exports so multi-agent commands describe and mutate the intended target instead of the requester.
+- Conversation bindings: normalize focused/current conversation ids, preserve binding metadata on account and Discord rebinds, avoid stale Discord lifecycle windows, and keep generic activity touches persisted so reply routing survives rebinds and restarts.
+- iMessage/self-chat: distinguish normal DM outbound rows from true self-chat using `destination_caller_id` plus chat participants, preserve multi-handle self-chat aliases, drop ambiguous reflected echoes, and strip wrapped imsg RPC text fields. (#61619, #63868, #63980, #63989, #64000) Thanks @neeravmakwana.
+- Matrix: keep multi-account room scoping consistent, keep packaged crypto migrations warning-only when appropriate, preserve ordered block streaming, add explicit Matrix block-streaming opt-in, and resolve verification/bootstrap from the packaged runtime entry. (#58449, #59249, #59266, #64373) Thanks @gumadeiras.
+- Telegram/security: tighten Telegram `allowFrom` sender validation and keep `/whoami` allowlist reporting in sync with command auth checks.
+- Agents/timeouts: extend the default LLM idle window to 120s and keep silent no-token idle timeouts on recovery paths, so slow models can retry or fall back before users see an error.
+- Gateway/agents: preserve configured model selection and richer `IDENTITY.md` content across agent create/update flows and workspace moves, and fail safely instead of silently overwriting unreadable identity files. (#61577) Thanks @samzong.
+- Skills/TaskFlow: restore valid frontmatter fences for the bundled `taskflow` and `taskflow-inbox-triage` skills and copy bundled `SKILL.md` files as hard dist-runtime copies so skills stay discoverable and loadable after updates. (#64166, #64469) Thanks @extrasmall0.
+- Skills: respect overridden home directories when loading personal skills so service, test, and custom launch environments read the intended user skill directory instead of the process home.
+- Windows/exec: settle supervisor waits from child exit state after stdout and stderr drain even when `close` never arrives, so CLI commands stop hanging or dying with forced `SIGKILL` on Windows. (#64072) Thanks @obviyus.
+- Browser/sandbox: prevent sandbox browser CDP startup hangs by recreating containers when the browser security hash changes and by waiting on the correct sandbox browser lifecycle. (#62873) Thanks @Syysean.
+- QQBot/streaming: make block streaming configurable per QQ bot account via `streaming.mode` (`"partial"` | `"off"`, default `"partial"`) instead of hardcoding it off, so responses can be delivered incrementally. (#63746)
+- QQBot/config: allow extra fields in `channels.qqbot` and `channels.qqbot.accounts.*` so extended qqbot builds can add new config options without gateway startup failing on schema validation. (#64075) Thanks @WideLee.
+- Dreaming/gateway: require `operator.admin` for persistent `/dreaming on|off` changes and treat missing gateway client scopes as unprivileged instead of silently allowing config writes. (#63872) Thanks @mbelinky.
+- Gateway/pairing: prefer explicit QR bootstrap auth over earlier Tailscale auth classification so iOS `/pair qr` silent bootstrap pairing does not fall through to `pairing required`. (#59232) Thanks @ngutman.
+- Browser/control: auto-generate browser-control auth tokens for `none` and `trusted-proxy` modes, and route browser auth/profile/doctor helpers through the public browser plugin facades. (#63280, #63957) Thanks @pgondhi987.
+- Browser/act: centralize `/act` request normalization and execution dispatch while adding stable machine-readable route-level error codes for invalid requests, selector misuse, evaluate-disabled gating, target mismatch, and existing-session unsupported actions. (#63977) Thanks @joshavant.
+- Security/QQBot: enforce media storage boundaries for all outbound local file paths and route image-size probes through SSRF-guarded media fetching instead of raw `fetch()`. (#63271, #63495) Thanks @pgondhi987.
+- Channel setup: ignore workspace plugin shadows when resolving trusted channel setup catalog entries so onboarding and setup flows keep using the bundled, trusted setup contract.
+- Gateway/memory startup: load the explicitly selected memory-slot plugin during gateway startup, while keeping restrictive allowlists and implicit default memory slots from auto-starting unrelated memory plugins. (#64423) Thanks @EronFan.
+- Config/plugins: let config writes keep disabled plugin entries without forcing required plugin config schemas or crashing raw plugin validation, and avoid re-activating plugin registry state during schema checks. (#54971, #63296) Thanks @fuller-stack-dev.
+- Config validation: surface the actual offending field for strict-schema union failures in bindings, including top-level unexpected keys on the matching ACP branch. (#40841) Thanks @Hollychou924.
+- Wizard/plugin config: coerce integer-typed plugin config fields from interactive text input so integer schema values persist as numbers instead of failing validation. (#63346) Thanks @jalehman.
+- Daemon/gateway install: preserve safe custom service env vars on forced reinstall, merge prior custom PATH segments behind the managed service PATH, and stop removed managed env keys from persisting as custom carryover. (#63136) Thanks @WarrenJones.
+- Cron/scheduling: treat `nextRunAtMs <= 0` as invalid across cron update, maintenance, timer, and stale-delivery paths so corrupted zero timestamps self-heal instead of causing immediate runs or skipped deliveries. (#63507) Thanks @WarrenJones.
+- Cron/auth: resolve auth profiles consistently for isolated cron jobs so scheduled runs use the same configured provider credentials as interactive sessions. (#62797) Thanks @neeravmakwana.
+- Tasks: let `openclaw tasks cancel` cancel stuck background tasks that never reached a normal terminal state. (#62506) Thanks @neeravmakwana.
+- Sessions/model selection: preserve catalog-backed session model labels, provider-qualified context limits, and already-qualified session model refs when catalog metadata is unavailable, so model selection and memory/context budgets survive reloads without bogus provider prefixes. (#61382, #62493) Thanks @Mule-ME.
+- Status: show configured fallback models in `/status` and shared session status cards so per-agent fallback configuration is visible before a live failover happens. (#33111) Thanks @AnCoSONG.
+- `/context detail` now compares the tracked prompt estimate with cached context usage and surfaces untracked provider/runtime overhead when present. (#28391) Thanks @ImLukeF.
+- Gateway/sessions: scope bare `sessions.create` aliases like `main` to the requested agent while preserving the canonical `global` and `unknown` sentinel keys. (#58207) Thanks @jalehman.
+- Gateway/session reset: emit the typed `before_reset` hook for gateway `/new` and `/reset`, preserving reset-hook behavior even when the previous transcript has already been archived. (#53872) Thanks @VACInc.
+- Plugins/commands: pass the active host `sessionKey` into plugin command contexts, and include `sessionId` when it is already available from the active session entry, so bundled and third-party commands can resolve the current conversation reliably. (#59044) Thanks @jalehman.
+- Agents/auth: honor `models.providers.*.authHeader` for pi embedded runner model requests by injecting `Authorization: Bearer <apiKey>` when requested. (#54390) Thanks @lndyzwdxhs.
+- Claude CLI: clear inherited Anthropic auth/header environment aliases before spawning Claude Code and add sanitized CLI backend auth-env diagnostics for debugging gateway-run provider selection.
+- Agents/failover: classify AbortError and stream-abort messages as timeout so Ollama NDJSON stream aborts stop showing `reason=unknown` in model fallback logs. (#58324) Thanks @yelog.
+- Fireworks/FirePass: disable Kimi K2.5 Turbo reasoning output by forcing thinking off on the FirePass path and hardening the provider wrapper so hidden reasoning no longer leaks into visible replies. (#63607) Thanks @frankekn.
+- Discord: update Carbon to v0.15.0. Thanks @thewilloftheshadow.
+- Config/Discord: coerce safe integer numeric Discord IDs to strings during config validation, keep unsafe or precision-losing numeric snowflakes rejected, and align `openclaw doctor` repair guidance with the same fail-closed behavior. (#45125) Thanks @moliendocode.
+- BlueBubbles/config: accept `enrichGroupParticipantsFromContacts` in the core strict config schema so gateways no longer fail validation or startup when the BlueBubbles plugin writes that field. (#56889) Thanks @zqchris.
+- Feishu/webhooks: read webhook bodies through the pre-auth guard so unauthenticated webhook traffic stays under the same body budget as other protected channel ingress paths.
+- Tools/web_fetch: add an opt-in `tools.web.fetch.ssrfPolicy.allowRfc2544BenchmarkRange` config so fake-IP proxy environments that resolve public sites into `198.18.0.0/15` can use `web_fetch` without weakening the default SSRF block. (#61830) Thanks @xing-xing-coder.
+- Dreaming/cron: reconcile managed dreaming cron from startup config and runtime lifecycle changes, but only recover managed dreaming cron state during heartbeat-triggered dreaming checks so ordinary chat traffic does not recreate removed jobs. (#63873, #63929, #63938) Thanks @mbelinky.
+- Memory/lancedb: accept `dreaming` config when `memory-lancedb` owns the memory slot so Dreaming surfaces can read slot-owner settings without schema rejection. (#63874) Thanks @mbelinky.
+- Control UI/dreaming: keep the Dreaming trace area contained and scrollable so overlays no longer cover tabs or blow out the page layout. (#63875) Thanks @mbelinky.
+- Dreaming/narrative: harden request-scoped diary fallback so scheduled dreaming only falls back on the dedicated subagent-runtime error, stop trusting spoofable raw error-code objects, and avoid leaking workspace paths when local fallback writes fail. (#64156) Thanks @mbelinky.
+- Dreaming/diary: add idempotent narrative subagent runs, preserve restrictive `DREAMS.md` permissions during atomic writes, and surface temp cleanup failures so repeated sweeps do not double-run the same narrative request or silently weaken diary safety. (#63876) Thanks @mbelinky.
+- Heartbeats/sessions: remove stale accumulated isolated heartbeat session keys when the next tick converges them back to the canonical sibling, so repaired sessions stop showing orphaned `:heartbeat:heartbeat` variants in session listings. (#59606) Thanks @rogerdigital.
+- Gateway/run cleanup: fix stale run-context TTL cleanup so the new maintenance sweep resets orphaned run sequence state and prevents unbounded run-context growth. (#52731) Thanks @artwalker.
+- UI/compaction: keep the compaction indicator in a retry-pending state until the run actually finishes, so the UI does not show `Context compacted` before compaction actually finishes. (#55132) Thanks @mpz4life.
+- Cron/tool schemas: keep cron tool schemas strict-model-friendly while still preserving `failureAlert=false`, nullable `agentId`/`sessionKey`, and flattened add/update recovery for the newly exposed cron job fields. (#55043) Thanks @brunolorente.
+- Git metadata: read commit ids from packed refs as well as loose refs so version and status metadata stay accurate after repository maintenance. (#63943)
+- Gateway: keep `commands.list` skill entries categorized under tools and include provider-aware plugin `nativeName` metadata even when `scope=text`, so remote clients can group skills correctly and map text-surface plugin commands back to native aliases. (#64147)
+- TUI: reset footer activity to idle when switching sessions so a stale streaming indicator cannot persist after the selection changes. (#63988) Thanks @neeravmakwana.
+- Claude CLI: stop marking spawned Claude Code runs as host-managed so they keep using normal CLI subscription behavior. (#64023) Thanks @Alex-Alaniz.
+- Codex auth: brand Codex OAuth flows as OpenClaw in user-visible auth prompts and diagnostics.
+- Gateway/pairing: fail closed for paired device records that have no device tokens, and reject pairing approvals whose requested scopes do not match the requested device roles.
+- ACP/gateway chat: classify lifecycle errors before forwarding them to ACP clients so refusals use ACP's refusal stop reason while transient backend errors continue to finish as normal turns.
+- Claude CLI/skills: pass eligible OpenClaw skills into CLI runs, including native Claude Code skill resolution via a temporary plugin plus per-run skill env/API key injection. (#62686, #62723) Thanks @zomars.
+- Discord: keep generated auto-thread names working with reasoning models by giving title generation enough output budget for thinking plus visible title text. (#64172) Thanks @hanamizuki.
+- Heartbeat: ignore doc-only Markdown fence markers in the default `HEARTBEAT.md` template so comment-only heartbeat scaffolds skip API calls again. (#61690, #63434) Thanks @ravyg.
+- Reply/skills: keep resolved skill and memory secret config stable through embedded reply runs so raw SecretRefs in secondary skill settings no longer crash replies when the gateway already has the live env. (#64249) Thanks @mbelinky.
+- Dreaming/startup: keep plugin-registered startup hooks alive across workspace hook reloads and include dreaming startup owners in the gateway startup plugin scope, so managed Dreaming cron registration comes back reliably after gateway boot. (#62327, #64258) Thanks @mbelinky.
+- Plugins: treat duplicate `registerService` calls from the same plugin id as idempotent so snapshot and activation loads no longer emit spurious `service already registered` diagnostics. (#62033, #64128) Thanks @ly85206559.
+- Discord/TTS: route auto voice replies through the native voice-note path so Discord receives Opus voice messages instead of regular audio attachments. (#64096) Thanks @LiuHuaize.
+- Config/plugins: use plugin-owned command alias metadata when `plugins.allow` contains runtime command names like `dreaming`, and point users at the owning plugin instead of stale plugin-not-found guidance. (#64191, #64242) Thanks @feiskyer.
+- Agents/Gemini: strip orphaned `required` entries from Gemini tool schemas so provider validation no longer rejects tools after schema cleanup or union flattening. (#64284) Thanks @xxxxxmax.
+- Assistant text: strip Qwen-style XML tool call payloads from visible replies so web and channel messages no longer show raw `<tool_call><function=...>` output. (#63999, #64214) Thanks @MoerAI.
+- Daemon/gateway: prevent systemd restart storms on configuration errors by exiting with `EX_CONFIG` and adding generated unit restart-prevention guards. (#63913) Thanks @neo1027144-creator.
+- Agents/exec: prevent gateway crash ("Agent listener invoked outside active run") when a subagent exec tool produces stdout/stderr after the agent run has ended or been aborted. (#62821) Thanks @openperf.
+- Gateway/OpenAI compat: return real `usage` for non-stream `/v1/chat/completions` responses, emit the final usage chunk when `stream_options.include_usage=true`, and bound usage-gated stream finalization after lifecycle end. (#62986) Thanks @Lellansin.
+- Agents/subagents: deduplicate delivered completion announces so retry or re-entry cleanup does not inject duplicate internal-context completion turns into the parent session. (#61525) Thanks @100yenadmin.
+- Agents/exec: keep sandboxed `tools.exec.host=auto` sessions from honoring per-call `host=node` or `host=gateway` overrides while a sandbox runtime is active, and stop advertising node routing in that state so exec stays on the sandbox host. (#63880)
+- Agents/subagents: preserve archived delete-mode runs until `sessions.delete` succeeds and prevent overlapping archive sweeps from duplicating in-flight cleanup attempts. (#61801) Thanks @100yenadmin.
+- Cron/isolated agent: run scheduled agent turns as non-owner senders so owner-only tools stay unavailable during cron execution. (#63878)
+- Discord/sandbox: include `image` in sandbox media param normalization so Discord event cover images cannot bypass sandbox path rewriting. (#64377) Thanks @mmaps.
+- Agents/exec: extend exec completion detection to cover local background exec formats so the owner-downgrade fires correctly for all exec paths. (#64376) Thanks @mmaps.
+- Hooks/security: mark agent hook system events as untrusted and sanitize hook display names before cron metadata reuse. (#64372) Thanks @eleqtrizit.
+- Daemon/launchd: keep `openclaw gateway stop` persistent without uninstalling the macOS LaunchAgent, re-enable it on explicit restart or repair, and harden launchd label handling. (#64447) Thanks @ngutman.
+- Plugins/context engines: preserve `plugins.slots.contextEngine` through normalization and keep explicitly selected workspace context-engine plugins enabled, so loader diagnostics and plugin activation stop dropping that slot selection. (#64192) Thanks @hclsys.
+- Heartbeat: stop top-level `interval:` and `prompt:` fields outside the `tasks:` block from bleeding into the last parsed heartbeat task. (#64488) Thanks @Rahulkumar070.
+- Agents/OpenAI replay: preserve malformed function-call arguments in stored assistant history, avoid double-encoding preserved raw strings on replay, and coerce replayed string args back to objects at Anthropic and Google provider boundaries. (#61956) Thanks @100yenadmin.
+- Heartbeat/config: accept and honor `agents.defaults.heartbeat.timeoutSeconds` and per-agent heartbeat timeout overrides for heartbeat agent turns. (#64491) Thanks @cedillarack.
+- CLI/devices: make implicit `openclaw devices approve` selection preview-only and require approving the exact request ID, preventing latest-request races during device pairing. (#64160) Thanks @coygeek.
+- Media/security: honor sender-scoped `toolsBySender` policy for outbound host-media reads so denied senders cannot trigger host file disclosure via attachment hydration. (#64459) Thanks @eleqtrizit.
+- Browser/security: reject strict-policy hostname navigation unless the hostname is an explicit allowlist exception or IP literal, and route CDP HTTP discovery through the pinned SSRF fetch path. (#64367) Thanks @eleqtrizit.
+- Models/vLLM: ignore empty `tool_calls` arrays from reasoning-model OpenAI-compatible replies, reset false `toolUse` stop reasons when no actual tool calls were parsed, and stop sending `tool_choice` unless tools are present so vLLM reasoning responses no longer hang indefinitely. (#61197, #61534) Thanks @balajisiva.
+
+## 2026.4.9
+
+### Changes
+
+- Memory/dreaming: add a grounded REM backfill lane with historical `rem-harness --path`, diary commit/reset flows, cleaner durable-fact extraction, and live short-term promotion integration so old daily notes can replay into Dreams and durable memory without a second memory stack. Thanks @mbelinky.
+- Control UI/dreaming: add a structured diary view with timeline navigation, backfill/reset controls, traceable dreaming summaries, and a grounded Scene lane with promotion hints plus a safe clear-grounded action for staged backfill signals. (#63395) Thanks @mbelinky.
+- QA/lab: add character-vibes evaluation reports with model selection and parallel runs so live QA can compare candidate behavior faster.
+- Plugins/provider-auth: let provider manifests declare `providerAuthAliases` so provider variants can share env vars, auth profiles, config-backed auth, and API-key onboarding choices without core-specific wiring.
+- iOS: pin release versioning to an explicit CalVer in `apps/ios/version.json`, keep TestFlight iteration on the same short version until maintainers intentionally promote the next gateway version, and add the documented `pnpm ios:version:pin -- --from-gateway` workflow for release trains. (#63001) Thanks @ngutman.
 
 ### Fixes
 
-- Control UI: guard stale session-history reloads during fast session switches so the selected session and rendered transcript stay in sync. (#62975) Thanks @scoootscooob.
-- Slack/media: preserve bearer auth across same-origin `files.slack.com` redirects while still stripping it on cross-origin Slack CDN hops, so `url_private_download` image attachments load again. (#62960) Thanks @vincentkoc.
-- Gateway/node exec events: mark remote node `exec.started`, `exec.finished`, and `exec.denied` summaries as untrusted system events and sanitize node-provided command/output/reason text before enqueueing them, so remote node output cannot inject trusted `System:` content into later turns. (#62659) Thanks @eleqtrizit.
-- Agents/timeouts: make the LLM idle timeout inherit `agents.defaults.timeoutSeconds` when configured, disable the unconfigured idle watchdog for cron runs, and point idle-timeout errors at `agents.defaults.llm.idleTimeoutSeconds`. Thanks @drvoss.
-- Security/dotenv: expand workspace `.env` filtering to block runtime-control variables like gateway routing, ClawHub endpoints/tokens, browser executable overrides, and skip/disable control families, so untrusted repositories cannot steer OpenClaw runtime behavior through repo-local dotenv files. (#62660) Thanks @eleqtrizit.
-- Agents/failover: classify Z.ai vendor code `1311` as billing and `1113` as auth, including long wrapped `1311` payloads, so these errors stop falling through to generic failover handling. (#49552) Thanks @1bcMax.
-- Browser/security: block browser-control module override and skip-server env vars from untrusted workspace `.env` files, and reject unsafe URL-style browser control override specifiers before lazy loading, so repo-local dotenv state cannot steer browser control module loading. (#62663) Thanks @eleqtrizit.
-- QQBot/media-tags: support HTML entity-encoded angle brackets (`&lt;`/`&gt;`) in media-tag regexes so entity-escaped `<qqimg>` tags from upstream are correctly parsed and normalized. (#60493) Thanks @ylc0919.
-- npm packaging: mirror bundled Slack, Telegram, Discord, and Feishu channel runtime deps at the root and harden published-install verification so fresh installs fail fast on manifest drift instead of missing-module crashes. (#63065) Thanks @scoootscooob.
-- npm packaging: derive required root runtime mirrors from bundled plugin manifests and built root chunks, then install packed release tarballs without the repo `node_modules` so release checks catch missing plugin deps before publish.
-- Reply/doctor: resolve reply-run SecretRefs before preflight helpers touch config, surface gateway OAuth reauth failures to users, and make `openclaw doctor` call out exact reauth commands.
-- Android/pairing: clear stale setup-code auth on new QR scans, bootstrap operator and node sessions from fresh pairing, prefer stored device tokens after bootstrap handoff, and pause pairing auto-retry while the app is backgrounded so scan-once Android pairing recovers reliably again. (#63199) Thanks @obviyus.
-- Auto-reply/NO_REPLY: strip glued leading `NO_REPLY` tokens before reply normalization and ACP-visible streaming so silent sentinel text no longer leaks into user-visible replies while preserving substantive `NO_REPLY ...` text. Thanks @frankekn.
-- Gateway/sessions: clear auto-fallback-pinned model overrides on `/reset` and `/new` while still preserving explicit user model selections, including legacy sessions created before override-source tracking existed. (#63155) Thanks @frankekn.
-- Codex CLI: pass OpenClaw's system prompt through Codex's `model_instructions_file` config override so fresh Codex CLI sessions receive the same prompt guidance as Claude CLI sessions.
-- Matrix/gateway: wait for Matrix sync readiness before marking startup successful, keep Matrix background handler failures contained, and route fatal Matrix sync stops through channel-level restart handling instead of crashing the whole gateway. (#62779) Thanks @gumadeiras.
 - Browser/security: re-run blocked-destination safety checks after interaction-driven main-frame navigations from click, evaluate, hook-triggered click, and batched action flows, so browser interactions cannot bypass the SSRF quarantine when they land on forbidden URLs. (#63226) Thanks @eleqtrizit.
+- Security/dotenv: block runtime-control env vars plus browser-control override and skip-server env vars from untrusted workspace `.env` files, and reject unsafe URL-style browser control override specifiers before lazy loading. (#62660, #62663) Thanks @eleqtrizit.
+- Gateway/node exec events: mark remote node `exec.started`, `exec.finished`, and `exec.denied` summaries as untrusted system events and sanitize node-provided command/output/reason text before enqueueing them, so remote node output cannot inject trusted `System:` content into later turns. (#62659) Thanks @eleqtrizit.
+- Plugins/onboarding auth choices: prevent untrusted workspace plugins from colliding with bundled provider auth-choice ids during non-interactive onboarding, so bundled provider setup keeps operator secrets out of untrusted workspace plugin handlers unless those plugins are explicitly trusted. (#62368) Thanks @pgondhi987.
+- Security/dependency audit: force `basic-ftp` to `5.2.1` for the CRLF command-injection fix and bump Hono plus `@hono/node-server` in production resolution paths.
+- Android/pairing: clear stale setup-code auth on new QR scans, bootstrap operator and node sessions from fresh pairing, prefer stored device tokens after bootstrap handoff, and pause pairing auto-retry while the app is backgrounded so scan-once Android pairing recovers reliably again. (#63199) Thanks @obviyus.
+- Matrix/gateway: wait for Matrix sync readiness before marking startup successful, keep Matrix background handler failures contained, and route fatal Matrix sync stops through channel-level restart handling instead of crashing the whole gateway. (#62779) Thanks @gumadeiras.
+- Slack/media: preserve bearer auth across same-origin `files.slack.com` redirects while still stripping it on cross-origin Slack CDN hops, so `url_private_download` image attachments load again. (#62960) Thanks @vincentkoc.
+- Reply/doctor: use the active runtime snapshot for queued reply runs, resolve reply-run SecretRefs before preflight helpers touch config, surface gateway OAuth reauth failures to users, and make `openclaw doctor` call out exact reauth commands. (#62693, #63217) Thanks @mbelinky.
+- Control UI: guard stale session-history reloads during fast session switches so the selected session and rendered transcript stay in sync. (#62975) Thanks @scoootscooob.
+- Gateway/chat: suppress exact and streamed `ANNOUNCE_SKIP` / `REPLY_SKIP` control replies across live chat updates and history sanitization so internal agent-to-agent control tokens no longer leak into user-facing gateway chat surfaces. (#51739) Thanks @Pinghuachiu.
+- Auto-reply/NO_REPLY: strip glued leading `NO_REPLY` tokens before reply normalization and ACP-visible streaming so silent sentinel text no longer leaks into user-visible replies while preserving substantive `NO_REPLY ...` text. Thanks @frankekn.
+- Sessions/routing: preserve established external routes on inter-session announce traffic so `sessions_send` follow-ups do not steal delivery from Telegram, Discord, or other external channels. (#58013) Thanks @duqaXxX.
+- Gateway/sessions: clear auto-fallback-pinned model overrides on `/reset` and `/new` while still preserving explicit user model selections, including legacy sessions created before override-source tracking existed. (#63155) Thanks @frankekn.
+- Slack/ACP: treat Slack ACP block replies as visible delivered output so OpenClaw stops re-sending the final fallback text after Slack already rendered the reply. (#62858) Thanks @gumadeiras.
+- Slack/partial streaming: key turn-local dedupe by dispatch kind and keep the final fallback reply path active when preview finalization fails so stale preview text cannot suppress the actual final answer. (#62859) Thanks @gumadeiras.
+- Matrix/doctor: migrate legacy `channels.matrix.dm.policy: "trusted"` configs back to compatible DM policies during `openclaw doctor --fix`, preserving explicit `allowFrom` boundaries as `allowlist` and defaulting empty legacy configs to `pairing`. (#62942) Thanks @lukeboyett.
+- npm packaging: mirror bundled channel runtime deps, stage Nostr runtime deps, derive required root mirrors from manifests and built chunks, and test packed release tarballs without repo `node_modules` so fresh installs fail fast on missing plugin deps instead of crashing at runtime. (#63065) Thanks @scoootscooob.
+- QA/live auth: fail fast when live QA scenarios hit classified auth or runtime failure replies, including raw scenario wait paths, and sanitize missing-key guidance so gateway auth problems surface as actionable errors instead of timeouts. (#63333) Thanks @shakkernerd.
+- Providers/OpenAI: default missing reasoning effort to `high` on OpenAI Responses, WebSocket, and compatible completions transports, while still honoring explicit per-run reasoning levels.
+- Providers/Ollama: allow Ollama models using the native `api: "ollama"` path to optionally display thinking output when `/think` is set to a non-off level. (#62712) Thanks @hoyyeva.
+- Codex CLI: pass OpenClaw's system prompt through Codex's `model_instructions_file` config override so fresh Codex CLI sessions receive the same prompt guidance as Claude CLI sessions.
+- Auth/profiles: persist explicit auth-profile upserts directly and skip external CLI sync for local writes so profile changes are saved without stale external credential state.
+- Agents/timeouts: make the LLM idle timeout inherit `agents.defaults.timeoutSeconds` when configured, disable the unconfigured idle watchdog for cron runs, and point idle-timeout errors at `agents.defaults.llm.idleTimeoutSeconds`. Thanks @drvoss.
+- Agents/failover: classify Z.ai vendor code `1311` as billing and `1113` as auth, including long wrapped `1311` payloads, so these errors stop falling through to generic failover handling. (#49552) Thanks @1bcMax.
+- QQBot/media-tags: support HTML entity-encoded angle brackets (`&lt;`/`&gt;`), URL slashes in attributes, and self-closing media tags so upstream `<qqimg>` payloads are correctly parsed and normalized. (#60493) Thanks @ylc0919.
+- Memory/dreaming: harden grounded backfill inputs, diary writes, status payloads, and diary action classification by preserving source-day labels, rejecting missing or symlinked targets cleanly, normalizing diary headings in gateway backfills, and tightening claim splitting plus diary source metadata. Thanks @mbelinky.
+- Memory/dreaming: accept embedded heartbeat trigger tokens so light and REM dreaming still run when runtime wrappers include extra heartbeat text.
+- Android/manual connect: allow blank port input only for TLS manual gateway endpoints so standard HTTPS Tailscale hosts default to `443` without silently changing cleartext manual connects. (#63134) Thanks @Tyler-RNG.
+- Windows/update: add heap headroom to Windows `pnpm build` steps during dev updates so update preflight builds stop failing on low default Node memory.
+- Plugin SDK: export the channel plugin base and web-search config contract through the public package so plugins can use them without private imports.
+- Plugins/contracts: keep test-only helpers out of production contract barrels, load shared contract harnesses through bundled test surfaces, and harden guardrails so indirect re-exports and canonical `*.test.ts` files stay blocked. (#63311) Thanks @altaywtf.
+- Control UI/models: preserve provider-qualified refs for OpenRouter catalog models whose ids already contain slashes so picker selections submit allowlist-compatible model refs instead of dropping the `openrouter/` prefix. (#63416) Thanks @sallyom.
+- Plugin SDK/command auth: split command status builders onto the lightweight `openclaw/plugin-sdk/command-status` subpath while preserving deprecated `command-auth` compatibility exports, so auth-only plugin imports no longer pull status/context warmup into CLI onboarding paths. (#63174) Thanks @hxy91819.
+- Wizard/plugin config: coerce integer-typed plugin config fields from interactive text input so integer schema values persist as numbers instead of failing validation. (#63346) Thanks @jalehman.
 
 ## 2026.4.8
 
@@ -81,7 +229,6 @@ Docs: https://docs.openclaw.ai
 - TUI: route `/status` through the shared session-status command, keep commentary hidden in history, strip raw envelope metadata from async command notices, preserve fallback streaming before per-attempt failures finalize, and restore Kitty keyboard state on exit or fatal crashes. (#49130, #59985, #60043, #61463) Thanks @biefan and contributors.
 - iOS/Watch exec approvals: keep Apple Watch review and approval recovery working while the iPhone is locked or backgrounded, including reconnect recovery, pending approval persistence, notification cleanup, and APNs-backed watch refresh recovery. (#61757) Thanks @ngutman.
 - Agents/context overflow: combine oversized and aggregate tool-result recovery in one pass and restore a total-context overflow backstop so recoverable sessions retry instead of failing early. (#61651) Thanks @Takhoffman.
-- Agents/OpenAI: default missing reasoning effort to `high` on OpenAI Responses, WebSocket, and compatible completions transports, while still honoring explicit per-run reasoning levels.
 - Auth/OpenAI Codex OAuth: reload fresh on-disk credentials inside the locked refresh path and retry once after `refresh_token_reused` rotates only the stored refresh token, so relogin/restart recovery stops getting stuck on stale cached auth state. Thanks @owen-ever.
 - Auth/OpenAI Codex OAuth: keep native `/model ...@profile` selections on the target session and honor explicit user-locked auth profiles even when per-agent auth order excludes them. (#62744) Thanks @jalehman.
 - Providers/Anthropic: preserve thinking blocks for Claude Opus 4.5+, Sonnet 4.5+, and newer Claude 4-family models so prompt-cache prefixes keep matching, and skip `service_tier` injection on OAuth-authenticated stream wrapper requests so Claude OAuth streaming stops failing with HTTP 401. (#60356, #61793)
@@ -92,7 +239,6 @@ Docs: https://docs.openclaw.ai
 - Gateway tool/exec config: block model-facing `gateway config.apply` and `config.patch` writes from changing exec approval paths such as `safeBins`, `safeBinProfiles`, `safeBinTrustedDirs`, and `strictInlineEval`, while still allowing unchanged structured values through. (#62001) Thanks @eleqtrizit.
 - Host exec/env sanitization: block dangerous Java, Rust, Cargo, Git, Kubernetes, cloud credential, config-path, and Helm env overrides so host-run tools cannot be redirected to attacker-chosen code, config, credentials, or repository state. (#59119, #62002, #62291) Thanks @eleqtrizit and contributors.
 - Commands/allowlist: require owner authorization for `/allowlist add` and `/allowlist remove` before channel resolution, so non-owner but command-authorized senders can no longer persistently rewrite allowlist policy state. (#62383) Thanks @pgondhi987.
-- Plugins/onboarding auth choices: prevent untrusted workspace plugins from colliding with bundled provider auth-choice ids during non-interactive onboarding, so bundled provider setup keeps operator secrets out of untrusted workspace plugin handlers unless those plugins are explicitly trusted. (#62368) Thanks @pgondhi987.
 - Feishu/docx uploads: honor `tools.fs.workspaceOnly` for local `upload_file` and `upload_image` paths by forwarding workspace-constrained `localRoots` into the media loader, so docx uploads can no longer read host-local files outside the workspace when workspace-only mode is active. (#62369) Thanks @pgondhi987.
 - Network/fetch guard: drop request bodies and body-describing headers on cross-origin `307` and `308` redirects by default, so attacker-controlled redirect hops cannot receive secret-bearing POST payloads from SSRF-guarded fetch flows unless a caller explicitly opts in. (#62357) Thanks @pgondhi987.
 - Browser/SSRF: treat main-frame `document` redirect hops as navigations even when Playwright does not flag them as `isNavigationRequest()`, so strict private-network blocking still stops forbidden redirect pivots before the browser reaches the internal target. (#62355) Thanks @pgondhi987.
@@ -149,6 +295,10 @@ Docs: https://docs.openclaw.ai
 - Agents/model resolution: let explicit `openai-codex/gpt-5.4` selection prefer provider runtime metadata when it reports a larger context window, keeping configured Codex runs aligned with the live provider limits. (#62694) Thanks @ruclaw7.
 - Agents/model resolution: keep explicit-model runtime comparisons on the configured workspace plugin registry, so workspace-installed providers do not silently fall back to stale explicit metadata during runtime model lookup.
 - Providers/Z.AI: default onboarding and endpoint detection to GLM-5.1 instead of GLM-5. (#61998) Thanks @serg0x.
+- Cron/isolated: resolve auth profiles without treating every isolated run as a brand-new auth session, so profile-based providers (for example OpenRouter) keep a stable credential choice instead of rotating or ignoring stored keys. (#62783) Thanks @neeravmakwana.
+- CLI/tasks: `openclaw tasks cancel` now records operator cancellation for CLI runtime tasks instead of returning "Task runtime does not support cancellation yet", so stuck `running` CLI tasks can be cleared. (#62419) Thanks @neeravmakwana.
+- Sessions/context: resolve context window limits using the active provider plus model (not bare model id alone) when persisting session usage, applying inline directives, and sizing memory-flush / preflight compaction thresholds, so duplicate model ids across providers no longer leak the wrong `contextTokens` into the session store or `/status`. (#62472) Thanks @neeravmakwana.
+- Channels/setup: exclude workspace shadow entries from channel setup catalog lookups and align trust checks with auto-enable so workspace-scoped overrides no longer bypass the trusted catalog. (`GHSA-82qx-6vj7-p8m2`) Thanks @zsxsoft.
 - Reply execution: prefer the active runtime snapshot over stale queued reply config during embedded reply and follow-up execution so SecretRef-backed reply turns stop crashing after secrets have already resolved. (#62693) Thanks @mbelinky.
 - Android/manual connect: allow blank port input only for TLS manual gateway endpoints so standard HTTPS Tailscale hosts default to `443` without silently changing cleartext manual connects. (#63134) Thanks @Tyler-RNG.
 - Matrix/agents: hide owner-only `set-profile` from embedded agent channel-action discovery so non-owner runs stop advertising profile updates they cannot execute. (#62662) Thanks @eleqtrizit.
@@ -371,7 +521,7 @@ Docs: https://docs.openclaw.ai
 - Agents/scheduling: steer background-now work toward automatic completion wake and treat `process` polling as on-demand inspection or intervention instead of default completion handling. (#60877) Thanks @vincentkoc.
 - Agents/skills: skip `.git` and `node_modules` when mirroring skills into sandbox workspaces so read-only sandboxes do not copy repo history or dependency trees. (#61090) Thanks @joelnishanth.
 - ACP/agents: inherit the target agent workspace for cross-agent ACP spawns and fall back safely when the inherited workspace no longer exists. (#58438) Thanks @zssggle-rgb.
-- ACPX/Windows: preserve backslashes and absolute `.exe` paths in Claude CLI parsing, and fail fast on wrapper-script targets with guidance to use `cmd.exe /c`, `powershell.exe -File`, or `node <script>`. (#60689) Thanks @steipete.
+- ACPX/Windows: preserve backslashes and absolute `.exe` paths in Claude CLI parsing, and fail fast on wrapper-script targets with guidance to use `cmd.exe /c`, `powershell.exe -File`, or `node <script>`. (#60689)
 - Auth/failover: persist selected fallback overrides before retrying, shorten `auth_permanent` lockouts, and refresh websocket/shared-auth sessions only when real auth changes occur so retries and secret rotations behave predictably. (#60404, #60323, #60387) Thanks @extrasmall0 and @mappel-nv.
 - Gateway/channels: pin the initial startup channel registry before later plugin-registry churn so configured channels stay visible and `channels.status` stops falling back to empty `channelOrder` / `channels` payloads after runtime plugin loads.
 - Prompt caching: order stable workspace project-context files before `HEARTBEAT.md` and keep `HEARTBEAT.md` below the system-prompt cache boundary so heartbeat churn does not invalidate the stable project-context prefix. (#58979) Thanks @yozu and @vincentkoc.
@@ -402,7 +552,6 @@ Docs: https://docs.openclaw.ai
 - Matrix: avoid failing startup when token auth already knows the user ID but still needs optional device metadata, retry transient auth bootstrap requests, and backfill missing device IDs after startup while keeping unknown-device storage reuse conservative until metadata is repaired. (#61383) Thanks @gumadeiras.
 - Agents/exec: stop streaming `tool_execution_update` events after an exec session backgrounds, preventing delayed background output from hitting a stale listener and crashing the gateway while keeping the output available through `process poll/log`. (#61627) Thanks @openperf.
 - Matrix: pass configured `deviceId` through health probes and keep probe-only client setup out of durable Matrix storage, so health checks preserve the correct device identity without rewriting `storage-meta.json` or related probe state on disk. (#61581) Thanks @MoerAI.
-  ||||||| parent of b4694a4ac7 (Telegram: add outbound chunker regression coverage)
 - Image generation/build: write stable runtime alias files into `dist/` and route provider-auth runtime lookups through those aliases so image-generation providers keep resolving auth/runtime modules after rebuilds instead of crashing on missing hashed chunk files.
 - Config/runtime: pin the first successful config load in memory for the running process and refresh that snapshot on successful writes/reloads, so hot paths stop reparsing `openclaw.json` between watcher-driven swaps.
 - Config/legacy cleanup: stop probing obsolete alternate legacy config names and service labels during local config/service detection, while keeping the active `~/.openclaw/openclaw.json` path canonical.
@@ -1343,7 +1492,7 @@ Docs: https://docs.openclaw.ai
 - Gateway/status: tolerate network interface discovery failures in status, onboarding control-UI links, and self-presence display paths so those surfaces fall back cleanly instead of crashing. (#52195) Thanks @meng-clb.
 - Gateway/Linux: auto-detect nvm-managed Node TLS CA bundle needs before CLI startup and refresh installed services that are missing `NODE_EXTRA_CA_CERTS`. (#51146) Thanks @GodsBoy.
 - Google auth/Node 25: patch `gaxios` to use native fetch without injecting `globalThis.window`, while translating proxy and mTLS transport settings so Google Vertex and Google Chat auth keep working on Node 25. (#47914) Thanks @pdd-cli.
-- Gateway/plugins: pin runtime webhook routes to the gateway startup registry so channel webhooks keep working across plugin-registry churn, and make plugin auth + dispatch resolve routes from the same live HTTP-route registry. (#47902) Fixes #46924 and #47041. Thanks @steipete.
+- Gateway/plugins: pin runtime webhook routes to the gateway startup registry so channel webhooks keep working across plugin-registry churn, and make plugin auth + dispatch resolve routes from the same live HTTP-route registry. (#47902) Fixes #46924 and #47041.
 - Gateway/restart: defer externally signaled unmanaged restarts through the in-process idle drain, and preserve the restored subagent run as remap fallback during orphan recovery so resumed sessions do not duplicate work. (#47719) Thanks @joeykrug.
 - Telegram/setup: seed fresh setups with `channels.telegram.groups["*"].requireMention=true` so new bots stay mention-gated in groups unless you explicitly open them up. Thanks @vincentkoc.
 - Inbound policy hardening: tighten callback and webhook sender checks across Mattermost and Google Chat, match Nextcloud Talk rooms by stable room token, and treat explicit empty Twitch allowlists as deny-all. (#46787) Thanks @zpbrent, @ijxpwastaken and @vincentkoc.
@@ -2756,7 +2905,7 @@ Docs: https://docs.openclaw.ai
 - Gemini OAuth/Auth flow: align OAuth project discovery metadata and endpoint fallback handling for Gemini CLI auth, including fallback coverage for environment-provided project IDs. (#16684) Thanks @vincentkoc.
 - Google Chat/Lifecycle: keep Google Chat `startAccount` pending until abort in webhook mode so startup is no longer interpreted as immediate exit, preventing auto-restart loops and webhook-target churn. (#27384) thanks @junsuwhy.
 - Temp dirs/Linux umask: force `0700` permissions after temp-dir creation and self-heal existing writable temp dirs before trust checks so `umask 0002` installs no longer crash-loop on startup. Landed from contributor PR #27860. (#27853) Thanks @stakeswky.
-- Nextcloud Talk/Lifecycle: keep `startAccount` pending until abort and stop the webhook monitor on shutdown, preventing `EADDRINUSE` restart loops when the gateway manages account lifecycle. (#27897) Thanks @steipete.
+- Nextcloud Talk/Lifecycle: keep `startAccount` pending until abort and stop the webhook monitor on shutdown, preventing `EADDRINUSE` restart loops when the gateway manages account lifecycle. (#27897)
 - Microsoft Teams/File uploads: acknowledge `fileConsent/invoke` immediately (`invokeResponse` before upload + file card send) so Teams no longer shows false "Something went wrong" timeout banners while upload completion continues asynchronously; includes updated async regression coverage. Landed from contributor PR #27641 by @scz2011.
 - Queue/Drain/Cron reliability: harden lane draining with guaranteed `draining` flag reset on synchronous pump failures, reject new queue enqueues during gateway restart drain windows (instead of silently killing accepted tasks), add `/stop` queued-backlog cutoff metadata with stale-message skipping (while avoiding cross-session native-stop cutoff bleed), and raise isolated cron `agentTurn` outer safety timeout to avoid false 10-minute timeout races against longer agent session timeouts. (#27407, #27332, #27427)
 - Typing/Main reply pipeline: always mark dispatch idle in `agent-runner` finalization so typing cleanup runs even when dispatcher `onIdle` does not fire, preventing stuck typing indicators after run completion. (#27250) Thanks @Sid-Qin.
@@ -2773,7 +2922,7 @@ Docs: https://docs.openclaw.ai
 - Agents/Canvas default node resolution: when multiple connected canvas-capable nodes exist and no single `mac-*` candidate is selected, default to the first connected candidate instead of failing with `node required` for implicit-node canvas tool calls. Landed from contributor PR #27444. Thanks @carbaj03.
 - TUI/stream assembly: preserve streamed text across real tool-boundary drops without keeping stale streamed text when non-text blocks appear only in the final payload. Landed from contributor PR #27711 by @scz2011. (#27674)
 - Hooks/Internal `message:sent`: forward `sessionKey` on outbound sends from agent delivery, cron isolated delivery, gateway receipt acks, heartbeat sends, session-maintenance warnings, and restart-sentinel recovery so internal `message:sent` hooks consistently dispatch with session context, including `openclaw agent --deliver` runs resumed via `--session-id` (without explicit `--session-key`). Landed from contributor PR #27584. Thanks @qualiobra.
-- Pi image-token usage: stop re-injecting history image blocks each turn, process image references from the current prompt only, and prune already-answered user-image blocks in stored history to prevent runaway token growth. (#27602) Thanks @steipete.
+- Pi image-token usage: stop re-injecting history image blocks each turn, process image references from the current prompt only, and prune already-answered user-image blocks in stored history to prevent runaway token growth. (#27602)
 - BlueBubbles/SSRF: auto-allowlist the configured `serverUrl` hostname for attachment fetches so localhost/private-IP BlueBubbles setups are no longer false-blocked by default SSRF checks. Landed from contributor PR #27648 by @lailoo. (#27599) Thanks @taylorhou for reporting.
 - Agents/Compaction + onboarding safety: prevent destructive double-compaction by stripping stale assistant usage around compaction boundaries, skipping post-compaction custom metadata writes in the same attempt, and cancelling safeguard compaction when there are no real conversation messages to summarize; harden workspace/bootstrap detection for memory-backed workspaces; and change `openclaw onboard --reset` default scope to `config+creds+sessions` (workspace deletion now requires `--reset-scope full`). (#26458, #27314) Thanks @jaden-clovervnd, @Sid-Qin, and @widingmarcus-cyber for fix direction in #26502, #26529, and #27492.
 - NO_REPLY suppression: suppress `NO_REPLY` before Slack API send and in sub-agent announce completion flow so sentinel text no longer leaks into user channels. Landed from contributor PRs #27529 (by @Sid-Qin) and #27535 (rewritten minimal landing by maintainers). (#27387, #27531)
@@ -2795,7 +2944,7 @@ Docs: https://docs.openclaw.ai
 - LINE/Inline directives auth: gate directive parsing (`/model`, `/think`, `/verbose`, `/reasoning`, `/queue`) on resolved authorization (`command.isAuthorizedSender`) so `commands.allowFrom`-authorized LINE senders are not silently stripped when raw `CommandAuthorized` is unset. Landed from contributor PR #27248 by @kevinWangSheng. (#27240)
 - Onboarding/Gateway: seed default Control UI `allowedOrigins` for non-loopback binds during onboarding (`localhost`/`127.0.0.1` plus custom bind host) so fresh non-loopback setups do not fail startup due to missing origin policy. (#26157) thanks @stakeswky.
 - Docker/GCP onboarding: reduce first-build OOM risk by capping Node heap during `pnpm install`, reuse existing gateway token during `docker-setup.sh` reruns so `.env` stays aligned with config, auto-bootstrap Control UI allowed origins for non-loopback Docker binds, and add GCP docs guidance for tokenized dashboard links + pairing recovery commands. (#26253) Thanks @pandego.
-- CLI/Gateway `--force` in non-root Docker: recover from `lsof` permission failures (`EACCES`/`EPERM`) by falling back to `fuser` kill + probe-based port checks, so `openclaw gateway --force` works for default container `node` user flows. (#27941) Thanks @steipete.
+- CLI/Gateway `--force` in non-root Docker: recover from `lsof` permission failures (`EACCES`/`EPERM`) by falling back to `fuser` kill + probe-based port checks, so `openclaw gateway --force` works for default container `node` user flows. (#27941)
 - Gateway/Bind visibility: emit a startup warning when binding to non-loopback addresses so operators get explicit exposure guidance in runtime logs. (#25397) thanks @let5sne.
 - Sessions cleanup/Doctor: add `openclaw sessions cleanup --fix-missing` to prune store entries whose transcript files are missing, including doctor guidance and CLI coverage. Landed from contributor PR #27508 by @Sid-Qin. (#27422)
 - Doctor/State integrity: ignore metadata-only slash routing sessions when checking recent missing transcripts so `openclaw doctor` no longer reports false-positive transcript-missing warnings for `*:slash:*` keys. (#27375) thanks @gumadeiras.
@@ -2857,24 +3006,24 @@ Docs: https://docs.openclaw.ai
 - Slack/Threading: stop forcing tool-call reply mode to `all` based on `ThreadLabel` alone; now force thread reply mode only when an explicit thread target exists (`MessageThreadId`/`ReplyToId`), so DM `replyToModeByChatType.direct` overrides are honored outside real thread replies. (#26251) Thanks @dbachelder.
 - Slack/Threading: when `replyToMode="all"` auto-threads top-level Slack DMs, seed the thread session key from the message `ts` so the initial message and later replies share the same isolated `:thread:` session instead of falling back to base DM context. (#26849) Thanks @calder-sandy.
 - Agents/Subagents delivery: refactor subagent completion announce dispatch into an explicit queue/direct/fallback state machine, recover outbound channel-plugin resolution in cold/stale plugin-registry states across announce/message/gateway send paths, finalize cleanup bookkeeping when announce flow rejects, and treat Telegram sends without `message_id` as delivery failures (instead of false-success `"unknown"` IDs). (#26867, #25961, #26803, #25069, #26741) Thanks @SmithLabsLLC and @docaohieu2808.
-- Telegram/Webhook: pre-initialize webhook bots, switch webhook processing to callback-mode JSON handling, and preserve full near-limit payload reads under delayed handlers to prevent webhook request hangs and dropped updates. (#26156) Thanks @steipete.
+- Telegram/Webhook: pre-initialize webhook bots, switch webhook processing to callback-mode JSON handling, and preserve full near-limit payload reads under delayed handlers to prevent webhook request hangs and dropped updates. (#26156)
 - Slack/Session threads: prevent oversized parent-session inheritance from silently bricking new thread sessions, surface embedded context-overflow empty-result failures to users, and add configurable `session.parentForkMaxTokens` (default `100000`, `0` disables). (#26912) Thanks @markshields-tl.
 - Cron/Message multi-account routing: honor explicit `delivery.accountId` for isolated cron delivery resolution, and when `message.send` omits `accountId`, fall back to the sending agent's bound channel account instead of defaulting to the global account. (#27015, #26975) Thanks @lbo728 and @stakeswky.
 - Gateway/Message media roots: thread `agentId` through gateway `send` RPC and prefer explicit `agentId` over session/default resolution so non-default agent workspace media sends no longer fail with `LocalMediaAccessError`; added regression coverage for agent precedence and blank-agent fallback. (#23249) Thanks @Sid-Qin.
 - Followups/Routing: when explicit origin routing fails, allow same-channel fallback dispatch (while still blocking cross-channel fallback) so followup replies do not get dropped on transient origin-adapter failures. (#26109) Thanks @Sid-Qin.
-- Cron/Announce duplicate guard: track attempted announce/direct delivery separately from confirmed `delivered`, and suppress fallback main-session cron summaries when delivery was already attempted to avoid duplicate end-user sends in uncertain-ack paths. (#27018) Thanks @steipete.
+- Cron/Announce duplicate guard: track attempted announce/direct delivery separately from confirmed `delivered`, and suppress fallback main-session cron summaries when delivery was already attempted to avoid duplicate end-user sends in uncertain-ack paths. (#27018)
 - LINE/Lifecycle: keep LINE `startAccount` pending until abort so webhook startup is no longer misread as immediate channel exit, preventing restart-loop storms on LINE provider boot. (#26528) Thanks @Sid-Qin.
 - Discord/Gateway: capture and drain startup-time gateway `error` events before lifecycle listeners attach so early `Fatal Gateway error: 4014` closes surface as actionable intent guidance instead of uncaught gateway crashes. (#23832) Thanks @theotarr.
 - Discord/Inbound text: preserve embed `title` + `description` fallback text in message and forwarded snapshot parsing so embed titles are not silently dropped from agent input. (#26946) Thanks @stakeswky.
 - Slack/Inbound media fallback: deliver file-only messages even when Slack media downloads fail by adding a filename placeholder fallback, capping fallback names to the shared media-file limit, and normalizing empty filenames to `file` so attachment-only messages are not silently dropped. (#25181) Thanks @justinhuangcode.
-- Telegram/Preview cleanup: keep finalized text previews when a later assistant message is media-only (for example mixed text plus voice turns) by skipping finalized preview archival at assistant-message boundaries, preventing cleanup from deleting already-visible final text messages. (#27042) Thanks @steipete.
+- Telegram/Preview cleanup: keep finalized text previews when a later assistant message is media-only (for example mixed text plus voice turns) by skipping finalized preview archival at assistant-message boundaries, preventing cleanup from deleting already-visible final text messages. (#27042)
 - Telegram/Markdown spoilers: keep valid `||spoiler||` pairs while leaving unmatched trailing `||` delimiters as literal text, avoiding false all-or-nothing spoiler suppression. (#26105) Thanks @Sid-Qin.
 - Slack/Allowlist channels: match channel IDs case-insensitively during channel allowlist resolution so lowercase config keys (for example `c0abc12345`) correctly match Slack runtime IDs (`C0ABC12345`) under `groupPolicy: "allowlist"`, preventing silent channel-event drops. (#26878) Thanks @lbo728.
 - Discord/Typing indicator: prevent stuck typing indicators by sealing channel typing keepalive callbacks after idle/cleanup and ensuring Discord dispatch always marks typing idle even if preview-stream cleanup fails. (#26295) Thanks @ngutman.
 - Channels/Typing indicator: guard typing keepalive start callbacks after idle/cleanup close so post-close ticks cannot re-trigger stale typing indicators. (#26325) Thanks @win4r.
 - Followups/Typing indicator: ensure followup turns mark dispatch idle on every exit path (including `NO_REPLY`, empty payloads, and agent errors) so typing keepalive cleanup always runs and channel typing indicators do not get stuck after queued/silent followups. (#26881) Thanks @codexGW.
-- Voice-call/TTS tools: hide the `tts` tool when the message provider is `voice`, preventing voice-call runs from selecting self-playback TTS and falling into silent no-output loops. (#27025) Thanks @steipete.
-- Agents/Tools: normalize non-standard plugin tool results that omit `content` so embedded runs no longer crash with `Cannot read properties of undefined (reading 'filter')` after tool completion (including `tesseramemo_query`). (#27007) Thanks @steipete.
+- Voice-call/TTS tools: hide the `tts` tool when the message provider is `voice`, preventing voice-call runs from selecting self-playback TTS and falling into silent no-output loops. (#27025)
+- Agents/Tools: normalize non-standard plugin tool results that omit `content` so embedded runs no longer crash with `Cannot read properties of undefined (reading 'filter')` after tool completion (including `tesseramemo_query`). (#27007)
 - Agents/Tool-call dispatch: trim whitespace-padded tool names in both transcript repair and live streamed embedded-runner responses so exact-match tool lookup no longer fails with `Tool ... not found` for model outputs like `" read "`. (#27094) Thanks @openperf and @Sid-Qin.
 - Cron/Model overrides: when isolated `payload.model` is no longer allowlisted, fall back to default model selection instead of failing the job, while still returning explicit errors for invalid model strings. (#26717) Thanks @Youyou972.
 - Agents/Model fallback: keep explicit text + image fallback chains reachable even when `agents.defaults.models` allowlists are present, prefer explicit run `agentId` over session-key parsing for followup fallback override resolution (with session-key fallback), treat agent-level fallback overrides as configured in embedded runner preflight, and classify `model_cooldown` / `cooling down` errors as `rate_limit` so failover continues. (#11972, #24137, #17231)
@@ -2920,7 +3069,7 @@ Docs: https://docs.openclaw.ai
 
 ### Changes
 
-- Auto-reply/Abort shortcuts: expand standalone stop phrases (`stop openclaw`, `stop action`, `stop run`, `stop agent`, `please stop`, and related variants), accept trailing punctuation (for example `STOP OPENCLAW!!!`), add multilingual stop keywords (including ES/FR/ZH/HI/AR/JP/DE/PT/RU forms), and treat exact `do not do that` as a stop trigger while preserving strict standalone matching. (#25103) Thanks @steipete and @vincentkoc.
+- Auto-reply/Abort shortcuts: expand standalone stop phrases (`stop openclaw`, `stop action`, `stop run`, `stop agent`, `please stop`, and related variants), accept trailing punctuation (for example `STOP OPENCLAW!!!`), add multilingual stop keywords (including ES/FR/ZH/HI/AR/JP/DE/PT/RU forms), and treat exact `do not do that` as a stop trigger while preserving strict standalone matching. (#25103) Thanks @vincentkoc.
 - Android/App UX: ship a native four-step onboarding flow, move post-onboarding into a five-tab shell (Connect, Chat, Voice, Screen, Settings), add a full Connect setup/manual mode screen, and refresh Android chat/settings surfaces for the new navigation model.
 - Talk/Gateway config: add provider-agnostic Talk configuration with legacy compatibility, and expose gateway Talk ElevenLabs config metadata for setup/status surfaces.
 - Security/Audit: add `security.trust_model.multi_user_heuristic` to flag likely shared-user ingress and clarify the personal-assistant trust model, with hardening guidance for intentional multi-user setups (`sandbox.mode="all"`, workspace-scoped FS, reduced tool surface, no personal/private identities on shared runtimes).
@@ -2930,7 +3079,7 @@ Docs: https://docs.openclaw.ai
 
 - Routing/Session isolation: harden followup routing so explicit cross-channel origin replies never fall back to the active dispatcher on route failure, preserve queued overflow summary routing metadata (`channel`/`to`/`thread`) across followup drain, and prefer originating channel context over internal provider tags for embedded followup runs. This prevents webchat/control-ui context from hijacking Discord-targeted replies in shared sessions. (#25864) Thanks @Gamedesigner.
 - Security/Routing: fail closed for shared-session cross-channel replies by binding outbound target resolution to the current turn's source channel metadata (instead of stale session route fallbacks), and wire those turn-source fields through gateway + command delivery planners with regression coverage. (#24571) Thanks @brandonwise.
-- Heartbeat routing: prevent heartbeat leakage/spam into Discord and other direct-message destinations by blocking direct-chat heartbeat delivery targets and keeping blocked-delivery cron/exec prompts internal-only. (#25871) Thanks @steipete.
+- Heartbeat routing: prevent heartbeat leakage/spam into Discord and other direct-message destinations by blocking direct-chat heartbeat delivery targets and keeping blocked-delivery cron/exec prompts internal-only. (#25871)
 - Heartbeat defaults/prompts: switch the implicit heartbeat delivery target from `last` to `none` (opt-in for external delivery), and use internal-only cron/exec heartbeat prompt wording when delivery is disabled so background checks do not nudge user-facing relay behavior. (#25871, #24638, #25851)
 - Auto-reply/Heartbeat queueing: drop heartbeat runs when a session already has an active run instead of enqueueing a stale followup, preventing duplicate heartbeat response branches after queue drain. (#25610, #25606) Thanks @mcaxtr.
 - Cron/Heartbeat delivery: stop inheriting cached session `lastThreadId` for heartbeat-mode target resolution unless a thread/topic is explicitly requested, so announce-mode cron and heartbeat deliveries stay on top-level destinations instead of leaking into active conversation threads. (#25730) Thanks @markshields-tl.
@@ -2963,7 +3112,7 @@ Docs: https://docs.openclaw.ai
 - Windows/Media safety checks: align async local-file identity validation with sync-safe-open behavior by treating win32 `dev=0` stats as unknown-device fallbacks (while keeping strict dev checks when both sides are non-zero), fixing false `Local media path is not safe to read` drops for local attachments/TTS/images. (#25708, #21989, #25699, #25878) Thanks @kevinWangSheng.
 - iMessage/Reasoning safety: harden iMessage echo suppression with outbound `messageId` matching (plus scoped text fallback), and enforce reasoning-payload suppression on routed outbound delivery paths to prevent hidden thinking text from being sent as user-visible channel messages. (#25897, #1649, #25757) Thanks @rmarr and @Iranb.
 - Providers/OpenRouter/Auth profiles: bypass auth-profile cooldown/disable windows for OpenRouter, so provider failures no longer put OpenRouter profiles into local cooldown and stale legacy cooldown markers are ignored in fallback and status selection paths. (#25892) Thanks @alexanderatallah for raising this and @vincentkoc for the fix.
-- Providers/Google reasoning: sanitize invalid negative `thinkingBudget` payloads for Gemini 3.1 requests by dropping `-1` budgets and mapping configured reasoning effort to `thinkingLevel`, preventing malformed reasoning payloads on `google-generative-ai`. (#25900) Thanks @steipete.
+- Providers/Google reasoning: sanitize invalid negative `thinkingBudget` payloads for Gemini 3.1 requests by dropping `-1` budgets and mapping configured reasoning effort to `thinkingLevel`, preventing malformed reasoning payloads on `google-generative-ai`. (#25900)
 - Providers/SiliconFlow: normalize `thinking="off"` to `thinking: null` for `Pro/*` model payloads to avoid provider-side 400 loops and misleading compaction retries. (#25435) Thanks @Zjianru.
 - Models/Bedrock auth: normalize additional Bedrock provider aliases (`bedrock`, `aws-bedrock`, `aws_bedrock`, `amazon bedrock`) to canonical `amazon-bedrock`, ensuring auth-mode resolution consistently selects AWS SDK fallback. (#25756) Thanks @fwhite13.
 - Models/Providers: preserve explicit user `reasoning` overrides when merging provider model config with built-in catalog metadata, so `reasoning: false` is no longer overwritten by catalog defaults. (#25314) Thanks @lbo728.
@@ -3059,7 +3208,7 @@ Docs: https://docs.openclaw.ai
 - Providers/Groq: avoid classifying Groq TPM limit errors as context overflow so throttling paths no longer trigger overflow recovery logic. (#16176) Thanks @dddabtc.
 - Gateway/Restart: treat child listener PIDs as owned by the service runtime PID during restart health checks to avoid false stale-process kills and restart timeouts on launchd/systemd. (#24696) Thanks @gumadeiras.
 - Config/Write: apply `unsetPaths` with immutable path-copy updates so config writes never mutate caller-provided objects, and harden `openclaw config get/set/unset` path traversal by rejecting prototype-key segments and inherited-property traversal. (#24134) thanks @frankekn.
-- Channels/WhatsApp: accept `channels.whatsapp.enabled` in config validation to match built-in channel auto-enable behavior, preventing `Unrecognized key: "enabled"` failures during channel setup. (#24263) Thanks @steipete.
+- Channels/WhatsApp: accept `channels.whatsapp.enabled` in config validation to match built-in channel auto-enable behavior, preventing `Unrecognized key: "enabled"` failures during channel setup. (#24263)
 - Security/Exec: detect obfuscated commands before exec allowlist decisions and require explicit approval for obfuscation patterns. (#8592) Thanks @CornBrother0x and @vincentkoc.
 - Security/ACP: harden ACP client permission auto-approval to require trusted core tool IDs, ignore untrusted `toolCall.kind` hints, and scope `read` auto-approval to the active working directory so unknown tool names and out-of-scope file reads always prompt. Thanks @nedlir for reporting.
 - Security/Skills: escape user-controlled prompt, filename, and output-path values in `openai-image-gen` HTML gallery generation to prevent stored XSS in generated `index.html` output. (#12538) Thanks @CornBrother0x.
@@ -3084,7 +3233,7 @@ Docs: https://docs.openclaw.ai
 - Update/Core: add an optional built-in auto-updater for package installs (`update.auto.*`), default-off, with stable rollout delay+jitter and beta hourly cadence.
 - CLI/Update: add `openclaw update --dry-run` to preview channel/tag/target/restart actions without mutating config, installing, syncing plugins, or restarting.
 - Config/UI: add tag-aware settings filtering and broaden config labels/help copy so fields are easier to discover and understand in the dashboard config screen.
-- Channels/Synology Chat: add a native Synology Chat channel plugin with webhook ingress, direct-message routing, outbound send/media support, per-account config, and DM policy controls. (#23012) Thanks @steipete.
+- Channels/Synology Chat: add a native Synology Chat channel plugin with webhook ingress, direct-message routing, outbound send/media support, per-account config, and DM policy controls. (#23012)
 - iOS/Talk: prefetch TTS segments and suppress expected speech-cancellation errors for smoother talk playback. (#22833) Thanks @ngutman.
 - Memory/FTS: add Spanish and Portuguese stop-word filtering for query expansion in FTS-only search mode, improving conversational recall for both languages. Thanks @vincentkoc.
 - Memory/FTS: add Japanese-aware query expansion tokenization and stop-word filtering (including mixed-script terms like ASCII + katakana) for FTS-only search mode. Thanks @vincentkoc.
@@ -3106,10 +3255,10 @@ Docs: https://docs.openclaw.ai
 - Agents/Moonshot: force `supportsDeveloperRole=false` for Moonshot-compatible `openai-completions` models (provider `moonshot` and Moonshot base URLs), so initial runs no longer send unsupported `developer` roles that trigger `ROLE_UNSPECIFIED` errors. (#21060, #22194) Thanks @ShengFuC.
 - Agents/Kimi: classify Moonshot `Your request exceeded model token limit` failures as context overflows so auto-compaction and user-facing overflow recovery trigger correctly instead of surfacing raw invalid-request errors. (#9562) Thanks @danilofalcao.
 - Providers/Moonshot: mark Kimi K2.5 as image-capable in implicit + onboarding model definitions, and refresh stale explicit provider capability fields (`input`/`reasoning`/context limits) from implicit catalogs so existing configs pick up Moonshot vision support without manual model rewrites. (#13135, #4459) Thanks @manikv12.
-- Agents/Transcript: enable consecutive-user turn merging for strict non-OpenAI `openai-completions` providers (for example Moonshot/Kimi), reducing `roles must alternate` ordering failures on OpenAI-compatible endpoints while preserving current OpenRouter/Opencode behavior. (#7693) Thanks @steipete.
+- Agents/Transcript: enable consecutive-user turn merging for strict non-OpenAI `openai-completions` providers (for example Moonshot/Kimi), reducing `roles must alternate` ordering failures on OpenAI-compatible endpoints while preserving current OpenRouter/Opencode behavior. (#7693)
 - Install/Discord Voice: make the native Opus decoder optional so `openclaw` install/update no longer hard-fails when native builds fail, while keeping `opusscript` as the runtime fallback decoder for Discord voice flows. (#23737, #23733, #23703) Thanks @jeadland, @Sheetaa, and @Breakyman.
 - Docker/Setup: precreate `$OPENCLAW_CONFIG_DIR/identity` during `docker-setup.sh` so CLI commands that need device identity (for example `devices list`) avoid `EACCES ... /home/node/.openclaw/identity` failures on restrictive bind mounts. (#23948) Thanks @ackson-beep.
-- Exec/Background: stop applying the default exec timeout to background sessions (`background: true` or explicit `yieldMs`) when no explicit timeout is set, so long-running background jobs are no longer terminated at the default timeout boundary. (#23303) Thanks @steipete.
+- Exec/Background: stop applying the default exec timeout to background sessions (`background: true` or explicit `yieldMs`) when no explicit timeout is set, so long-running background jobs are no longer terminated at the default timeout boundary. (#23303)
 - Slack/Threading: sessions: keep parent-session forking and thread-history context active beyond first turn by removing first-turn-only gates in session init, thread-history fetch, and reply prompt context injection. (#23843, #23090) Thanks @vincentkoc and @Taskle.
 - Slack/Threading: respect `replyToMode` when Slack auto-populates top-level `thread_ts`, and ignore inline `replyToId` directive tags when `replyToMode` is `off` so thread forcing stays disabled unless explicitly configured. (#23839, #23320, #23513) Thanks @vincentkoc and @dorukardahan.
 - Slack/Extension: forward `message read` `threadId` to `readMessages` and use delivery-context `threadId` as outbound `thread_ts` fallback so extension replies/reads stay in the correct Slack thread. (#22216, #22485, #23836) Thanks @vincentkoc, @lan17 and @dorukardahan.
@@ -3129,7 +3278,7 @@ Docs: https://docs.openclaw.ai
 - Telegram/Webhook: add `channels.telegram.webhookPort` config support and pass it through plugin startup wiring to the monitor listener.
 - Browser/Extension Relay: refactor the MV3 worker to preserve debugger attachments across relay drops, auto-reconnect with bounded backoff+jitter, persist and rehydrate attached tab state via `chrome.storage.session`, recover from `target_closed` navigation detaches, guard stale socket handlers, enforce per-tab operation locks and per-request timeouts, and add lifecycle keepalive/badge refresh hooks (`alarms`, `webNavigation`). (#15099, #6175, #8468, #9807)
 - Browser/Relay: treat extension websocket as connected only when `OPEN`, allow reconnect when a stale `CLOSING/CLOSED` extension socket lingers, and guard stale socket message/close handlers so late events cannot clear active relay state; includes regression coverage for live-duplicate `409` rejection and immediate reconnect-after-close races. (#15099, #18698, #20688)
-- Browser/Remote CDP: extend stale-target recovery so `ensureTabAvailable()` now reuses the sole available tab for remote CDP profiles (same behavior as extension profiles) while preserving strict `tab not found` errors when multiple tabs exist; includes remote-profile regression tests. (#15989) Thanks @steipete.
+- Browser/Remote CDP: extend stale-target recovery so `ensureTabAvailable()` now reuses the sole available tab for remote CDP profiles (same behavior as extension profiles) while preserving strict `tab not found` errors when multiple tabs exist; includes remote-profile regression tests. (#15989)
 - Gateway/Pairing: treat `operator.admin` as satisfying other `operator.*` scope checks during device-auth verification so local CLI/TUI sessions stop entering pairing-required loops for pairing/approval-scoped commands. (#22062, #22193, #21191) Thanks @Botaccess, @jhartshorn, and @ctbritt.
 - Gateway/Pairing: auto-approve loopback `scope-upgrade` pairing requests (including device-token reconnects) so local clients do not disconnect on pairing-required scope elevation. (#23708) Thanks @widingmarcus-cyber.
 - Gateway/Scopes: include `operator.read` and `operator.write` in default operator connect scope bundles across CLI, Control UI, and macOS clients so write-scoped announce/sub-agent follow-up calls no longer hit `pairing required` disconnects on loopback gateways. (#22582) thanks @YuzuruS.
@@ -3167,25 +3316,25 @@ Docs: https://docs.openclaw.ai
 - Security/Group policy: harden `channels.*.groups.*.toolsBySender` matching by requiring explicit sender-key types (`id:`, `e164:`, `username:`, `name:`), preventing cross-identifier collisions across mutable/display-name fields while keeping legacy untyped keys on a deprecated ID-only path. Thanks @jiseoung for reporting.
 - Channels/Group policy: fail closed when `groupPolicy: "allowlist"` is set without explicit `groups`, honor account-level `groupPolicy` overrides, and enforce `groupPolicy: "disabled"` as a hard group block. (#22215) Thanks @etereo.
 - Telegram/Discord extensions: propagate trusted `mediaLocalRoots` through extension outbound `sendMedia` options so extension direct-send media paths honor agent-scoped local-media allowlists. (#20029, #21903, #23227)
-- Agents/Exec: honor explicit agent context when resolving `tools.exec` defaults for runs with opaque/non-agent session keys, so per-agent `host/security/ask` policies are applied consistently. (#11832) Thanks @steipete.
+- Agents/Exec: honor explicit agent context when resolving `tools.exec` defaults for runs with opaque/non-agent session keys, so per-agent `host/security/ask` policies are applied consistently. (#11832)
 - CLI/Sessions: resolve implicit session-store path templates with the configured default agent ID so named-agent setups do not silently read/write stale `agent:main` session/auth stores. (#22685) Thanks @sene1337.
-- Doctor/Security: add an explicit warning that `approvals.exec.enabled=false` disables forwarding only, while enforcement remains driven by host-local `exec-approvals.json` policy. (#15047) Thanks @steipete.
-- Sandbox/Docker: default sandbox container user to the workspace owner `uid:gid` when `agents.*.sandbox.docker.user` is unset, fixing non-root gateway file-tool permissions under capability-dropped containers. (#20979) Thanks @steipete.
+- Doctor/Security: add an explicit warning that `approvals.exec.enabled=false` disables forwarding only, while enforcement remains driven by host-local `exec-approvals.json` policy. (#15047)
+- Sandbox/Docker: default sandbox container user to the workspace owner `uid:gid` when `agents.*.sandbox.docker.user` is unset, fixing non-root gateway file-tool permissions under capability-dropped containers. (#20979)
 - Plugins/Media sandbox: propagate trusted `mediaLocalRoots` through plugin action dispatch (including Discord/Telegram action adapters) so plugin send paths enforce the same agent-scoped local-media sandbox roots as core outbound sends. (#20258, #22718)
-- Agents/Workspace guard: map sandbox container-workdir file-tool paths (for example `/workspace/...` and `file:///workspace/...`) to host workspace roots before workspace-only validation, preventing false `Path escapes sandbox root` rejections for sandbox file tools. (#9560) Thanks @steipete.
-- Gateway/Exec approvals: expire approval requests immediately when no approval-capable gateway clients are connected and no forwarding targets are available, avoiding delayed approvals after restarts/offline approver windows. (#22144) Thanks @steipete.
+- Agents/Workspace guard: map sandbox container-workdir file-tool paths (for example `/workspace/...` and `file:///workspace/...`) to host workspace roots before workspace-only validation, preventing false `Path escapes sandbox root` rejections for sandbox file tools. (#9560)
+- Gateway/Exec approvals: expire approval requests immediately when no approval-capable gateway clients are connected and no forwarding targets are available, avoiding delayed approvals after restarts/offline approver windows. (#22144)
 - Security/Exec approvals: when approving wrapper commands with allow-always in allowlist mode, persist inner executable paths for known dispatch wrappers (`env`, `nice`, `nohup`, `stdbuf`, `timeout`) and fail closed (no persisted entry) when wrapper unwrapping is not safe, preventing wrapper-path approval bypasses. Thanks @tdjackey for reporting.
-- Node/macOS exec host: default headless macOS node `system.run` to local execution and only route through the companion app when `OPENCLAW_NODE_EXEC_HOST=app` is explicitly set, avoiding companion-app filesystem namespace mismatches during exec. (#23547) Thanks @steipete.
+- Node/macOS exec host: default headless macOS node `system.run` to local execution and only route through the companion app when `OPENCLAW_NODE_EXEC_HOST=app` is explicitly set, avoiding companion-app filesystem namespace mismatches during exec. (#23547)
 - Sandbox/Media: map container workspace paths (`/workspace/...` and `file:///workspace/...`) back to the host sandbox root for outbound media validation, preventing false deny errors for sandbox-generated local media. (#23083) Thanks @echo931.
 - Sandbox/Docker: apply custom bind mounts after workspace mounts and prioritize bind-source resolution on overlapping paths, so explicit workspace binds are no longer ignored. (#22669) Thanks @tasaankaeris.
 - Exec approvals/Forwarding: restore Discord text forwarding when component approvals are not configured, and carry request snapshots through resolve events so resolved notices still forward after cache misses/restarts. (#22988) Thanks @bubmiller.
 - Control UI/WebSocket: stop and clear the browser gateway client on UI teardown so remounts cannot leave orphan websocket clients that create duplicate active connections. (#23422) Thanks @floatinggball-design.
 - Control UI/WebSocket: send a stable per-tab `instanceId` in websocket connect frames so reconnect cycles keep a consistent client identity for diagnostics and presence tracking. (#23616) Thanks @zq58855371-ui.
 - Config/Memory: allow `"mistral"` in `agents.defaults.memorySearch.provider` and `agents.defaults.memorySearch.fallback` schema validation. (#14934) Thanks @ThomsenDrake.
-- Feishu/Commands: in group chats, command authorization now falls back to top-level `channels.feishu.allowFrom` when per-group `allowFrom` is not set, so `/command` no longer gets blocked by an unintended empty allowlist. (#23756) Thanks @steipete.
+- Feishu/Commands: in group chats, command authorization now falls back to top-level `channels.feishu.allowFrom` when per-group `allowFrom` is not set, so `/command` no longer gets blocked by an unintended empty allowlist. (#23756)
 - Dev tooling: prevent `CLAUDE.md` symlink target regressions by excluding CLAUDE symlink sentinels from `oxfmt` and marking them `-text` in `.gitattributes`, so formatter/EOL normalization cannot reintroduce trailing-newline targets. Thanks @vincentkoc.
 - Agents/Compaction: restore embedded compaction safeguard/context-pruning extension loading in production by wiring bundled extension factories into the resource loader instead of runtime file-path resolution. (#22349; landed from contributor PR #5005 by @Diaspar4u) Thanks @Diaspar4u.
-- Feishu/Media: for inbound video messages that include both `file_key` (video) and `image_key` (thumbnail), prefer `file_key` when downloading media so video attachments are saved instead of silently failing on thumbnail keys. (#23633) Thanks @steipete.
+- Feishu/Media: for inbound video messages that include both `file_key` (video) and `image_key` (thumbnail), prefer `file_key` when downloading media so video attachments are saved instead of silently failing on thumbnail keys. (#23633)
 - Hooks/Loader: avoid redundant hook-module recompilation on gateway restart by skipping cache-busting for bundled hooks and using stable file metadata keys (`mtime+size`) for mutable workspace/managed/plugin hook imports. (#16953) Thanks @mudrii.
 - Hooks/Cron: suppress duplicate main-session events for delivered hook turns and mark `SILENT_REPLY_TOKEN` (`NO_REPLY`) early exits as delivered to prevent hook context pollution. (#20678) Thanks @JonathanWorks.
 - Providers/OpenRouter: inject `cache_control` on system prompts for OpenRouter Anthropic models to improve prompt-cache reuse. (#17473) Thanks @rrenamed.
@@ -3664,7 +3813,7 @@ Docs: https://docs.openclaw.ai
 - Ollama/Qwen: handle Qwen 3 reasoning field format in Ollama responses. (#18631) Thanks @mr-sk.
 - OpenAI/Transcripts: always drop orphaned reasoning blocks from transcript repair. (#18632) Thanks @TySabs.
 - Fix types in all tests. Typecheck the whole repository.
-- Gateway/Channels: wire `gateway.channelHealthCheckMinutes` into strict config validation, treat implicit account status as managed for health checks, and harden channel auto-restart flow (preserve restart-attempt caps across crash loops, propagate enabled/configured runtime flags, and stop pending restart backoff after manual stop). Thanks @steipete.
+- Gateway/Channels: wire `gateway.channelHealthCheckMinutes` into strict config validation, treat implicit account status as managed for health checks, and harden channel auto-restart flow (preserve restart-attempt caps across crash loops, propagate enabled/configured runtime flags, and stop pending restart backoff after manual stop).
 - Gateway/WebChat: hard-cap `chat.history` oversized payloads by truncating high-cost fields and replacing over-budget entries with placeholders, so history fetches stay within configured byte limits and avoid chat UI freezes. (#18505)
 - UI/Usage: replace lingering undefined `var(--text-muted)` usage with `var(--muted)` in usage date-range and chart styles to keep muted text visible across themes. (#17975) Thanks @jogelin.
 - UI/Usage: preserve selected-range totals when timeline data is downsampled by bucket-aggregating timeseries points (instead of dropping intermediate points), so filtered tokens/cost stay accurate. (#17959) Thanks @jogelin.
@@ -4674,21 +4823,21 @@ Docs: https://docs.openclaw.ai
 
 - Providers: Ollama discovery + docs; Venice guide upgrades + cross-links. (#1606) Thanks @abhaymundhara. https://docs.openclaw.ai/providers/ollama https://docs.openclaw.ai/providers/venice
 - Channels: LINE plugin (Messaging API) with rich replies + quick replies. (#1630) Thanks @plum-dawg.
-- TTS: Edge fallback (keyless) + `/tts` auto modes. (#1668, #1667) Thanks @steipete, @sebslight. https://docs.openclaw.ai/tts
+- TTS: Edge fallback (keyless) + `/tts` auto modes. (#1668, #1667) Thanks @sebslight. https://docs.openclaw.ai/tts
 - Exec approvals: approve in-chat via `/approve` across all channels (including plugins). (#1621) Thanks @czekaj. https://docs.openclaw.ai/tools/exec-approvals https://docs.openclaw.ai/tools/slash-commands
 - Telegram: DM topics as separate sessions + outbound link preview toggle. (#1597, #1700) Thanks @rohannagpal, @zerone0x. https://docs.openclaw.ai/channels/telegram
 
 ### Changes
 
 - Channels: add LINE plugin (Messaging API) with rich replies, quick replies, and plugin HTTP registry. (#1630) Thanks @plum-dawg.
-- TTS: add Edge TTS provider fallback, defaulting to keyless Edge with MP3 retry on format failures. (#1668) Thanks @steipete. https://docs.openclaw.ai/tts
+- TTS: add Edge TTS provider fallback, defaulting to keyless Edge with MP3 retry on format failures. (#1668) https://docs.openclaw.ai/tts
 - TTS: add auto mode enum (off/always/inbound/tagged) with per-session `/tts` override. (#1667) Thanks @sebslight. https://docs.openclaw.ai/tts
 - Telegram: treat DM topics as separate sessions and keep DM history limits stable with thread suffixes. (#1597) Thanks @rohannagpal.
 - Telegram: add `channels.telegram.linkPreview` to toggle outbound link previews. (#1700) Thanks @zerone0x. https://docs.openclaw.ai/channels/telegram
 - Web search: add Brave freshness filter parameter for time-scoped results. (#1688) Thanks @JonUleis. https://docs.openclaw.ai/tools/web
 - UI: refresh Control UI dashboard design system (colors, icons, typography). (#1745, #1786) Thanks @EnzeD, @mousberg.
 - Exec approvals: forward approval prompts to chat with `/approve` for all channels (including plugins). (#1621) Thanks @czekaj. https://docs.openclaw.ai/tools/exec-approvals https://docs.openclaw.ai/tools/slash-commands
-- Gateway: expose config.patch in the gateway tool with safe partial updates + restart sentinel. (#1653) Thanks @steipete.
+- Gateway: expose config.patch in the gateway tool with safe partial updates + restart sentinel. (#1653)
 - Diagnostics: add diagnostic flags for targeted debug logs (config + env override). https://docs.openclaw.ai/diagnostics/flags
 - Docs: expand FAQ (migration, scheduling, concurrency, model recommendations, OpenAI subscription auth, Pi sizing, hackable install, docs SSL workaround).
 - Docs: add verbose installer troubleshooting guidance.
@@ -4701,9 +4850,9 @@ Docs: https://docs.openclaw.ai
 
 - Web UI: fix config/debug layout overflow, scrolling, and code block sizing. (#1715) Thanks @saipreetham589.
 - Web UI: show Stop button during active runs, swap back to New session when idle. (#1664) Thanks @ndbroadbent.
-- Web UI: clear stale disconnect banners on reconnect; allow form saves with unsupported schema paths but block missing schema. (#1707) Thanks @steipete.
+- Web UI: clear stale disconnect banners on reconnect; allow form saves with unsupported schema paths but block missing schema. (#1707)
 - Web UI: hide internal `message_id` hints in chat bubbles.
-- Gateway: allow Control UI token-only auth to skip device pairing even when device identity is present (`gateway.controlUi.allowInsecureAuth`). (#1679) Thanks @steipete.
+- Gateway: allow Control UI token-only auth to skip device pairing even when device identity is present (`gateway.controlUi.allowInsecureAuth`). (#1679)
 - Matrix: decrypt E2EE media attachments with preflight size guard. (#1744) Thanks @araa47.
 - BlueBubbles: route phone-number targets to DMs, avoid leaking routing IDs, and auto-create missing DMs (Private API required). (#1751) Thanks @tyler6204. https://docs.openclaw.ai/channels/bluebubbles
 - BlueBubbles: keep part-index GUIDs in reply tags when short IDs are missing.
@@ -4776,7 +4925,7 @@ Docs: https://docs.openclaw.ai
 - Heartbeat: accept plugin channel ids for heartbeat target validation + UI hints.
 - Messaging/Sessions: mirror outbound sends into target session keys (threads + dmScope), create session entries on send, and normalize session key casing. (#1520, commit 4b6cdd1d3)
 - Sessions: reject array-backed session stores to prevent silent wipes. (#1469)
-- Gateway: compare Linux process start time to avoid PID recycling lock loops; keep locks unless stale. (#1572) Thanks @steipete.
+- Gateway: compare Linux process start time to avoid PID recycling lock loops; keep locks unless stale. (#1572)
 - Gateway: accept null optional fields in exec approval requests. (#1511) Thanks @pvoo.
 - Exec approvals: persist allowlist entry ids to keep macOS allowlist rows stable. (#1521) Thanks @ngutman.
 - Exec: honor tools.exec ask/security defaults for elevated approvals (avoid unwanted prompts). (commit 5662a9cdf)
@@ -5082,7 +5231,7 @@ Docs: https://docs.openclaw.ai
 - macOS: bundle Textual resources in packaged app builds to avoid code block crashes. (#1006)
 - Daemon: include HOME in service environments to avoid missing HOME errors. (#1214)
 
-Thanks @AlexMikhalev, @CoreyH, @John-Rood, @KrauseFx, @MaudeBot, @Nachx639, @NicholaiVogel, @RyanLisse, @ThePickle31, @VACInc, @Whoaa512, @YuriNachos, @aaronveklabs, @abdaraxus, @alauppe, @ameno-, @artuskg, @austinm911, @bradleypriest, @cheeeee, @dougvk, @fogboots, @gnarco, @gumadeiras, @jdrhyne, @joelklabo, @longmaba, @mukhtharcm, @odysseus0, @oscargavin, @rhjoh, @sebslight, @sibbl, @sleontenko, @steipete, @suminhthanh, @thewilloftheshadow, @tyler6204, @vignesh07, @visionik, @ysqander, @zerone0x.
+Thanks @AlexMikhalev, @CoreyH, @John-Rood, @KrauseFx, @MaudeBot, @Nachx639, @NicholaiVogel, @RyanLisse, @ThePickle31, @VACInc, @Whoaa512, @YuriNachos, @aaronveklabs, @abdaraxus, @alauppe, @ameno-, @artuskg, @austinm911, @bradleypriest, @cheeeee, @dougvk, @fogboots, @gnarco, @gumadeiras, @jdrhyne, @joelklabo, @longmaba, @mukhtharcm, @odysseus0, @oscargavin, @rhjoh, @sebslight, @sibbl, @sleontenko, @suminhthanh, @thewilloftheshadow, @tyler6204, @vignesh07, @visionik, @ysqander, @zerone0x.
 
 ### Breaking
 
@@ -5451,7 +5600,7 @@ Thanks @AlexMikhalev, @CoreyH, @John-Rood, @KrauseFx, @MaudeBot, @Nachx639, @Nic
 - Gateway/CLI: honor `CLAWDBOT_LAUNCHD_LABEL` / `CLAWDBOT_SYSTEMD_UNIT` overrides; `agents.list` respects explicit config; reduce noisy loopback WS logs during tests; run `openclaw doctor --non-interactive` during updates. (#781) - thanks @ronyrus.
 - Onboarding/Control UI: refuse invalid configs (run doctor first); quote Windows browser URLs for OAuth; keep chat scroll position unless the user is near the bottom. (#764) - thanks @mukhtharcm; (#794) - thanks @roshanasingh4; (#217) - thanks @thewilloftheshadow.
 - Tools/UI: harden tool input schemas for strict providers; drop null-only union variants for Gemini schema cleanup; treat `maxChars: 0` as unlimited; keep TUI last streamed response instead of "(no output)". (#782) - thanks @AbhisekBasu1; (#796) - thanks @gabriel-trigo; (#747) - thanks @thewilloftheshadow.
-- Connections UI: polish multi-account account cards. (#816) - thanks @steipete.
+- Connections UI: polish multi-account account cards. (#816)
 
 ### Installer
 
@@ -5491,7 +5640,7 @@ Thanks @AlexMikhalev, @CoreyH, @John-Rood, @KrauseFx, @MaudeBot, @Nachx639, @Nic
 - Tests: add Docker plugin loader + tgz-install smoke test.
 - Tests: extend Docker plugin E2E to cover installing from local folders (`plugins.load.paths`) and `file:` npm specs.
 - Tests: add coverage for pre-compaction memory flush settings.
-- Tests: modernize live model smoke selection for current releases and enforce tools/images/thinking-high coverage. (#769) - thanks @steipete.
+- Tests: modernize live model smoke selection for current releases and enforce tools/images/thinking-high coverage. (#769)
 - Agents/Tools: add `apply_patch` tool for multi-file edits (experimental; gated by tools.exec.applyPatch; OpenAI-only).
 - Agents/Tools: rename the bash tool to exec (config alias maintained). (#748) - thanks @myfunc.
 - Agents: add pre-compaction memory flush config (`agents.defaults.compaction.*`) with a soft threshold + system prompt.
@@ -5511,8 +5660,8 @@ Thanks @AlexMikhalev, @CoreyH, @John-Rood, @KrauseFx, @MaudeBot, @Nachx639, @Nic
 ### Fixes
 
 - Models/Onboarding: configure MiniMax (minimax.io) via Anthropic-compatible `/anthropic` endpoint by default (keep `minimax-api` as a legacy alias).
-- Models: normalize Gemini 3 Pro/Flash IDs to preview names for live model lookups. (#769) - thanks @steipete.
-- CLI: fix guardCancel typing for configure prompts. (#769) - thanks @steipete.
+- Models: normalize Gemini 3 Pro/Flash IDs to preview names for live model lookups. (#769)
+- CLI: fix guardCancel typing for configure prompts. (#769)
 - Gateway/WebChat: include handshake validation details in the WebSocket close reason for easier debugging; preserve close codes.
 - Gateway/Auth: send invalid connect responses before closing the handshake; stabilize invalid-connect auth test.
 - Gateway: tighten gateway listener detection.
@@ -5529,7 +5678,7 @@ Thanks @AlexMikhalev, @CoreyH, @John-Rood, @KrauseFx, @MaudeBot, @Nachx639, @Nic
 - Auto-reply: align `/think` default display with model reasoning defaults. (#751) - thanks @gabriel-trigo.
 - Auto-reply: flush block reply buffers on tool boundaries. (#750) - thanks @sebslight.
 - Auto-reply: allow sender fallback for command authorization when `SenderId` is empty (WhatsApp self-chat). (#755) - thanks @juanpablodlc.
-- Auto-reply: treat whitespace-only sender ids as missing for command authorization (WhatsApp self-chat). (#766) - thanks @steipete.
+- Auto-reply: treat whitespace-only sender ids as missing for command authorization (WhatsApp self-chat). (#766)
 - Heartbeat: refresh prompt text for updated defaults.
 - Memory/QMD: prefer `qmd collection add --glob` for current QMD releases and fall back to legacy `--mask` when older builds reject it. (#55123) Thanks @ForceConstant and @vincentkoc.
 - Agents/Tools: use PowerShell on Windows to capture system utility output. (#748) - thanks @myfunc.

CONTRIBUTING.md

@@ -102,6 +102,11 @@ For coordinated change sets that genuinely need more than 10 PRs, join the **#cl
   - For targeted shared-surface work, use `pnpm test:contracts:channels` or `pnpm test:contracts:plugins`
   - These commands also cover the shared seam/smoke files that the default unit lane skips
   - If you changed broader runtime behavior, still run the relevant wider lanes (`pnpm test:extensions`, `pnpm test:channels`, or `pnpm test`) before asking for review
+- If you touched bundled-plugin boundaries in shared code, run the matching inventories:
+  - `node scripts/check-src-extension-import-boundary.mjs --json` for `src/**`
+  - `node scripts/check-sdk-package-extension-import-boundary.mjs --json` for `src/plugin-sdk/**` and `packages/**`
+  - `node scripts/check-test-helper-extension-import-boundary.mjs --json` for `test/helpers/**`
+- Shared test helpers must use `src/test-utils/bundled-plugin-public-surface.ts` instead of repo-relative `extensions/**` imports. Keep plugin-local deep mocks inside the owning bundled plugin package.
 - If you have access to Codex, run `codex review --base origin/main` locally before opening or updating your PR. Treat this as the current highest standard of AI review, even if GitHub Codex review also runs.
 - Do not submit refactor-only PRs unless a maintainer explicitly requested that refactor for an active fix or deliverable.
 - Do not submit test or CI-config fixes for failures already red on `main` CI. If a failure is already visible in the [main branch CI runs](https://github.com/openclaw/openclaw/actions), it's a known issue the Maintainer team is tracking, and a PR that only addresses those failures will be closed automatically. If you spot a _new_ regression not yet shown in main CI, report it as an issue first.

INCIDENT_RESPONSE.md

@@ -0,0 +1,53 @@
+# OpenClaw Incident Response Plan
+
+## 1. Detection and triage
+
+We monitor security signals from:
+
+- GitHub Security Advisories (GHSA) and private vulnerability reports.
+- Public GitHub issues/discussions when reports are not sensitive.
+- Official plublic discussion groups and channels (i.e. Discord and X).
+- Automated signals (for example Dependabot, CodeQL, npm advisories, and secret scanning).
+
+Initial triage:
+
+1. Confirm affected component, version, and trust boundary impact.
+2. Classify as security issue vs hardening/no-action using the repository `SECURITY.md` scope and out-of-scope rules.
+3. An incident owner responds accordingly.
+
+## 2. Assessment
+
+Severity guide:
+
+- **Critical:** Package/release/repository compromise, active exploitation, or unauthenticated trust-boundary bypass with high-impact control or data exposure.
+- **High:** Verified trust-boundary bypass requiring limited preconditions (for example authenticated but unauthorized high-impact action), or exposure of OpenClaw-owned sensitive credentials.
+- **Medium:** Significant security weakness with practical impact but constrained exploitability or substantial prerequisites.
+- **Low:** Defense-in-depth findings, narrowly scoped denial-of-service, or hardening/parity gaps without a demonstrated trust-boundary bypass.
+
+## 3. Response
+
+1. Acknowledge receipt to the reporter (private when sensitive).
+2. Reproduce on supported releases and latest `main`, then implement and validate a patch with regression coverage.
+3. For critical/high incidents, prepare patched release(s) as fast as practical.
+4. For medium/low incidents, patch in normal release flow and document mitigation guidance.
+
+## 4. Communication
+
+We communicate through:
+
+- GitHub Security Advisories in the affected repository.
+- Release notes/changelog entries for fixed versions.
+- Direct reporter follow-up on status and resolution.
+
+Disclosure policy:
+
+- Critical/high incidents should receive coordinated disclosure, with CVE issuance when appropriate.
+- Low-risk hardening findings may be documented in release notes or advisories without CVE, depending on impact and user exposure.
+
+## 5. Recovery and follow-up
+
+After shipping the fix:
+
+1. Verify remediations in CI and release artifacts.
+2. Run a short post-incident review (timeline, root cause, detection gap, prevention plan).
+3. Add follow-up hardening/tests/docs tasks and track them to completion.

appcast.xml

@@ -2,6 +2,63 @@
 <rss xmlns:sparkle="http://www.andymatuschak.org/xml-namespaces/sparkle" version="2.0">
     <channel>
         <title>OpenClaw</title>
+        <item>
+            <title>2026.4.9</title>
+            <pubDate>Thu, 09 Apr 2026 02:38:08 +0000</pubDate>
+            <link>https://raw.githubusercontent.com/openclaw/openclaw/main/appcast.xml</link>
+            <sparkle:version>2026040990</sparkle:version>
+            <sparkle:shortVersionString>2026.4.9</sparkle:shortVersionString>
+            <sparkle:minimumSystemVersion>15.0</sparkle:minimumSystemVersion>
+            <description><![CDATA[<h2>OpenClaw 2026.4.9</h2>
+<h3>Changes</h3>
+<ul>
+<li>Memory/dreaming: add a grounded REM backfill lane with historical <code>rem-harness --path</code>, diary commit/reset flows, cleaner durable-fact extraction, and live short-term promotion integration so old daily notes can replay into Dreams and durable memory without a second memory stack. Thanks @mbelinky.</li>
+<li>Control UI/dreaming: add a structured diary view with timeline navigation, backfill/reset controls, traceable dreaming summaries, and a grounded Scene lane with promotion hints plus a safe clear-grounded action for staged backfill signals. (#63395) Thanks @mbelinky.</li>
+<li>QA/lab: add character-vibes evaluation reports with model selection and parallel runs so live QA can compare candidate behavior faster.</li>
+<li>Plugins/provider-auth: let provider manifests declare <code>providerAuthAliases</code> so provider variants can share env vars, auth profiles, config-backed auth, and API-key onboarding choices without core-specific wiring.</li>
+<li>iOS: pin release versioning to an explicit CalVer in <code>apps/ios/version.json</code>, keep TestFlight iteration on the same short version until maintainers intentionally promote the next gateway version, and add the documented <code>pnpm ios:version:pin -- --from-gateway</code> workflow for release trains. (#63001) Thanks @ngutman.</li>
+</ul>
+<h3>Fixes</h3>
+<ul>
+<li>Browser/security: re-run blocked-destination safety checks after interaction-driven main-frame navigations from click, evaluate, hook-triggered click, and batched action flows, so browser interactions cannot bypass the SSRF quarantine when they land on forbidden URLs. (#63226) Thanks @eleqtrizit.</li>
+<li>Security/dotenv: block runtime-control env vars plus browser-control override and skip-server env vars from untrusted workspace <code>.env</code> files, and reject unsafe URL-style browser control override specifiers before lazy loading. (#62660, #62663) Thanks @eleqtrizit.</li>
+<li>Gateway/node exec events: mark remote node <code>exec.started</code>, <code>exec.finished</code>, and <code>exec.denied</code> summaries as untrusted system events and sanitize node-provided command/output/reason text before enqueueing them, so remote node output cannot inject trusted <code>System:</code> content into later turns. (#62659) Thanks @eleqtrizit.</li>
+<li>Plugins/onboarding auth choices: prevent untrusted workspace plugins from colliding with bundled provider auth-choice ids during non-interactive onboarding, so bundled provider setup keeps operator secrets out of untrusted workspace plugin handlers unless those plugins are explicitly trusted. (#62368) Thanks @pgondhi987.</li>
+<li>Security/dependency audit: force <code>basic-ftp</code> to <code>5.2.1</code> for the CRLF command-injection fix and bump Hono plus <code>@hono/node-server</code> in production resolution paths.</li>
+<li>Android/pairing: clear stale setup-code auth on new QR scans, bootstrap operator and node sessions from fresh pairing, prefer stored device tokens after bootstrap handoff, and pause pairing auto-retry while the app is backgrounded so scan-once Android pairing recovers reliably again. (#63199) Thanks @obviyus.</li>
+<li>Matrix/gateway: wait for Matrix sync readiness before marking startup successful, keep Matrix background handler failures contained, and route fatal Matrix sync stops through channel-level restart handling instead of crashing the whole gateway. (#62779) Thanks @gumadeiras.</li>
+<li>Slack/media: preserve bearer auth across same-origin <code>files.slack.com</code> redirects while still stripping it on cross-origin Slack CDN hops, so <code>url_private_download</code> image attachments load again. (#62960) Thanks @vincentkoc.</li>
+<li>Reply/doctor: use the active runtime snapshot for queued reply runs, resolve reply-run SecretRefs before preflight helpers touch config, surface gateway OAuth reauth failures to users, and make <code>openclaw doctor</code> call out exact reauth commands. (#62693, #63217) Thanks @mbelinky.</li>
+<li>Control UI: guard stale session-history reloads during fast session switches so the selected session and rendered transcript stay in sync. (#62975) Thanks @scoootscooob.</li>
+<li>Gateway/chat: suppress exact and streamed <code>ANNOUNCE_SKIP</code> / <code>REPLY_SKIP</code> control replies across live chat updates and history sanitization so internal agent-to-agent control tokens no longer leak into user-facing gateway chat surfaces. (#51739) Thanks @Pinghuachiu.</li>
+<li>Auto-reply/NO_REPLY: strip glued leading <code>NO_REPLY</code> tokens before reply normalization and ACP-visible streaming so silent sentinel text no longer leaks into user-visible replies while preserving substantive <code>NO_REPLY ...</code> text. Thanks @frankekn.</li>
+<li>Sessions/routing: preserve established external routes on inter-session announce traffic so <code>sessions_send</code> follow-ups do not steal delivery from Telegram, Discord, or other external channels. (#58013) Thanks @duqaXxX.</li>
+<li>Gateway/sessions: clear auto-fallback-pinned model overrides on <code>/reset</code> and <code>/new</code> while still preserving explicit user model selections, including legacy sessions created before override-source tracking existed. (#63155) Thanks @frankekn.</li>
+<li>Slack/ACP: treat Slack ACP block replies as visible delivered output so OpenClaw stops re-sending the final fallback text after Slack already rendered the reply. (#62858) Thanks @gumadeiras.</li>
+<li>Slack/partial streaming: key turn-local dedupe by dispatch kind and keep the final fallback reply path active when preview finalization fails so stale preview text cannot suppress the actual final answer. (#62859) Thanks @gumadeiras.</li>
+<li>Matrix/doctor: migrate legacy <code>channels.matrix.dm.policy: "trusted"</code> configs back to compatible DM policies during <code>openclaw doctor --fix</code>, preserving explicit <code>allowFrom</code> boundaries as <code>allowlist</code> and defaulting empty legacy configs to <code>pairing</code>. (#62942) Thanks @lukeboyett.</li>
+<li>npm packaging: mirror bundled channel runtime deps, stage Nostr runtime deps, derive required root mirrors from manifests and built chunks, and test packed release tarballs without repo <code>node_modules</code> so fresh installs fail fast on missing plugin deps instead of crashing at runtime. (#63065) Thanks @scoootscooob.</li>
+<li>QA/live auth: fail fast when live QA scenarios hit classified auth or runtime failure replies, including raw scenario wait paths, and sanitize missing-key guidance so gateway auth problems surface as actionable errors instead of timeouts. (#63333) Thanks @shakkernerd.</li>
+<li>Providers/OpenAI: default missing reasoning effort to <code>high</code> on OpenAI Responses, WebSocket, and compatible completions transports, while still honoring explicit per-run reasoning levels.</li>
+<li>Providers/Ollama: allow Ollama models using the native <code>api: "ollama"</code> path to optionally display thinking output when <code>/think</code> is set to a non-off level. (#62712) Thanks @hoyyeva.</li>
+<li>Codex CLI: pass OpenClaw's system prompt through Codex's <code>model_instructions_file</code> config override so fresh Codex CLI sessions receive the same prompt guidance as Claude CLI sessions.</li>
+<li>Auth/profiles: persist explicit auth-profile upserts directly and skip external CLI sync for local writes so profile changes are saved without stale external credential state.</li>
+<li>Agents/timeouts: make the LLM idle timeout inherit <code>agents.defaults.timeoutSeconds</code> when configured, disable the unconfigured idle watchdog for cron runs, and point idle-timeout errors at <code>agents.defaults.llm.idleTimeoutSeconds</code>. Thanks @drvoss.</li>
+<li>Agents/failover: classify Z.ai vendor code <code>1311</code> as billing and <code>1113</code> as auth, including long wrapped <code>1311</code> payloads, so these errors stop falling through to generic failover handling. (#49552) Thanks @1bcMax.</li>
+<li>QQBot/media-tags: support HTML entity-encoded angle brackets (<code>&lt;</code>/<code>&gt;</code>), URL slashes in attributes, and self-closing media tags so upstream <code><qqimg></code> payloads are correctly parsed and normalized. (#60493) Thanks @ylc0919.</li>
+<li>Memory/dreaming: harden grounded backfill inputs, diary writes, status payloads, and diary action classification by preserving source-day labels, rejecting missing or symlinked targets cleanly, normalizing diary headings in gateway backfills, and tightening claim splitting plus diary source metadata. Thanks @mbelinky.</li>
+<li>Memory/dreaming: accept embedded heartbeat trigger tokens so light and REM dreaming still run when runtime wrappers include extra heartbeat text.</li>
+<li>Android/manual connect: allow blank port input only for TLS manual gateway endpoints so standard HTTPS Tailscale hosts default to <code>443</code> without silently changing cleartext manual connects. (#63134) Thanks @Tyler-RNG.</li>
+<li>Windows/update: add heap headroom to Windows <code>pnpm build</code> steps during dev updates so update preflight builds stop failing on low default Node memory.</li>
+<li>Plugin SDK: export the channel plugin base and web-search config contract through the public package so plugins can use them without private imports.</li>
+<li>Plugins/contracts: keep test-only helpers out of production contract barrels, load shared contract harnesses through bundled test surfaces, and harden guardrails so indirect re-exports and canonical <code>*.test.ts</code> files stay blocked. (#63311) Thanks @altaywtf.</li>
+<li>Control UI/models: preserve provider-qualified refs for OpenRouter catalog models whose ids already contain slashes so picker selections submit allowlist-compatible model refs instead of dropping the <code>openrouter/</code> prefix. (#63416) Thanks @sallyom.</li>
+<li>Plugin SDK/command auth: split command status builders onto the lightweight <code>openclaw/plugin-sdk/command-status</code> subpath while preserving deprecated <code>command-auth</code> compatibility exports, so auth-only plugin imports no longer pull status/context warmup into CLI onboarding paths. (#63174) Thanks @hxy91819.</li>
+</ul>
+<p><a href="https://github.com/openclaw/openclaw/blob/main/CHANGELOG.md">View full changelog</a></p>
+]]></description>
+            <enclosure url="https://github.com/openclaw/openclaw/releases/download/v2026.4.9/OpenClaw-2026.4.9.zip" length="25336730" type="application/octet-stream" sparkle:edSignature="zFKTcKpejPyGEHj6Bdop3EBDfRrHyQMtJzrpVKsIkBq3I/jbTNvsxQveKEy9r7dqkZVsldFYv7eSunP3SUmaAw=="/>
+        </item>
         <item>
             <title>2026.4.8</title>
             <pubDate>Wed, 08 Apr 2026 06:12:50 +0000</pubDate>
@@ -132,253 +189,5 @@
 ]]></description>
             <enclosure url="https://github.com/openclaw/openclaw/releases/download/v2026.4.7/OpenClaw-2026.4.7.zip" length="25324827" type="application/octet-stream" sparkle:edSignature="RyFWRz1trE/qvOiInD4vR6je9wx7fUTtHpZ94W8rMlZDByux9CyXOm/Anai96b9KyjTeQyC7YnJp5SRnYY3iCg=="/>
         </item>
-        <item>
-            <title>2026.4.5</title>
-            <pubDate>Mon, 06 Apr 2026 04:55:17 +0100</pubDate>
-            <link>https://raw.githubusercontent.com/openclaw/openclaw/main/appcast.xml</link>
-            <sparkle:version>2026040590</sparkle:version>
-            <sparkle:shortVersionString>2026.4.5</sparkle:shortVersionString>
-            <sparkle:minimumSystemVersion>15.0</sparkle:minimumSystemVersion>
-            <description><![CDATA[<h2>OpenClaw 2026.4.5</h2>
-<h3>Breaking</h3>
-<ul>
-<li>Config: remove legacy public config aliases such as <code>talk.voiceId</code> / <code>talk.apiKey</code>, <code>agents.*.sandbox.perSession</code>, <code>browser.ssrfPolicy.allowPrivateNetwork</code>, <code>hooks.internal.handlers</code>, and channel/group/room <code>allow</code> toggles in favor of the canonical public paths and <code>enabled</code>, while keeping load-time compatibility and <code>openclaw doctor --fix</code> migration support for existing configs. (#60726) Thanks @vincentkoc.</li>
-</ul>
-<h3>Changes</h3>
-<ul>
-<li>Agents/video generation: add the built-in <code>video_generate</code> tool so agents can create videos through configured providers and return the generated media directly in the reply.</li>
-<li>Agents/music generation: ignore unsupported optional hints such as <code>durationSeconds</code> with a warning instead of hard-failing requests on providers like Google Lyria.</li>
-<li>Providers/ComfyUI: add a bundled <code>comfy</code> workflow media plugin for local ComfyUI and Comfy Cloud workflows, including shared <code>image_generate</code>, <code>video_generate</code>, and workflow-backed <code>music_generate</code> support, with prompt injection, optional reference-image upload, live tests, and output download.</li>
-<li>Tools/music generation: add the built-in <code>music_generate</code> tool with bundled Google (Lyria) and MiniMax providers plus workflow-backed Comfy support, including async task tracking and follow-up delivery of finished audio.</li>
-<li>Providers: add bundled Qwen, Fireworks AI, and StepFun providers, plus MiniMax TTS, Ollama Web Search, and MiniMax Search integrations for chat, speech, and search workflows. (#60032, #55921, #59318, #54648)</li>
-<li>Providers/Amazon Bedrock: add bundled Mantle support plus inference-profile discovery and automatic request-region injection so Bedrock-hosted Claude, GPT-OSS, Qwen, Kimi, GLM, and similar routes work with less manual setup. (#61296, #61299) Thanks @wirjo.</li>
-<li>Control UI/multilingual: add localized control UI support for Simplified Chinese, Traditional Chinese, Brazilian Portuguese, German, Spanish, Japanese, Korean, French, Turkish, Indonesian, Polish, and Ukrainian. Thanks @vincentkoc.</li>
-<li>Plugins: add plugin-config TUI prompts to guided onboarding/setup flows, and add <code>openclaw plugins install --force</code> so existing plugin and hook-pack targets can be replaced without using the dangerous-code override flag. (#60590, #60544)</li>
-<li>Control UI/skills: add ClawHub search, detail, and install flows directly in the Skills panel. (#60134) Thanks @samzong.</li>
-<li>iOS/exec approvals: add generic APNs approval notifications that open an in-app exec approval modal, fetch command details only after authenticated operator reconnect, and clear stale notification state when the approval resolves. (#60239) Thanks @ngutman.</li>
-<li>Matrix/exec approvals: add Matrix-native exec approval prompts with account-scoped approvers, channel-or-DM delivery, and room-thread aware resolution handling. (#58635) Thanks @gumadeiras.</li>
-<li>Channels/context visibility: add configurable <code>contextVisibility</code> per channel (<code>all</code>, <code>allowlist</code>, <code>allowlist_quote</code>) so supplemental quote, thread, and fetched history context can be filtered by sender allowlists instead of always passing through as received.</li>
-<li>Providers/request overrides: add shared model and media request transport overrides across OpenAI-, Anthropic-, Google-, and compatible provider paths, including headers, auth, proxy, and TLS controls. (#60200)</li>
-<li>Providers/OpenAI: add forward-compat <code>openai-codex/gpt-5.4-mini</code>, an opt-in GPT personality, and provider-owned GPT-5 prompt contributions so Codex/GPT runs stay cache-stable and compatible with bundled catalog lag.</li>
-<li>Agents/Claude CLI: expose OpenClaw tools to background Claude CLI runs through a loopback MCP bridge and switch bundled runs to stdin + <code>stream-json</code> partial-message streaming so prompts stop riding argv, long replies show live progress, and final session/usage metadata still land cleanly. (#35676) Thanks @mylukin.</li>
-<li>ACPX/runtime: embed the ACP runtime directly in the bundled <code>acpx</code> plugin, remove the extra external ACP CLI hop, harden live ACP session binding and reuse, and add a generic <code>reply_dispatch</code> hook so bundled plugins like ACPX can own reply interception without hardcoded ACP paths in core auto-reply routing. (#61319)</li>
-<li>Agents/progress: add experimental structured plan updates and structured execution item events so compatible UIs can show clearer step-by-step progress during long-running runs.</li>
-<li>Providers/Anthropic: remove the Claude CLI backend and setup-token from new onboarding, keep existing configured legacy profiles runnable, and have <code>openclaw doctor</code> repair or remove stale <code>anthropic:claude-cli</code> state during migration.</li>
-<li>Tools/video generation: add bundled xAI (<code>grok-imagine-video</code>), Alibaba Model Studio Wan, and Runway video providers, plus live-test/default model wiring for all three.</li>
-<li>Memory/search: add Amazon Bedrock embeddings for Titan, Cohere, Nova, and TwelveLabs models, with AWS credential-chain auto-detection for <code>provider: "auto"</code> and provider-specific dimension controls. Thanks @wirjo.</li>
-<li>Providers/Amazon Bedrock Mantle: generate bearer tokens from the AWS credential chain so Mantle auto-discovery can use IAM auth without manually exporting <code>AWS_BEARER_TOKEN_BEDROCK</code>. Thanks @wirjo.</li>
-<li>Memory/dreaming (experimental): add weighted short-term recall promotion, a <code>/dreaming</code> command, Dreams UI, multilingual conceptual tagging, and doctor/status repair support, while refactoring dreaming from competing modes into three cooperative phases (light, deep, REM) with independent schedules and recovery behavior so durable memory promotion can run in the background with less manual setup. (#60569, #60697) Thanks @vignesh07.</li>
-<li>Memory/dreaming: add configurable aging controls (<code>recencyHalfLifeDays</code>, <code>maxAgeDays</code>) plus optional verbose logging so operators can tune recall decay and inspect promotion decisions more easily.</li>
-<li>Memory/dreaming: add REM preview tooling (<code>openclaw memory rem-harness</code>, <code>promote-explain</code>), surface possible lasting truths during REM staging, and make deep promotion replay-safe so reruns reconcile instead of duplicating <code>MEMORY.md</code> entries.</li>
-<li>Memory/dreaming: write dreaming trail content to top-level <code>dreams.md</code> instead of daily memory notes, update <code>/dreaming</code> help text to point there, and keep <code>dreams.md</code> available for explicit reads without pulling it into default recall. Thanks @davemorin.</li>
-<li>Memory/dreaming: add the Dream Diary surface in Dreams, simplify user-facing dreaming config to <code>enabled</code> plus optional <code>frequency</code>, treat phases as implementation detail in docs/UI, and keep the lobster animation visible above diary content. Thanks @vignesh07.</li>
-<li>Prompt caching: keep prompt prefixes more reusable across transport fallback, deterministic MCP tool ordering, compaction, embedded image history, normalized system-prompt fingerprints, <code>openclaw status --verbose</code> cache diagnostics, and the removal of duplicate in-band tool inventories from agent system prompts so follow-up turns hit cache more reliably. (#58036, #58037, #58038, #59054, #60603, #60691) Thanks @bcherny and @vincentkoc.</li>
-<li>Agents/cache: diagnostics: add prompt-cache break diagnostics, trace live cache scenarios through embedded runner paths, and show cache reuse explicitly in <code>openclaw status --verbose</code>. Thanks @vincentkoc.</li>
-<li>Agents/cache: stabilize cache-relevant system prompt fingerprints by normalizing equivalent structured prompt whitespace, line endings, hook-added system context, and runtime capability ordering so semantically unchanged prompts reuse KV/cache more reliably. Thanks @vincentkoc.</li>
-<li>Agents/tool prompts: remove the duplicate in-band tool inventory from agent system prompts so tool-calling models rely on the structured tool definitions as the single source of truth, improving prompt stability and reducing stale tool guidance.</li>
-<li>Config/schema: enrich the exported <code>openclaw config schema</code> JSON Schema with field titles and descriptions so editors, agents, and other schema consumers receive the same config help metadata. (#60067) Thanks @solavrc.</li>
-<li>Providers/CLI: remove bundled CLI text-provider backends and the <code>agents.defaults.cliBackends</code> surface, while keeping ACP harness sessions and Gemini media understanding on the native bundled providers.</li>
-<li>Matrix/exec approvals: clarify unavailable-approval replies so Matrix no longer claims chat approvals are unsupported when native exec approvals are merely unconfigured. (#61424) Thanks @gumadeiras.</li>
-<li>Docs/IRC: replace public IRC hostname examples with <code>irc.example.com</code> and recommend private servers for bot coordination while listing common public networks for intentional use.</li>
-<li>Memory/dreaming: group nearby daily-note lines into short coherent chunks before staging them for dreaming, so one-off context from recent notes reaches REM/deep with better evidence and less line-level noise.</li>
-<li>Memory/dreaming: drop generic date/day headings from daily-note chunk prefixes while keeping meaningful section labels, so staged snippets stay cleaner and more reusable. (#61597) Thanks @mbelinky.</li>
-<li>Plugins/Lobster: run bundled Lobster workflows in process instead of spawning the external CLI, reducing transport overhead and unblocking native runtime integration. (#61523) Thanks @mbelinky.</li>
-<li>Plugins/Lobster: harden managed resume validation so invalid TaskFlow resume calls fail earlier, and memoize embedded runtime loading per runner while keeping failed loads retryable. (#61566) Thanks @mbelinky.</li>
-</ul>
-<h3>Fixes</h3>
-<ul>
-<li>Security: preserve restrictive plugin-only tool allowlists, require owner access for <code>/allowlist add</code> and <code>/allowlist remove</code>, fail closed when <code>before_tool_call</code> hooks crash, block browser SSRF redirect bypasses earlier, and keep non-interactive auth-choice inference scoped to bundled and already-trusted plugins. (#58476, #59836, #59822, #58771, #59120) Thanks @eleqtrizit and @pgondhi987.</li>
-<li>Providers/OpenAI: make GPT-5 and Codex runs act sooner with lower-verbosity defaults, visible progress during tool work, and a one-shot retry when a turn only narrates the plan instead of taking action.</li>
-<li>Providers/OpenAI and reply delivery: preserve native <code>reasoning.effort: "none"</code> and strict schemas where supported, add GPT-5.4 assistant <code>phase</code> metadata across replay and the Gateway <code>/v1/responses</code> layer, and keep commentary buffered until <code>final_answer</code> so web chat, session previews, embedded replies, and Telegram partials stop leaking planning text. Fixes #59150, #59643, #61282.</li>
-<li>Telegram: fix current-model checks in the model picker, HTML-format non-default <code>/model</code> confirmations, explicit topic replies, persisted reaction ownership across restarts, caption-media placeholder and <code>file_id</code> preservation on download failure, and upgraded-install inbound image reads. (#60384, #60042, #59634, #59207, #59948, #59971) Thanks @sfuminya, @GitZhangChi, @dashhuang, @samzong, @v1p0r, and @neeravmakwana.</li>
-<li>Telegram: restore DM voice-note preflight transcription so direct-message audio stops arriving as raw <code><media:audio></code> placeholders. (#61008) Thanks @manueltarouca.</li>
-<li>Telegram/reasoning: only create a Telegram reasoning preview lane when the session is explicitly <code>reasoning:stream</code>, so hidden <code><think></code> traces from streamed replies stop surfacing as chat previews on normal sessions. Thanks @vincentkoc.</li>
-<li>Telegram/native command menu: trim long menu descriptions before dropping commands so sub-100 command sets can still fit Telegram's payload budget and keep more <code>/</code> entries visible. (#61129) Thanks @neeravmakwana.</li>
-<li>Discord: keep REST, webhook, and monitor traffic on the configured proxy, preserve component-only media sends, honor <code>@everyone</code> and <code>@here</code> mention gates, keep ACK reactions on the active account, and split voice connect/playback timeouts so auto-join is more reliable. (#57465, #60361, #60345) Thanks @geekhuashan.</li>
-<li>Discord/reply tags: strip leaked <code>[[reply_to_current]]</code> control tags from preview text and honor explicit reply-tag threading during final delivery, so Discord replies stay attached to the triggering message instead of printing reply metadata into chat.</li>
-<li>Discord/replies: replace the unshipped <code>replyToOnlyWhenBatched</code> flag with <code>replyToMode: "batched"</code> so native reply references only attach on debounced multi-message turns while explicit reply tags still work.</li>
-<li>Discord/image generation: include the real generated <code>MEDIA:</code> paths in tool output, avoid duplicate plain-output media requeueing, and persist volatile workspace-generated media into durable outbound media before final reply delivery so generated image replies stop pointing at missing local files.</li>
-<li>Slack: route live DM replies back to the concrete inbound DM channel while keeping persisted routing metadata user-scoped, so normal assistant replies stop disappearing when pairing and system messages still arrive. (#59030) Thanks @afurm.</li>
-<li>WhatsApp: restore <code>channels.whatsapp.blockStreaming</code> and reset watchdog timeouts after reconnect so quiet chats stop falling into reconnect loops. (#60007, #60069) Thanks @MonkeyLeeT and @mcaxtr.</li>
-<li>Android/Talk Mode: cancel in-flight <code>talk.speak</code> playback when speech is explicitly stopped, and restore spoken replies on both node-scoped and gateway-backed sessions by keeping reply routing and embedded transport overrides aligned with the current playback path. (#60306, #61164, #61214)</li>
-<li>Voice-call/OpenAI: pass full plugin config into realtime transcription provider resolution so streaming calls can discover the bundled OpenAI realtime transcription provider again. Fixes #60936. Thanks @sliekens and @vincentkoc.</li>
-<li>Matrix/exec approvals: anchor seeded approval reactions to the primary Matrix prompt event, resolve them from event metadata instead of prompt text, and clean up chunked approval prompts correctly. (#60931) Thanks @gumadeiras.</li>
-<li>Matrix: recover more reliably when secret storage or recovery keys are missing by recreating secret storage during repair and backup reset, hold crypto snapshot locks during persistence, and surface explicit too-large attachment markers. (#59846, #59851, #60599, #60289) Thanks @al3mart, @emonty, and @efe-arv.</li>
-<li>Matrix/DM sessions: add <code>channels.matrix.dm.sessionScope</code>, shared-session collision notices, and aligned outbound session reuse so separate Matrix DM rooms can keep distinct context when configured. (#61373) Thanks @gumadeiras.</li>
-<li>Matrix: move legacy top-level <code>avatarUrl</code> into the default account during multi-account promotion and keep env-backed account setup avatar config persisted. (#61437) Thanks @gumadeiras.</li>
-<li>MS Teams: download inline DM images via Graph API and preserve channel reply threading in proactive fallback. (#52212, #55198) Thanks @Ted-developer and @hyojin.</li>
-<li>MS Teams: replace the deprecated Teams SDK HttpPlugin stub with <code>httpServerAdapter</code> so recurring gateway deprecation warnings stop firing and the Express 5 compatibility workaround stays on the supported SDK path. (#60939) Thanks @coolramukaka-sys.</li>
-<li>Control UI/chat: add a per-session thinking-level picker in the chat header and mobile chat settings, and keep the browser bundle on UI-local thinking/session-key helpers so Safari no longer crashes on Node-only imports before rendering chat controls.</li>
-<li>Sandbox/SSH: reject hardlinked files during cross-device rename fallback so EXDEV file copies preserve the same pinned file-boundary checks as direct reads.</li>
-<li>Control UI: keep Stop visible during tool-only execution, preserve pending-send busy state, and clear stale ClawHub search results as soon as the query changes. (#54528, #59800, #60267) Thanks @chziyue and @frankekn.</li>
-<li>Control UI/avatar: honor <code>ui.assistant.avatar</code> when serving <code>/avatar/:agentId</code> so Appearance UI avatar paths stop falling back to initials placeholders. (#60778) Thanks @hannasdev.</li>
-<li>Control UI/cron: highlight the Cron refresh button while refresh is in flight so the page's loading state stays visible even when prior data remains on screen. (#60394) Thanks @coder-zhuzm.</li>
-<li>Control UI/Overview: prevent gateway access token/password visibility toggle buttons from overlapping their inputs at narrow widths. (#56924) Thanks @bbddbb1.</li>
-<li>Auto-reply: unify reply lifecycle ownership across preflight compaction, session rotation, CLI-backed runs, and gateway restart handling so <code>/stop</code> and same-session overlap checks target the right active turn and restart-interrupted turns return the restart notice instead of being silently dropped. (#61267) Thanks @dutifulbob.</li>
-<li>Reply delivery: prevent duplicate block replies on <code>text_end</code> channels so providers that emit explicit text-end boundaries no longer double-send the same final message. (#61530)</li>
-<li>Gateway/startup: default <code>gateway.mode</code> to <code>local</code> when unset, detect PID recycling in gateway lock files on Windows and macOS, and show startup progress so healthy restarts stop getting blocked by stale locks. (#54801, #60085, #59843) Thanks @BradGroux and @TonyDerek-dot.</li>
-<li>Gateway/macOS: let launchd <code>KeepAlive</code> own in-process gateway restarts again, adding a short supervised-exit delay so rapid restarts avoid launchd crash-loop unloads while <code>openclaw gateway restart</code> still reports real LaunchAgent errors synchronously.</li>
-<li>Gateway/macOS: re-bootstrap the LaunchAgent if <code>launchctl kickstart -k</code> unloads it during restart so failed restarts do not leave the gateway unmanaged until manual repair.</li>
-<li>Gateway/macOS: recover installed-but-unloaded LaunchAgents during <code>openclaw gateway start</code> and <code>restart</code>, while still preferring live unmanaged gateways during restart recovery. (#43766) Thanks @HenryC-3.</li>
-<li>Gateway/Windows scheduled tasks: preserve Task Scheduler settings on reinstall, fail loudly when <code>/Run</code> does not start, and report fast failed restarts accurately instead of pretending they timed out after 60 seconds. (#59335) Thanks @tmimmanuel.</li>
-<li>Windows/restart: fall back to the installed Startup-entry launcher when the scheduled task was never registered, so <code>/restart</code> can relaunch the gateway on Windows setups where <code>schtasks</code> install fell back during onboarding. (#58943) Thanks @imechZhangLY.</li>
-<li>Windows/restart: clean up stale gateway listeners before Windows self-restart and treat listener and argv probe failures as inconclusive, so scheduled-task relaunch no longer falls into an <code>EADDRINUSE</code> retry loop. (#60480) Thanks @arifahmedjoy.</li>
-<li>Update/npm: prefer the npm binary that owns the installed global OpenClaw prefix so mixed Homebrew-plus-nvm setups update the right install. (#60153) Thanks @jayeshp19.</li>
-<li>Agents/music and video generation: add <code>tools.media.asyncCompletion.directSend</code> as an opt-in direct-delivery path for finished async media tasks, while keeping the legacy requester-session wake/model-delivery flow as the default.</li>
-<li>CLI/skills JSON: route <code>skills list --json</code>, <code>skills info --json</code>, and <code>skills check --json</code> output to stdout instead of stderr so machine-readable consumers receive JSON on the expected stream again. (#60914; fixes #57599; landed from contributor PR #57611 by @Aftabbs) Thanks @Aftabbs.</li>
-<li>CLI/Commander: preserve Commander-computed exit codes for argument and help-error paths, and cover the user-argv parse mode in the regression tests so invalid CLI invocations no longer report success when exits are intercepted. (#60923) Thanks @Linux2010.</li>
-<li>Cron: replay interrupted recurring jobs on the first gateway restart instead of waiting for a second restart. (#60583) Thanks @joelnishanth.</li>
-<li>Cron: send failure notifications through the job's primary delivery channel using the same session context as successful delivery when no explicit <code>failureDestination</code> is configured. (#60622) Thanks @artwalker.</li>
-<li>Exec/remote skills: stop advertising <code>exec host=node</code> when the current exec policy cannot route to a node, and clarify blocked exec-host override errors with both the requested host and allowed config path.</li>
-<li>Agents/Claude CLI/security: clear inherited Claude Code config-root and plugin-root env overrides like <code>CLAUDE_CONFIG_DIR</code> and <code>CLAUDE_CODE_PLUGIN_*</code>, so OpenClaw-launched Claude CLI runs cannot be silently pointed at an alternate Claude config/plugin tree with different hooks, plugins, or auth context. Thanks @vincentkoc.</li>
-<li>Agents/Claude CLI/security: clear inherited Claude Code provider-routing and managed-auth env overrides, and mark OpenClaw-launched Claude CLI runs as host-managed, so Claude CLI backdoor sessions cannot be silently redirected to proxy, Bedrock, Vertex, Foundry, or parent-managed token contexts. Thanks @vincentkoc.</li>
-<li>Agents/Claude CLI/security: force host-managed Claude CLI backdoor runs to <code>--setting-sources user</code>, even under custom backend arg overrides, so repo-local <code>.claude</code> project/local settings, hooks, and plugin discovery do not silently execute inside non-interactive OpenClaw sessions. Thanks @vincentkoc.</li>
-<li>Agents/Claude CLI: treat malformed bare <code>--permission-mode</code> backend overrides as missing and fail safe back to <code>bypassPermissions</code>, so custom <code>cliBackends.claude-cli.args</code> security config cannot accidentally consume the next flag as a bogus permission mode. Thanks @vincentkoc.</li>
-<li>Gateway/device pairing: require non-admin paired-device sessions to manage only their own device for token rotate/revoke and paired-device removal, blocking cross-device token theft inside pairing-scoped sessions. (#50627) Thanks @coygeek.</li>
-<li>Gateway/plugin routes: keep gateway-auth plugin runtime routes on write-only fallback scopes unless a trusted-proxy caller explicitly declares narrower <code>x-openclaw-scopes</code>, so plugin HTTP handlers no longer mint admin-level runtime scopes on missing or untrusted HTTP scope headers. (#59815) Thanks @pgondhi987.</li>
-<li>Build/types: fix the Node <code>createRequire(...)</code> helper typing so provider-runtime lazy loads compile cleanly again and <code>pnpm build</code> no longer fails in the Pi embedded provider error-pattern path.</li>
-<li>Gateway/security: scope loopback browser-origin auth throttling by normalized origin so one localhost Control UI tab cannot lock out a different localhost browser origin after repeated auth failures.</li>
-<li>Gateway/auth: serialize async shared-secret auth attempts per client so concurrent Tailscale-capable failures cannot overrun the intended auth rate-limit budget. Thanks @Telecaster2147.</li>
-<li>Device pairing/security: keep non-operator device scope checks bound to the requested role prefix so bootstrap verification cannot redeem <code>operator.*</code> scopes through <code>node</code> auth. (#57258) Thanks @jlapenna.</li>
-<li>Device pairing: reject rotating device tokens into roles that were never approved during pairing, and keep reconnect role checks bounded to the paired device's approved role set. (#60462) Thanks @eleqtrizit.</li>
-<li>Gateway/device auth: reuse cached device-token scopes only for cached-token reconnects, while keeping explicit <code>deviceToken</code> scope requests and empty-cache fallbacks intact so reconnects preserve <code>operator.read</code> without breaking explicit auth flows. (#46032) Thanks @caicongyang.</li>
-<li>Mobile pairing/security: fail closed for internal <code>/pair</code> setup-code issuance, cleanup, and approval paths when gateway pairing scopes are missing, and keep approval-time requested-scope enforcement on the internal command path. (#55996) Thanks @coygeek.</li>
-<li>Mobile pairing/bootstrap: keep QR bootstrap handoff tokens bounded to the mobile-safe contract so node handoff stays unscoped and operator handoff drops mixed <code>node.*</code>, <code>operator.admin</code>, and <code>operator.pairing</code> scopes.</li>
-<li>Mobile pairing/Android: tighten secure endpoint handling so Tailscale and public remote setup reject cleartext endpoints, private LAN pairing still works, merged-role approvals mint both node and operator device tokens, and bootstrap tokens survive node auto-pair until operator approval finishes. (#60128, #60208, #60221) Thanks @obviyus.</li>
-<li>Android/canvas security: require exact normalized A2UI URL matches before forwarding canvas bridge actions, rejecting query mismatches and descendant paths while still allowing fragment-only A2UI navigation.</li>
-<li>Synology Chat/security: default low-level HTTPS helper TLS verification to on so helper/API defaults match the shipped safe account default, and only explicit <code>allowInsecureSsl: true</code> opts out.</li>
-<li>Synology Chat/security: route webhook token comparison through the shared constant-time secret helper for consistency with other bundled plugins.</li>
-<li>Plugins/marketplace: block remote marketplace symlink escapes without breaking ordinary local marketplace install paths. (#60556) Thanks @eleqtrizit.</li>
-<li>Telegram/local Bot API: honor <code>channels.telegram.apiRoot</code> for buffered media downloads, add <code>channels.telegram.network.dangerouslyAllowPrivateNetwork</code> for trusted fake-IP setups, and require <code>channels.telegram.trustedLocalFileRoots</code> before reading absolute Bot API <code>file_path</code> values. (#59544, #60705) Thanks @SARAMALI15792 and @obviyus.</li>
-<li>Outbound/sanitizer: strip leaked <code><tool_call></code>, <code><function_calls></code>, and model special tokens from shared user-visible assistant text, including truncated tool-call streams, so internal scaffolding no longer bleeds into replies across surfaces. (#60619) Thanks @oliviareid-svg.</li>
-<li>Agents/errors: surface an explicit disk-full message when local session or transcript writes fail with <code>ENOSPC</code>/<code>disk full</code>, so those runs stop degrading into opaque <code>NO_REPLY</code>-style failures. Thanks @vincentkoc.</li>
-<li>Exec approvals: remove heuristic command-obfuscation gating from host exec so gateway and node runs rely on explicit policy, allowlist, and strict inline-eval rules only.</li>
-<li>Agents/tool results: cap live tool-result persistence and overflow-recovery truncation at 40k characters so oversized tool output stays bounded without discarding recent context entirely.</li>
-<li>Discord/video replies: split text-plus-video deliveries into a text reply followed by a media-only send, and let live provider auth checks honor manifest-declared API key env vars like <code>MODELSTUDIO_API_KEY</code>.</li>
-<li>Config/All Settings: keep the raw config view intact when sensitive fields are blank instead of corrupting or dropping the rendered snapshot. (#28214) Thanks @solodmd.</li>
-<li>Plugin SDK/facades: back-fill bundled plugin facade sentinels before plugin-id tracking re-enters config loading, so CLI/provider startup no longer crashes with <code>shouldNormalizeGoogleProviderConfig is not a function</code> or other empty-facade reads during bundled plugin re-entry. Thanks @adam91holt.</li>
-<li>Plugins/facades: back-fill facade sentinels before tracked-plugin resolution re-enters config loading, so facade exports stay defined during circular provider normalization. (#61180) Thanks @adam91holt.</li>
-<li>QA lab: restore typed mock OpenAI gateway config wiring so QA-lab config helpers compile cleanly again and <code>pnpm check</code> / <code>pnpm build</code> stay green.</li>
-<li>Discord/image generation: include the real generated <code>MEDIA:</code> paths in tool output and avoid duplicate plain-output media requeueing so Discord image replies stop pointing at missing local files.</li>
-<li>Slack: route live DM replies back to the concrete inbound DM channel while keeping persisted routing metadata user-scoped, so normal assistant replies stop disappearing when pairing and system messages still arrive. (#59030) Thanks @afurm.</li>
-<li>Discord/reply tags: strip leaked <code>[[reply_to_current]]</code> control tags from preview text and honor explicit reply-tag threading during final delivery, so Discord replies stay attached to the triggering message instead of printing reply metadata into chat.</li>
-<li>Telegram: fix current-model checks in the model picker, HTML-format non-default <code>/model</code> confirmations, explicit topic replies, persisted reaction ownership across restarts, caption-media placeholder and <code>file_id</code> preservation on download failure, and upgraded-install inbound image reads. (#60384, #60042, #59634, #59207, #59948, #59971) Thanks @sfuminya, @GitZhangChi, @dashhuang, @samzong, @v1p0r, and @neeravmakwana.</li>
-<li>Telegram: restore DM voice-note preflight transcription so direct-message audio stops arriving as raw <code><media:audio></code> placeholders. (#61008) Thanks @manueltarouca.</li>
-<li>Telegram/reasoning: only create a Telegram reasoning preview lane when the session is explicitly <code>reasoning:stream</code>, so hidden <code><think></code> traces from streamed replies stop surfacing as chat previews on normal sessions. Thanks @vincentkoc.</li>
-<li>Telegram/native command menu: trim long menu descriptions before dropping commands so sub-100 command sets can still fit Telegram's payload budget and keep more <code>/</code> entries visible. (#61129) Thanks @neeravmakwana.</li>
-<li>Feishu/reasoning: only expose streamed reasoning previews when the session is explicitly <code>reasoning:stream</code>, so hidden reasoning traces do not surface on normal streaming sessions. Thanks @vincentkoc.</li>
-<li>Discord: keep REST, webhook, and monitor traffic on the configured proxy, preserve component-only media sends, honor <code>@everyone</code> and <code>@here</code> mention gates, keep ACK reactions on the active account, and split voice connect/playback timeouts so auto-join is more reliable. (#57465, #60361, #60345) Thanks @geekhuashan.</li>
-<li>WhatsApp: restore <code>channels.whatsapp.blockStreaming</code> and reset watchdog timeouts after reconnect so quiet chats stop falling into reconnect loops. (#60007, #60069) Thanks @MonkeyLeeT and @mcaxtr.</li>
-<li>Memory: keep <code>memory-core</code> builtin embedding registration on the already-registered path so selecting <code>memory-core</code> no longer recurses through plugin discovery and crashes during startup. (#61402) Thanks @ngutman.</li>
-<li>Agents/tool results: keep large <code>read</code> outputs visible longer, preserve the latest <code>read</code> output when older tool output can absorb the overflow budget, and fall back to Pi's normal overflow compaction/retry path before replacing a fresh <code>read</code> with a compacted stub. Thanks @vincentkoc.</li>
-<li>Memory/QMD: prefer modern <code>qmd collection add --glob</code>, accept newer single-line JSON hit metadata while keeping legacy line fields, refresh QMD docs/doctor install guidance and model-override guidance, and keep older QMD releases working. Thanks @vincentkoc.</li>
-<li>MS Teams: download inline DM images via Graph API and preserve channel reply threading in proactive fallback. (#52212, #55198) Thanks @Ted-developer and @hyojin.</li>
-<li>MS Teams: replace the deprecated Teams SDK HttpPlugin stub with <code>httpServerAdapter</code> so recurring gateway deprecation warnings stop firing and the Express 5 compatibility workaround stays on the supported SDK path. (#60939) Thanks @coolramukaka-sys.</li>
-<li>Matrix/exec approvals: anchor seeded approval reactions to the primary Matrix prompt event, resolve them from event metadata instead of prompt text, and clean up chunked approval prompts correctly. (#60931) Thanks @gumadeiras.</li>
-<li>Matrix: recover more reliably when secret storage or recovery keys are missing by recreating secret storage during repair and backup reset, hold crypto snapshot locks during persistence, and surface explicit too-large attachment markers. (#59846, #59851, #60599, #60289) Thanks @al3mart, @emonty, and @efe-arv.</li>
-<li>Android/Talk Mode: cancel in-flight <code>talk.speak</code> playback when speech is explicitly stopped, so stale replies stop starting after barge-in or manual stop. (#61164) Thanks @obviyus.</li>
-<li>Android/Talk Mode: restore spoken assistant replies on node-scoped sessions by keeping reply routing synced to the resolved node session key and pausing mic capture during reply playback. (#60306) Thanks @MKV21.</li>
-<li>Android/Talk Mode: restore voice replies on gateway-backed talk mode sessions by updating embedded runner transport overrides to the current agent transport API. (#61214) Thanks @obviyus.</li>
-<li>Voice-call/OpenAI: pass full plugin config into realtime transcription provider resolution so streaming calls can discover the bundled OpenAI realtime transcription provider again. Fixes #60936. Thanks @sliekens and @vincentkoc.</li>
-<li>Control UI/chat: add a per-session thinking-level picker in the chat header and mobile chat settings, and keep the browser bundle on UI-local thinking/session-key helpers so Safari no longer crashes on Node-only imports before rendering chat controls.</li>
-<li>Control UI: keep Stop visible during tool-only execution, preserve pending-send busy state, and clear stale ClawHub search results as soon as the query changes. (#54528, #59800, #60267) Thanks @chziyue and @frankekn.</li>
-<li>Control UI/avatar: honor <code>ui.assistant.avatar</code> when serving <code>/avatar/:agentId</code> so Appearance UI avatar paths stop falling back to initials placeholders. (#60778) Thanks @hannasdev.</li>
-<li>Control UI/cron: highlight the Cron refresh button while refresh is in flight so the page's loading state stays visible even when prior data remains on screen. (#60394) Thanks @coder-zhuzm.</li>
-<li>Control UI/Overview: prevent gateway access token/password visibility toggle buttons from overlapping their inputs at narrow widths. (#56924) Thanks @bbddbb1.</li>
-<li>CLI/skills JSON: route <code>skills list --json</code>, <code>skills info --json</code>, and <code>skills check --json</code> output to stdout instead of stderr so machine-readable consumers receive JSON on the expected stream again. (#60914; fixes #57599; landed from contributor PR #57611 by @Aftabbs) Thanks @Aftabbs.</li>
-<li>CLI/Commander: preserve Commander-computed exit codes for argument and help-error paths, and cover the user-argv parse mode in the regression tests so invalid CLI invocations no longer report success when exits are intercepted. (#60923) Thanks @Linux2010.</li>
-<li>Cron: replay interrupted recurring jobs on the first gateway restart instead of waiting for a second restart. (#60583) Thanks @joelnishanth.</li>
-<li>Cron: send failure notifications through the job's primary delivery channel using the same session context as successful delivery when no explicit <code>failureDestination</code> is configured. (#60622) Thanks @artwalker.</li>
-<li>Live model switching: only treat explicit user-driven model changes as pending live switches, so fallback rotation, heartbeat overrides, and compaction no longer trip <code>LiveSessionModelSwitchError</code> before making an API call. (#60266) Thanks @kiranvk-2011.</li>
-<li>Exec approvals: reuse durable exact-command <code>allow-always</code> approvals in allowlist mode so identical reruns stop prompting, and tighten Windows interpreter/path approval handling so wrapper and malformed-path cases fail closed more consistently. (#59880, #59780, #58040, #59182) Thanks @luoyanglang, @SnowSky1, and @pgondhi987.</li>
-<li>Node exec approvals: keep node-host <code>system.run</code> approvals bound to the prepared execution plan across async forwarding, so mutable script operands still get approval-time binding and drift revalidation instead of dropping back to unbound execution.</li>
-<li>Agents/exec approvals: let <code>exec-approvals.json</code> agent security override stricter gateway tool defaults so approved subagents can use <code>security: “full”</code> without falling back to allowlist enforcement again. (#60310) Thanks @lml2468.</li>
-<li>Agents/exec: restore <code>host=node</code> routing for node-pinned and <code>host=auto</code> sessions, while still blocking sandboxed <code>auto</code> sessions from jumping to gateway. (#60788) Thanks @openperf.</li>
-<li>Exec/heartbeat: use the canonical <code>exec-event</code> wake reason for <code>notifyOnExit</code> so background exec completions still trigger follow-up turns when <code>HEARTBEAT.md</code> is empty or comments-only. (#41479) Thanks @rstar327.</li>
-<li>Heartbeat: skip wake delivery when the target session lane is already busy so the pending event is retried instead of getting drained too early. (#40526) Thanks @lucky7323.</li>
-<li>Group chats/agent prompts: tell models to minimize empty lines and use normal chat-style spacing so group replies avoid document-style blank-line formatting.</li>
-<li>Providers/OpenAI GPT: treat short approval turns like <code>ok do it</code> and <code>go ahead</code> as immediate action turns, and trim overly memo-like GPT-5 chat confirmations so OpenAI replies stay shorter and more conversational by default.</li>
-<li>Providers/OpenAI Codex: split native <code>contextWindow</code> from runtime <code>contextTokens</code>, keep the default effective cap at <code>272000</code>, and expose a per-model <code>contextTokens</code> override on <code>models.providers.*.models[]</code>.</li>
-<li>Providers/OpenAI-compatible WS: compute fallback token totals from normalized usage when providers omit or zero <code>total_tokens</code>, so DashScope-compatible sessions stop storing zero totals after alias normalization. (#54940) Thanks @lyfuci.</li>
-<li>Agents/OpenAI: mark Claude-compatible file tool schemas as <code>additionalProperties: false</code> so direct OpenAI GPT-5 routes stop rejecting the <code>read</code> tool with invalid strict-schema errors.</li>
-<li>Agents/OpenAI: fall back to <code>strict: false</code> for native OpenAI tool calls when a tool schema is not strict-compatible, and normalize empty-object tool schemas to include <code>required: []</code>, so direct GPT-5 routes stop failing with invalid strict-schema errors like missing <code>path</code> in <code>required</code>.</li>
-<li>Agents/GPT: add explicit work-item lifecycle events for embedded runs, use them to surface real progress more reliably, and stop counting tool-started turns as planning-only retries.</li>
-<li>Plugins/OpenAI: enable <code>gpt-image-1</code> reference-image edits through <code>/images/edits</code> multipart uploads, and stop inferring unsupported resolution overrides when no explicit <code>size</code> or <code>resolution</code> is provided.</li>
-<li>Agents/replay: remove the malformed assistant-content canonicalization repair from replay history sanitization instead of extending that legacy repair path into replay validation.</li>
-<li>Plugins/OpenAI: tune the OpenAI prompt overlay for live-chat cadence so GPT replies stay shorter, more human, and less wall-of-text by default.</li>
-<li>Providers/compat: stop forcing OpenAI-only defaults on proxy and custom OpenAI-compatible routes, preserve native vendor-specific reasoning/tool/streaming behavior across Anthropic-compatible, Moonshot, Mistral, ModelStudio, OpenRouter, xAI, and Z.ai endpoints, and route GitHub Copilot Claude models through Anthropic Messages instead of OpenAI Responses.</li>
-<li>Providers/GitHub Copilot: send IDE identity headers on runtime model requests and GitHub token exchange so IDE-authenticated Copilot runs stop failing with missing <code>Editor-Version</code>. (#60641) Thanks @VACInc and @vincentkoc.</li>
-<li>Providers/OpenRouter failover: classify <code>403 “Key limit exceeded”</code> spending-limit responses as billing so model fallback continues instead of stopping on generic auth. (#59892) Thanks @rockcent.</li>
-<li>Providers/Anthropic: keep <code>claude-cli/*</code> auth on live Claude CLI credentials at runtime, avoid persisting stale bearer-token profiles, and suppress macOS Keychain prompts during non-interactive Claude CLI setup. (#61234) Thanks @darkamenosa.</li>
-<li>Providers/Anthropic: when Claude CLI auth becomes the default, write a real <code>claude-cli</code> auth profile so local and gateway agent runs can use Claude CLI immediately without missing-API-key failures. Thanks @vincentkoc.</li>
-<li>Providers/Anthropic Vertex: honor <code>cacheRetention: “long”</code> with the real 1-hour prompt-cache TTL on Vertex AI endpoints, and default <code>anthropic-vertex</code> cache retention like direct Anthropic. (#60888) Thanks @affsantos.</li>
-<li>Agents/Anthropic: preserve native <code>toolu_*</code> replay ids on direct Anthropic and Anthropic Vertex paths so cache-sensitive history stops rewriting known-valid Anthropic tool-use ids. (#52612)</li>
-<li>Providers/Google: add model-level <code>cacheRetention</code> support for direct Gemini system prompts by creating, reusing, and refreshing <code>cachedContents</code> automatically on Google AI Studio runs. (#51372) Thanks @rafaelmariano-glitch.</li>
-<li>Google Gemini CLI auth: detect bundled npm installs by scanning packaged bundle files for the Gemini OAuth client config, so <code>npm install -g @google/gemini-cli</code> layouts work again. (#60486) Thanks @wzfmini01.</li>
-<li>Google Gemini CLI auth: detect personal OAuth mode from local Gemini settings and skip Code Assist project discovery for those logins, so personal Google accounts stop failing with <code>loadCodeAssist 400 Bad Request</code>. (#49226) Thanks @bobworrall.</li>
-<li>Google Gemini CLI auth: improve OAuth credential discovery across Windows nvm and Homebrew libexec installs, and align Code Assist metadata so Gemini login stops failing on packaged CLI layouts. (#40729) Thanks @hughcube.</li>
-<li>Google Gemini CLI models: add forward-compat support for stable <code>gemini-2.5-*</code> model ids by letting the bundled CLI provider clone them from Google templates, so <code>gemini-2.5-flash-lite</code> and related configured models stop showing up as missing. (#35274) Thanks @mySebbe.</li>
-<li>Google image generation: disable pinned DNS for Gemini image requests and honor explicit <code>pinDns</code> overrides in shared provider HTTP helpers so proxy-backed image generation works again. (#59873) Thanks @luoyanglang.</li>
-<li>Providers/Microsoft Foundry: preserve explicit image capability on normalized Foundry deployments, repair stale GPT/o-series text-only model metadata across gateway and runtime paths, and keep unknown fallback models from borrowing unrelated image support.</li>
-<li>Providers/Model Studio: preserve native streaming usage reporting for DashScope-compatible endpoints even when they are configured under a generic provider key, so streamed token totals stop sticking at zero. (#52395) Thanks @IVY-AI-gif.</li>
-<li>Providers/Z.AI: preserve explicitly registered <code>glm-5-*</code> variants like <code>glm-5-turbo</code> instead of intercepting them with the generic GLM-5 forward-compat shim. (#48185) Thanks @haoyu-haoyu.</li>
-<li>Amazon Bedrock/aws-sdk auth: stop injecting the fake <code>AWS_PROFILE</code> apiKey marker when no AWS auth env vars exist, so instance-role and other default-chain setups keep working without poisoning provider config. (#61194) Thanks @wirjo.</li>
-<li>Agents/Kimi tool-call repair: preserve tool arguments that were already present on streamed tool calls when later malformed deltas fail reevaluation, while still dropping stale repair-only state before <code>toolcall_end</code>.</li>
-<li>Plugins/Kimi Coding: parse tagged tool calls and keep Anthropic-native tool payloads so Kimi coding endpoints execute tools instead of echoing raw markup. (#60051, #60391) Thanks @obviyus and @Eric-Guo.</li>
-<li>Media understanding: auto-register image-capable config providers for vision routing, so custom GLM-style provider ids with image models stop failing with “no media-understanding provider registered”. (#51418) Thanks @xydt-610.</li>
-<li>Plugins/media understanding: enable bundled Groq and Deepgram providers by default so configured transcription models work without extra plugin activation config. (#59982) Thanks @yxjsxy.</li>
-<li>MiniMax/pricing: keep bundled MiniMax highspeed pricing distinct in provider catalogs and preserve the lower M2.5 cache-read pricing when onboarding older MiniMax models. (#54214) Thanks @octo-patch.</li>
-<li>MiniMax: advertise image input on bundled <code>MiniMax-M2.7</code> and <code>MiniMax-M2.7-highspeed</code> model definitions so image-capable flows can route through the M2.7 family correctly. (#54843) Thanks @MerlinMiao88888888.</li>
-<li>Models/MiniMax: honor <code>MINIMAX_API_HOST</code> for implicit bundled MiniMax provider catalogs so China-hosted API-key setups pick <code>api.minimaxi.com/anthropic</code> without manual provider config. (#34524) Thanks @caiqinghua.</li>
-<li>Usage/MiniMax: invert remaining-style <code>usage_percent</code> fields when MiniMax reports only remaining percentage data, so usage bars stop showing nearly-full remaining quota as nearly-exhausted usage. (#60254) Thanks @jwchmodx.</li>
-<li>Usage/MiniMax: let usage snapshots treat <code>minimax-portal</code> and MiniMax CN aliases as the same MiniMax quota surface, and prefer stored MiniMax OAuth before falling back to Coding Plan keys.</li>
-<li>Usage/MiniMax: prefer the chat-model <code>model_remains</code> entry and derive Coding Plan window labels from MiniMax interval timestamps so MiniMax usage snapshots stop picking zero-budget media rows and misreporting 4h windows as <code>5h</code>. (#52349) Thanks @IVY-AI-gif.</li>
-<li>Model picker/providers: treat bundled BytePlus and Volcengine plan aliases as their native providers during setup, and expose their bundled standard/coding catalogs before auth so setup can suggest the right models. (#58819) Thanks @Luckymingxuan.</li>
-<li>Tools/web_search (Kimi): when <code>tools.web.search.kimi.baseUrl</code> is unset, inherit native Moonshot chat <code>baseUrl</code> (<code>.ai</code> / <code>.cn</code>) so China console keys authenticate on the same host as chat. Fixes #44851. (#56769) Thanks @tonga54.</li>
-<li>Agents/Claude CLI: keep non-interactive <code>--permission-mode bypassPermissions</code> when custom <code>cliBackends.claude-cli.args</code> override defaults, including fallback resolution before the runtime plugin registry is active, so cron and heartbeat Claude CLI runs do not regress to interactive approval mode. (#61114) Thanks @cathrynlavery and @thewilloftheshadow.</li>
-<li>Agents/Claude CLI: persist explicit <code>openclaw agent --session-id</code> runs under a stable session key so follow-ups can reuse the stored CLI binding and resume the same underlying Claude session.</li>
-<li>Agents/Claude CLI: persist routed Claude session bindings, rotate them on <code>/new</code> and <code>/reset</code>, and keep live Claude CLI model switches moving across the configured Claude family so resumed sessions follow the real active thread and model. Thanks @vincentkoc.</li>
-<li>Agents/CLI backends: invalidate stored CLI session reuse when local CLI login state or the selected auth profile credential changes, so relogin and token rotation stop resuming stale sessions.</li>
-<li>Agents/Claude CLI/images: reuse stable hydrated image file paths and preserve shared media extensions like HEIC when passing image refs to local CLI runs, so Claude CLI image prompts stop thrashing KV cache prefixes and oddball image formats do not fall back to <code>.bin</code>. Thanks @vincentkoc.</li>
-<li>Agents/compaction: keep assistant tool calls and displaced tool results in the same compaction chunk so strict summarization providers stop rejecting orphaned tool pairs. (#58849) Thanks @openperf.</li>
-<li>Agents/failover: scope Anthropic <code>An unknown error occurred</code> failover matching by provider so generic internal unknown-error text no longer triggers retryable timeout fallback. (#59325) Thanks @aaron-he-zhu.</li>
-<li>Agents/subagents: honor allowlist validation, auth-profile handoff, and session override state when a subagent retries after <code>LiveSessionModelSwitchError</code>. (#58178) Thanks @openperf.</li>
-<li>Agents/runtime: make default subagent allowlists, inherited skills/workspaces, and duplicate session-id resolution behave more predictably, and include value-shape hints in missing-parameter tool errors. (#59944, #59992, #59858, #55317) Thanks @hclsys, @gumadeiras, @joelnishanth, and @priyansh19.</li>
-<li>Agents/pairing: merge completion announce delivery context with the requester session fallback so missing <code>to</code> still reaches the original channel, and include <code>operator.talk.secrets</code> in CLI default operator scopes for node-role device pairing approvals. (#56481) Thanks @maxpetrusenko.</li>
-<li>Agents/scheduling: steer background-now work toward automatic completion wake and treat <code>process</code> polling as on-demand inspection or intervention instead of default completion handling. (#60877) Thanks @vincentkoc.</li>
-<li>Agents/skills: skip <code>.git</code> and <code>node_modules</code> when mirroring skills into sandbox workspaces so read-only sandboxes do not copy repo history or dependency trees. (#61090) Thanks @joelnishanth.</li>
-<li>ACP/agents: inherit the target agent workspace for cross-agent ACP spawns and fall back safely when the inherited workspace no longer exists. (#58438) Thanks @zssggle-rgb.</li>
-<li>ACPX/Windows: preserve backslashes and absolute <code>.exe</code> paths in Claude CLI parsing, and fail fast on wrapper-script targets with guidance to use <code>cmd.exe /c</code>, <code>powershell.exe -File</code>, or <code>node <script></code>. (#60689) Thanks @steipete.</li>
-<li>Auth/failover: persist selected fallback overrides before retrying, shorten <code>auth_permanent</code> lockouts, and refresh websocket/shared-auth sessions only when real auth changes occur so retries and secret rotations behave predictably. (#60404, #60323, #60387) Thanks @extrasmall0 and @mappel-nv.</li>
-<li>Gateway/channels: pin the initial startup channel registry before later plugin-registry churn so configured channels stay visible and <code>channels.status</code> stops falling back to empty <code>channelOrder</code> / <code>channels</code> payloads after runtime plugin loads.</li>
-<li>Prompt caching: order stable workspace project-context files before <code>HEARTBEAT.md</code> and keep <code>HEARTBEAT.md</code> below the system-prompt cache boundary so heartbeat churn does not invalidate the stable project-context prefix. (#58979) Thanks @yozu and @vincentkoc.</li>
-<li>Prompt caching: route Codex Responses and Anthropic Vertex through boundary-aware cache shaping, and report the actual outbound system prompt in cache traces so cache reuse and misses line up with what providers really receive. Thanks @vincentkoc.</li>
-<li>Agents/cache: preserve the full 3-turn prompt-cache image window across tool loops, keep colliding bundled MCP tool definitions deterministic, and reapply Anthropic Vertex cache shaping after payload hook replacements so KV/cache reuse stays stable. Thanks @vincentkoc.</li>
-<li>Status/cache: restore <code>cacheRead</code> and <code>cacheWrite</code> in transcript fallback so <code>/status</code> keeps showing cache hit percentages when session logs are the only complete usage source. (#59247) Thanks @stuartsy.</li>
-<li>Status/usage: let <code>/status</code> and <code>session_status</code> fall back to transcript token totals when the session meta store stayed at zero, so LM Studio, Ollama, DashScope, and similar OpenAI-compatible providers stop showing <code>Context: 0/...</code>. (#55041) Thanks @jjjojoj.</li>
-<li>Mattermost/config schema: accept <code>groups.*.requireMention</code> again so existing Mattermost configs no longer fail strict validation after upgrade. (#58271) Thanks @MoerAI.</li>
-<li>Doctor/config: compare normalized <code>talk</code> configs by deep structural equality instead of key-order-sensitive serialization so <code>openclaw doctor --fix</code> stops repeatedly reporting/applying no-op <code>talk.provider/providers</code> normalization. (#59911) Thanks @ejames-dev.</li>
-<li>Anthropic CLI onboarding: rewrite migrated fallback model refs during non-interactive Claude CLI setup too, so onboarding and scripted setup no longer keep stale <code>anthropic/*</code> fallbacks after switching the primary model to <code>claude-cli/*</code>. Thanks @vincentkoc.</li>
-<li>Models/Anthropic CLI auth: replace migrated <code>agents.defaults.models</code> allowlists when <code>openclaw models auth login --provider anthropic --method cli --set-default</code> switches to <code>claude-cli/*</code>, so stale <code>anthropic/*</code> entries do not linger beside the migrated Claude CLI defaults. Thanks @vincentkoc.</li>
-<li>Doctor/Claude CLI: add dedicated Claude CLI health checks so <code>openclaw doctor</code> can spot missing local installs or broken auth before agent runs fail. Thanks @vincentkoc.</li>
-<li>Plugins/auth-choice: apply provider-owned auth config patches without recursively preserving replaced default-model maps, so Anthropic Claude CLI and similar migrations can intentionally swap model allowlists during onboarding and setup instead of accumulating stale entries. Thanks @vincentkoc.</li>
-<li>Plugins/onboarding: write dotted plugin uiHint paths like Brave <code>webSearch.mode</code> as nested plugin config so <code>llm-context</code> setup stops failing validation. (#61159) Thanks @obviyus.</li>
-<li>Plugins/install: preserve unsafe override flags across linked plugin and hook-pack probes so local <code>--link</code> installs honor the documented override behavior. (#60624) Thanks @JerrettDavis.</li>
-<li>Plugins/cache: inherit the active gateway workspace for provider, web-search, and web-fetch snapshot loads when callers omit <code>workspaceDir</code>, so compatible plugin registries and snapshot caches stop missing on gateway-owned runtime paths. (#61138) Thanks @jzakirov.</li>
-<li>Plugin SDK/context engines: export the missing context-engine result and subagent lifecycle types from <code>openclaw/plugin-sdk</code> so context engine plugins can type <code>ContextEngine</code> implementations without local workarounds. (#61251) Thanks @DaevMithran.</li>
-<li>Tasks/maintenance: reconcile stale cron and chat-backed CLI task rows against live cron-job and agent-run ownership instead of treating any persisted session key as proof that the task is still running. (#60310) Thanks @lml2468.</li>
-<li>Plugins: suppress trust-warning noise during non-activating snapshot and CLI metadata loads. (#61427) Thanks @gumadeiras.</li>
-<li>Agents/video generation: accept <code>agents.defaults.videoGenerationModel</code> in strict config validation and <code>openclaw config set/get</code>, so gateways using <code>video_generate</code> no longer fail to boot after enabling a video model.</li>
-<li>Matrix/streaming: add a quiet preview mode for streamed Matrix replies, keep legacy <code>partial</code> preview-first behavior, and finalize quiet media captions correctly so previews stop notifying early without dropping final text semantics. (#61450) Thanks @gumadeiras.</li>
-<li>Gateway/shutdown: bound websocket-server shutdown even when no tracked clients remain, so gateway restarts stop hanging until the watchdog kills the process. (#61565) Thanks @mbelinky.</li>
-<li>Control UI/multilingual: localize the remaining shared channel, instances, nodes, and gateway-confirmation strings so the dashboard stops mixing translated UI with hardcoded English labels. Thanks @vincentkoc.</li>
-<li>Discord/media: raise the default inbound and outbound media cap to <code>100MB</code> so Discord matches Telegram more closely and larger attachments stop failing on the old low default.</li>
-<li>Matrix: keep direct transport requests on the pinned dispatcher by routing them through undici runtime fetch, so Matrix clients resume syncing on newer runtimes without dropping the validated address binding. (#61595) Thanks @gumadeiras.</li>
-<li>Plugins/facades: resolve globally installed bundled-plugin runtime facades from registry roots so bundled channels like LINE still boot when the winning plugin install lives under the global extensions directory with an encoded scoped folder name. (#61297) Thanks @openperf.</li>
-</ul>
-<p><a href="https://github.com/openclaw/openclaw/blob/main/CHANGELOG.md">View full changelog</a></p>
-]]></description>
-            <enclosure url="https://github.com/openclaw/openclaw/releases/download/v2026.4.5/OpenClaw-2026.4.5.zip" length="25050620" type="application/octet-stream" sparkle:edSignature="gVbB/73byllY0utwGIi3P5t0FyvLldeR0Uq2pAa6LTBr8VyZlwNCZ2xPlt2zDFshSUBFKxicYzohOmfJ28ACBg=="/>
-        </item>
     </channel>
 </rss>

apps/android/app/build.gradle.kts

@@ -65,8 +65,8 @@ android {
         applicationId = "ai.openclaw.app"
         minSdk = 31
         targetSdk = 36
-        versionCode = 2026040901
-        versionName = "2026.4.9"
+        versionCode = 2026041001
+        versionName = "2026.4.10"
         ndk {
             // Support all major ABIs — native libs are tiny (~47 KB per ABI)
             abiFilters += listOf("armeabi-v7a", "arm64-v8a", "x86", "x86_64")

apps/ios/CHANGELOG.md

@@ -8,6 +8,10 @@
 
 ### Fixed
 
+## 2026.4.10 - 2026-04-10
+
+Maintenance update for the current OpenClaw release.
+
 ## 2026.4.6 - 2026-04-06
 
 First App Store release of OpenClaw for iPhone. Pair with your OpenClaw Gateway to use chat, voice, sharing, and device actions from iOS.

apps/ios/Config/Version.xcconfig

@@ -2,8 +2,8 @@
 // Source of truth: apps/ios/version.json
 // Generated by scripts/ios-sync-versioning.ts.
 
-OPENCLAW_IOS_VERSION = 2026.4.6
-OPENCLAW_MARKETING_VERSION = 2026.4.6
+OPENCLAW_IOS_VERSION = 2026.4.10
+OPENCLAW_MARKETING_VERSION = 2026.4.10
 OPENCLAW_BUILD_VERSION = 1
 
 #include? "../build/Version.xcconfig"

apps/ios/fastlane/metadata/en-US/release_notes.txt

@@ -1 +1 @@
-First App Store release of OpenClaw for iPhone. Pair with your OpenClaw Gateway to use chat, voice, sharing, and device actions from iOS.
+Maintenance update for the current OpenClaw release.

apps/ios/version.json

@@ -1,3 +1,3 @@
 {
-  "version": "2026.4.6"
+  "version": "2026.4.10"
 }

apps/macos/Package.resolved

@@ -1,5 +1,5 @@
 {
-  "originHash" : "fb90e7b1977f43661ac91681d16da11f9ddd85630407ef170eaada0a6ee39972",
+  "originHash" : "31972864afdac74537794e1a3b7bd22484c09ec1be8e3624fb9ea582e9222ad9",
   "pins" : [
     {
       "identity" : "axorcist",
@@ -28,6 +28,15 @@
         "version" : "0.1.0"
       }
     },
+    {
+      "identity" : "eventsource",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/mattt/EventSource.git",
+      "state" : {
+        "revision" : "a3a85a85214caf642abaa96ae664e4c772a59f6e",
+        "version" : "1.4.1"
+      }
+    },
     {
       "identity" : "menubarextraaccess",
       "kind" : "remoteSourceControl",
@@ -37,6 +46,33 @@
         "version" : "1.2.2"
       }
     },
+    {
+      "identity" : "mlx-audio-swift",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/Blaizzy/mlx-audio-swift",
+      "state" : {
+        "revision" : "fcbd04daa1bfebe881932f630af2ba6ce9af3274",
+        "version" : "0.1.2"
+      }
+    },
+    {
+      "identity" : "mlx-swift",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/ml-explore/mlx-swift.git",
+      "state" : {
+        "revision" : "61b9e011e09a62b489f6bd647958f1555bdf2896",
+        "version" : "0.31.3"
+      }
+    },
+    {
+      "identity" : "mlx-swift-lm",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/ml-explore/mlx-swift-lm.git",
+      "state" : {
+        "revision" : "25b00d4e22e61ec9c41efda47990cd2084ec87ff",
+        "version" : "2.31.3"
+      }
+    },
     {
       "identity" : "peekaboo",
       "kind" : "remoteSourceControl",
@@ -64,6 +100,33 @@
         "version" : "1.2.1"
       }
     },
+    {
+      "identity" : "swift-asn1",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/apple/swift-asn1.git",
+      "state" : {
+        "revision" : "9f542610331815e29cc3821d3b6f488db8715517",
+        "version" : "1.6.0"
+      }
+    },
+    {
+      "identity" : "swift-atomics",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/apple/swift-atomics.git",
+      "state" : {
+        "revision" : "b601256eab081c0f92f059e12818ac1d4f178ff7",
+        "version" : "1.3.0"
+      }
+    },
+    {
+      "identity" : "swift-collections",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/apple/swift-collections.git",
+      "state" : {
+        "revision" : "6675bc0ff86e61436e615df6fc5174e043e57924",
+        "version" : "1.4.1"
+      }
+    },
     {
       "identity" : "swift-concurrency-extras",
       "kind" : "remoteSourceControl",
@@ -73,6 +136,33 @@
         "version" : "1.3.2"
       }
     },
+    {
+      "identity" : "swift-crypto",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/apple/swift-crypto.git",
+      "state" : {
+        "revision" : "bb4ba815dab96d4edc1e0b86d7b9acf9ff973a84",
+        "version" : "4.3.1"
+      }
+    },
+    {
+      "identity" : "swift-huggingface",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/huggingface/swift-huggingface.git",
+      "state" : {
+        "revision" : "b721959445b617d0bf03910b2b4aced345fd93bf",
+        "version" : "0.9.0"
+      }
+    },
+    {
+      "identity" : "swift-jinja",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/huggingface/swift-jinja.git",
+      "state" : {
+        "revision" : "0aeefadec459ce8e11a333769950fb86183aca43",
+        "version" : "2.3.5"
+      }
+    },
     {
       "identity" : "swift-log",
       "kind" : "remoteSourceControl",
@@ -82,6 +172,15 @@
         "version" : "1.10.1"
       }
     },
+    {
+      "identity" : "swift-nio",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/apple/swift-nio.git",
+      "state" : {
+        "revision" : "558f24a4647193b5a0e2104031b71c55d31ff83a",
+        "version" : "2.97.1"
+      }
+    },
     {
       "identity" : "swift-numerics",
       "kind" : "remoteSourceControl",
@@ -109,6 +208,15 @@
         "version" : "1.6.4"
       }
     },
+    {
+      "identity" : "swift-transformers",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/huggingface/swift-transformers.git",
+      "state" : {
+        "revision" : "58c4bc11963a140358d791f678a60a2745a23146",
+        "version" : "1.2.1"
+      }
+    },
     {
       "identity" : "swiftui-math",
       "kind" : "remoteSourceControl",
@@ -126,6 +234,15 @@
         "revision" : "5b06b811c0f5313b6b84bbef98c635a630638c38",
         "version" : "0.3.1"
       }
+    },
+    {
+      "identity" : "yyjson",
+      "kind" : "remoteSourceControl",
+      "location" : "https://github.com/ibireme/yyjson.git",
+      "state" : {
+        "revision" : "8b4a38dc994a110abaec8a400615567bd996105f",
+        "version" : "0.12.0"
+      }
     }
   ],
   "version" : 3

apps/macos/Package.swift

@@ -20,6 +20,7 @@ let package = Package(
         .package(url: "https://github.com/apple/swift-log.git", from: "1.10.1"),
         .package(url: "https://github.com/sparkle-project/Sparkle", from: "2.9.0"),
         .package(url: "https://github.com/steipete/Peekaboo.git", branch: "main"),
+        .package(url: "https://github.com/Blaizzy/mlx-audio-swift", exact: "0.1.2"),
         .package(path: "../shared/OpenClawKit"),
         .package(path: "../../Swabble"),
     ],
@@ -54,6 +55,7 @@ let package = Package(
                 .product(name: "Sparkle", package: "Sparkle"),
                 .product(name: "PeekabooBridge", package: "Peekaboo"),
                 .product(name: "PeekabooAutomationKit", package: "Peekaboo"),
+                .product(name: "MLXAudioTTS", package: "mlx-audio-swift"),
             ],
             exclude: [
                 "Resources/Info.plist",

apps/macos/Sources/OpenClaw/CommandResolver.swift

@@ -235,7 +235,8 @@ enum CommandResolver {
         extraArgs: [String] = [],
         defaults: UserDefaults = .standard,
         configRoot: [String: Any]? = nil,
-        searchPaths: [String]? = nil) -> [String]
+        searchPaths: [String]? = nil,
+        projectRoot: URL? = nil) -> [String]
     {
         let settings = self.connectionSettings(defaults: defaults, configRoot: configRoot)
         if settings.mode == .remote, let ssh = self.sshNodeCommand(
@@ -246,7 +247,7 @@ enum CommandResolver {
             return ssh
         }
 
-        let root = self.projectRoot()
+        let root = projectRoot ?? self.projectRoot()
         if let openclawPath = self.projectOpenClawExecutable(projectRoot: root) {
             return [openclawPath, subcommand] + extraArgs
         }
@@ -289,14 +290,16 @@ enum CommandResolver {
         extraArgs: [String] = [],
         defaults: UserDefaults = .standard,
         configRoot: [String: Any]? = nil,
-        searchPaths: [String]? = nil) -> [String]
+        searchPaths: [String]? = nil,
+        projectRoot: URL? = nil) -> [String]
     {
         self.openclawNodeCommand(
             subcommand: subcommand,
             extraArgs: extraArgs,
             defaults: defaults,
             configRoot: configRoot,
-            searchPaths: searchPaths)
+            searchPaths: searchPaths,
+            projectRoot: projectRoot)
     }
 
     // MARK: - SSH helpers

apps/macos/Sources/OpenClaw/HostEnvSanitizer.swift

@@ -8,6 +8,8 @@ struct HostEnvOverrideDiagnostics: Equatable {
 enum HostEnvSanitizer {
     /// Generated from src/infra/host-env-security-policy.json via scripts/generate-host-env-security-policy-swift.mjs.
     /// Parity is validated by src/infra/host-env-security.policy-parity.test.ts.
+    private static let blockedInheritedKeys = HostEnvSecurityPolicy.blockedInheritedKeys
+    private static let blockedInheritedPrefixes = HostEnvSecurityPolicy.blockedInheritedPrefixes
     private static let blockedKeys = HostEnvSecurityPolicy.blockedKeys
     private static let blockedPrefixes = HostEnvSecurityPolicy.blockedPrefixes
     private static let blockedOverrideKeys = HostEnvSecurityPolicy.blockedOverrideKeys
@@ -28,6 +30,11 @@ enum HostEnvSanitizer {
         return self.blockedPrefixes.contains(where: { upperKey.hasPrefix($0) })
     }
 
+    private static func isBlockedInherited(_ upperKey: String) -> Bool {
+        if self.blockedInheritedKeys.contains(upperKey) { return true }
+        return self.blockedInheritedPrefixes.contains(where: { upperKey.hasPrefix($0) })
+    }
+
     private static func isBlockedOverride(_ upperKey: String) -> Bool {
         if self.blockedOverrideKeys.contains(upperKey) { return true }
         return self.blockedOverridePrefixes.contains(where: { upperKey.hasPrefix($0) })
@@ -113,7 +120,7 @@ enum HostEnvSanitizer {
             let key = rawKey.trimmingCharacters(in: .whitespacesAndNewlines)
             guard !key.isEmpty else { continue }
             let upper = key.uppercased()
-            if self.isBlocked(upper) { continue }
+            if self.isBlockedInherited(upper) { continue }
             merged[key] = value
         }
 

apps/macos/Sources/OpenClaw/HostEnvSecurityPolicy.generated.swift

@@ -5,20 +5,232 @@
 import Foundation
 
 enum HostEnvSecurityPolicy {
+    static let blockedInheritedKeys: Set<String> = [
+        "_JAVA_OPTIONS",
+        "AMQP_URL",
+        "ANSIBLE_CALLBACK_PLUGINS",
+        "ANSIBLE_COLLECTIONS_PATH",
+        "ANSIBLE_CONFIG",
+        "ANSIBLE_CONNECTION_PLUGINS",
+        "ANSIBLE_FILTER_PLUGINS",
+        "ANSIBLE_INVENTORY_PLUGINS",
+        "ANSIBLE_LIBRARY",
+        "ANSIBLE_LOOKUP_PLUGINS",
+        "ANSIBLE_MODULE_UTILS",
+        "ANSIBLE_REMOTE_TEMP",
+        "ANSIBLE_ROLES_PATH",
+        "ANSIBLE_STRATEGY_PLUGINS",
+        "ANT_OPTS",
+        "AWS_ACCESS_KEY_ID",
+        "AWS_CONTAINER_CREDENTIALS_FULL_URI",
+        "AWS_CONTAINER_CREDENTIALS_RELATIVE_URI",
+        "AWS_SECRET_ACCESS_KEY",
+        "AWS_SECURITY_TOKEN",
+        "AWS_SESSION_TOKEN",
+        "AZURE_CLIENT_ID",
+        "AZURE_CLIENT_SECRET",
+        "BASH_ENV",
+        "BROWSER",
+        "BUN_CONFIG_REGISTRY",
+        "BUNDLE_GEMFILE",
+        "BZR_EDITOR",
+        "BZR_PLUGIN_PATH",
+        "BZR_SSH",
+        "C_INCLUDE_PATH",
+        "CARGO_BUILD_RUSTC",
+        "CARGO_BUILD_RUSTC_WRAPPER",
+        "CARGO_HOME",
+        "CATALINA_OPTS",
+        "CC",
+        "CFLAGS",
+        "CGO_CFLAGS",
+        "CGO_LDFLAGS",
+        "CLASSPATH",
+        "CMAKE_C_COMPILER",
+        "CMAKE_CXX_COMPILER",
+        "CMAKE_TOOLCHAIN_FILE",
+        "COMPOSER_HOME",
+        "CONFIG_SHELL",
+        "CONFIG_SITE",
+        "CORECLR_PROFILER",
+        "CORECLR_PROFILER_PATH",
+        "CPATH",
+        "CPLUS_INCLUDE_PATH",
+        "CURL_HOME",
+        "CXX",
+        "DATABASE_URL",
+        "DENO_DIR",
+        "DOTNET_ADDITIONAL_DEPS",
+        "DOTNET_STARTUP_HOOKS",
+        "EDITOR",
+        "ELIXIR_ERL_OPTIONS",
+        "EMACSLOADPATH",
+        "ENV",
+        "ERL_AFLAGS",
+        "ERL_FLAGS",
+        "ERL_ZFLAGS",
+        "EXINIT",
+        "FCEDIT",
+        "GCONV_PATH",
+        "GEM_HOME",
+        "GEM_PATH",
+        "GH_TOKEN",
+        "GIT_ALTERNATE_OBJECT_DIRECTORIES",
+        "GIT_ASKPASS",
+        "GIT_COMMON_DIR",
+        "GIT_DIR",
+        "GIT_EDITOR",
+        "GIT_EXEC_PATH",
+        "GIT_EXTERNAL_DIFF",
+        "GIT_HOOK_PATH",
+        "GIT_INDEX_FILE",
+        "GIT_NAMESPACE",
+        "GIT_OBJECT_DIRECTORY",
+        "GIT_PROXY_COMMAND",
+        "GIT_SEQUENCE_EDITOR",
+        "GIT_SSH",
+        "GIT_SSH_COMMAND",
+        "GIT_SSL_CAINFO",
+        "GIT_SSL_CAPATH",
+        "GIT_SSL_NO_VERIFY",
+        "GIT_TEMPLATE_DIR",
+        "GIT_WORK_TREE",
+        "GITHUB_TOKEN",
+        "GITLAB_TOKEN",
+        "GLIBC_TUNABLES",
+        "GOENV",
+        "GOFLAGS",
+        "GONOPROXY",
+        "GONOSUMCHECK",
+        "GONOSUMDB",
+        "GOPATH",
+        "GOPRIVATE",
+        "GOPROXY",
+        "GRADLE_OPTS",
+        "GVIMINIT",
+        "HELM_HOME",
+        "HELM_PLUGINS",
+        "HGRCPATH",
+        "HOSTALIASES",
+        "IFS",
+        "JAVA_OPTS",
+        "JAVA_TOOL_OPTIONS",
+        "JDK_JAVA_OPTIONS",
+        "JULIA_EDITOR",
+        "LDFLAGS",
+        "LESSCLOSE",
+        "LESSOPEN",
+        "LIBRARY_PATH",
+        "LUA_CPATH",
+        "LUA_INIT",
+        "LUA_INIT_5_1",
+        "LUA_INIT_5_2",
+        "LUA_INIT_5_3",
+        "LUA_INIT_5_4",
+        "LUA_PATH",
+        "MAKEFLAGS",
+        "MAVEN_OPTS",
+        "MFLAGS",
+        "MONGODB_URI",
+        "MYVIMRC",
+        "NODE_AUTH_TOKEN",
+        "NODE_OPTIONS",
+        "NODE_PATH",
+        "NPM_TOKEN",
+        "OBJC_INCLUDE_PATH",
+        "OPENSSL_CONF",
+        "OPENSSL_ENGINES",
+        "PACKER_PLUGIN_PATH",
+        "PERL5DB",
+        "PERL5DBCMD",
+        "PERL5LIB",
+        "PERL5OPT",
+        "PHP_INI_SCAN_DIR",
+        "PHPRC",
+        "PIP_CONFIG_FILE",
+        "PIP_EXTRA_INDEX_URL",
+        "PIP_FIND_LINKS",
+        "PIP_INDEX_URL",
+        "PIP_PYPI_URL",
+        "PIP_TRUSTED_HOST",
+        "PROMPT_COMMAND",
+        "PS4",
+        "PYTHONBREAKPOINT",
+        "PYTHONHOME",
+        "PYTHONPATH",
+        "PYTHONSTARTUP",
+        "PYTHONUSERBASE",
+        "R_ENVIRON",
+        "R_ENVIRON_USER",
+        "R_LIBS_USER",
+        "R_PROFILE",
+        "R_PROFILE_USER",
+        "REDIS_URL",
+        "RUBYLIB",
+        "RUBYOPT",
+        "RUBYSHELL",
+        "RUSTC_WRAPPER",
+        "RUSTFLAGS",
+        "SBT_OPTS",
+        "SHELL",
+        "SHELLOPTS",
+        "SSH_ASKPASS",
+        "SSLKEYLOGFILE",
+        "SUDO_ASKPASS",
+        "SUDO_EDITOR",
+        "SVN_EDITOR",
+        "SVN_SSH",
+        "TF_CLI_CONFIG_FILE",
+        "TF_PLUGIN_CACHE_DIR",
+        "UV_DEFAULT_INDEX",
+        "UV_EXTRA_INDEX_URL",
+        "UV_INDEX",
+        "UV_INDEX_URL",
+        "UV_PYTHON",
+        "VAGRANT_VAGRANTFILE",
+        "VIMINIT",
+        "VIRTUAL_ENV",
+        "VISUAL",
+        "WGETRC",
+        "XDG_CONFIG_DIRS",
+        "XDG_CONFIG_HOME",
+        "YARN_RC_FILENAME"
+    ]
+
+    static let blockedInheritedPrefixes: [String] = [
+        "BASH_FUNC_",
+        "DYLD_",
+        "LD_"
+    ]
+
     static let blockedKeys: Set<String> = [
         "_JAVA_OPTIONS",
         "ANT_OPTS",
         "BASH_ENV",
         "BROWSER",
+        "BZR_EDITOR",
+        "BZR_PLUGIN_PATH",
+        "BZR_SSH",
         "CARGO_BUILD_RUSTC",
         "CARGO_BUILD_RUSTC_WRAPPER",
+        "CATALINA_OPTS",
         "CC",
         "CMAKE_C_COMPILER",
         "CMAKE_CXX_COMPILER",
+        "CMAKE_TOOLCHAIN_FILE",
+        "CONFIG_SHELL",
+        "CONFIG_SITE",
+        "CORECLR_PROFILER",
         "CXX",
         "DOTNET_ADDITIONAL_DEPS",
         "DOTNET_STARTUP_HOOKS",
+        "ELIXIR_ERL_OPTIONS",
+        "EMACSLOADPATH",
         "ENV",
+        "ERL_AFLAGS",
+        "ERL_FLAGS",
+        "ERL_ZFLAGS",
+        "EXINIT",
         "GCONV_PATH",
         "GIT_ALTERNATE_OBJECT_DIRECTORIES",
         "GIT_COMMON_DIR",
@@ -26,6 +238,7 @@ enum HostEnvSecurityPolicy {
         "GIT_EDITOR",
         "GIT_EXEC_PATH",
         "GIT_EXTERNAL_DIFF",
+        "GIT_HOOK_PATH",
         "GIT_INDEX_FILE",
         "GIT_NAMESPACE",
         "GIT_OBJECT_DIRECTORY",
@@ -37,42 +250,85 @@ enum HostEnvSecurityPolicy {
         "GIT_WORK_TREE",
         "GLIBC_TUNABLES",
         "GRADLE_OPTS",
+        "GVIMINIT",
+        "HELM_PLUGINS",
         "HGRCPATH",
+        "HOSTALIASES",
         "IFS",
         "JAVA_OPTS",
         "JAVA_TOOL_OPTIONS",
         "JDK_JAVA_OPTIONS",
+        "JULIA_EDITOR",
+        "LUA_INIT",
+        "LUA_INIT_5_1",
+        "LUA_INIT_5_2",
+        "LUA_INIT_5_3",
+        "LUA_INIT_5_4",
         "MAKEFLAGS",
         "MAVEN_OPTS",
         "MFLAGS",
+        "MYVIMRC",
         "NODE_OPTIONS",
         "NODE_PATH",
+        "PACKER_PLUGIN_PATH",
         "PERL5LIB",
         "PERL5OPT",
         "PS4",
         "PYTHONBREAKPOINT",
         "PYTHONHOME",
         "PYTHONPATH",
+        "R_ENVIRON",
+        "R_ENVIRON_USER",
+        "R_PROFILE",
+        "R_PROFILE_USER",
         "RUBYLIB",
         "RUBYOPT",
+        "RUBYSHELL",
         "RUSTC_WRAPPER",
         "SBT_OPTS",
         "SHELL",
         "SHELLOPTS",
-        "SSLKEYLOGFILE"
+        "SSLKEYLOGFILE",
+        "SUDO_ASKPASS",
+        "SVN_EDITOR",
+        "SVN_SSH",
+        "VAGRANT_VAGRANTFILE",
+        "VIMINIT"
     ]
 
     static let blockedOverrideKeys: Set<String> = [
         "ALL_PROXY",
+        "AMQP_URL",
+        "ANSIBLE_CALLBACK_PLUGINS",
+        "ANSIBLE_COLLECTIONS_PATH",
+        "ANSIBLE_CONFIG",
+        "ANSIBLE_CONNECTION_PLUGINS",
+        "ANSIBLE_FILTER_PLUGINS",
+        "ANSIBLE_INVENTORY_PLUGINS",
+        "ANSIBLE_LIBRARY",
+        "ANSIBLE_LOOKUP_PLUGINS",
+        "ANSIBLE_MODULE_UTILS",
+        "ANSIBLE_REMOTE_TEMP",
+        "ANSIBLE_ROLES_PATH",
+        "ANSIBLE_STRATEGY_PLUGINS",
+        "AWS_ACCESS_KEY_ID",
         "AWS_CONFIG_FILE",
+        "AWS_CONTAINER_CREDENTIALS_FULL_URI",
+        "AWS_CONTAINER_CREDENTIALS_RELATIVE_URI",
+        "AWS_SECRET_ACCESS_KEY",
+        "AWS_SECURITY_TOKEN",
+        "AWS_SESSION_TOKEN",
         "AWS_SHARED_CREDENTIALS_FILE",
         "AWS_WEB_IDENTITY_TOKEN_FILE",
         "AZURE_AUTH_LOCATION",
+        "AZURE_CLIENT_ID",
+        "AZURE_CLIENT_SECRET",
         "BUN_CONFIG_REGISTRY",
         "BUNDLE_GEMFILE",
         "C_INCLUDE_PATH",
         "CARGO_BUILD_RUSTC_WRAPPER",
         "CARGO_HOME",
+        "CFLAGS",
         "CGO_CFLAGS",
         "CGO_LDFLAGS",
         "CLASSPATH",
@@ -82,6 +338,7 @@ enum HostEnvSecurityPolicy {
         "CPLUS_INCLUDE_PATH",
         "CURL_CA_BUNDLE",
         "CURL_HOME",
+        "DATABASE_URL",
         "DENO_DIR",
         "DOCKER_CERT_PATH",
         "DOCKER_CONTEXT",
@@ -91,6 +348,7 @@ enum HostEnvSecurityPolicy {
         "FCEDIT",
         "GEM_HOME",
         "GEM_PATH",
+        "GH_TOKEN",
         "GIT_ALTERNATE_OBJECT_DIRECTORIES",
         "GIT_ASKPASS",
         "GIT_COMMON_DIR",
@@ -106,6 +364,8 @@ enum HostEnvSecurityPolicy {
         "GIT_SSL_CAPATH",
         "GIT_SSL_NO_VERIFY",
         "GIT_WORK_TREE",
+        "GITHUB_TOKEN",
+        "GITLAB_TOKEN",
         "GOENV",
         "GOFLAGS",
         "GONOPROXY",
@@ -123,6 +383,7 @@ enum HostEnvSecurityPolicy {
         "HTTP_PROXY",
         "HTTPS_PROXY",
         "KUBECONFIG",
+        "LDFLAGS",
         "LESSCLOSE",
         "LESSOPEN",
         "LIBRARY_PATH",
@@ -131,9 +392,12 @@ enum HostEnvSecurityPolicy {
         "MAKEFLAGS",
         "MANPAGER",
         "MFLAGS",
+        "MONGODB_URI",
         "NO_PROXY",
+        "NODE_AUTH_TOKEN",
         "NODE_EXTRA_CA_CERTS",
         "NODE_TLS_REJECT_UNAUTHORIZED",
+        "NPM_TOKEN",
         "OBJC_INCLUDE_PATH",
         "OPENSSL_CONF",
         "OPENSSL_ENGINES",
@@ -151,13 +415,18 @@ enum HostEnvSecurityPolicy {
         "PROMPT_COMMAND",
         "PYTHONSTARTUP",
         "PYTHONUSERBASE",
+        "R_LIBS_USER",
+        "REDIS_URL",
         "REQUESTS_CA_BUNDLE",
         "RUSTC_WRAPPER",
         "RUSTFLAGS",
         "SSH_ASKPASS",
+        "SSH_AUTH_SOCK",
         "SSL_CERT_DIR",
         "SSL_CERT_FILE",
         "SUDO_EDITOR",
+        "TF_CLI_CONFIG_FILE",
+        "TF_PLUGIN_CACHE_DIR",
         "UV_DEFAULT_INDEX",
         "UV_EXTRA_INDEX_URL",
         "UV_INDEX",
@@ -166,6 +435,7 @@ enum HostEnvSecurityPolicy {
         "VIRTUAL_ENV",
         "VISUAL",
         "WGETRC",
+        "XDG_CONFIG_DIRS",
         "XDG_CONFIG_HOME",
         "YARN_RC_FILENAME",
         "ZDOTDIR"
@@ -174,7 +444,8 @@ enum HostEnvSecurityPolicy {
     static let blockedOverridePrefixes: [String] = [
         "CARGO_REGISTRIES_",
         "GIT_CONFIG_",
-        "NPM_CONFIG_"
+        "NPM_CONFIG_",
+        "TF_VAR_"
     ]
 
     static let blockedPrefixes: [String] = [

apps/macos/Sources/OpenClaw/Resources/Info.plist

@@ -15,9 +15,9 @@
     <key>CFBundlePackageType</key>
     <string>APPL</string>
     <key>CFBundleShortVersionString</key>
-    <string>2026.4.9</string>
+    <string>2026.4.10</string>
     <key>CFBundleVersion</key>
-    <string>2026040901</string>
+    <string>2026041001</string>
     <key>CFBundleIconFile</key>
     <string>OpenClaw</string>
     <key>CFBundleURLTypes</key>

apps/macos/Sources/OpenClaw/ShellExecutor.swift

@@ -73,8 +73,10 @@ enum ShellExecutor {
                 group.addTask { await waitTask.value }
                 group.addTask {
                     try? await Task.sleep(nanoseconds: nanos)
-                    if process.isRunning { process.terminate() }
-                    _ = await waitTask.value // drain pipes after termination
+                    guard process.isRunning else {
+                        return await waitTask.value
+                    }
+                    process.terminate()
                     return ShellResult(
                         stdout: "",
                         stderr: "",

apps/macos/Sources/OpenClaw/TalkMLXSpeechSynthesizer.swift

@@ -0,0 +1,178 @@
+import Foundation
+import MLXAudioTTS
+import OSLog
+
+// swiftformat:disable wrap wrapMultilineStatementBraces trailingCommas redundantSelf extensionAccessControl
+/// Runtime access stays serialized through `TalkModeRuntime` actor helper methods.
+final class TalkMLXSpeechSynthesizer {
+    enum SynthesizeError: Error {
+        case canceled
+        case modelLoadFailed(String)
+        case audioGenerationFailed
+        case audioPlaybackFailed
+        case timedOut
+    }
+
+    static let shared = TalkMLXSpeechSynthesizer()
+    static let defaultModelRepo = "mlx-community/Soprano-80M-bf16"
+
+    private let logger = Logger(subsystem: "ai.openclaw", category: "talk.mlx")
+    private var currentToken = UUID()
+    private var modelRepo: String?
+    private var model: (any SpeechGenerationModel)?
+
+    private init() {}
+
+    func stop() {
+        self.currentToken = UUID()
+    }
+
+    func synthesize(
+        text: String,
+        modelRepo: String?,
+        language: String?,
+        voicePreset: String?) async throws -> Data {
+        let trimmed = text.trimmingCharacters(in: .whitespacesAndNewlines)
+        guard !trimmed.isEmpty else { return Data() }
+
+        self.stop()
+        let token = UUID()
+        self.currentToken = token
+
+        let resolvedRepo = Self.resolvedModelRepo(modelRepo)
+        let rawModel = try await self.loadModel(
+            modelRepo: resolvedRepo,
+            token: token)
+        let model = UncheckedSpeechModel(raw: rawModel)
+        guard self.currentToken == token else {
+            throw SynthesizeError.canceled
+        }
+
+        let audioData: Data
+        do {
+            let audio = try await model.generateAudio(
+                text: trimmed,
+                voice: voicePreset,
+                language: language)
+            audioData = Self.makeWavData(
+                samples: audio,
+                sampleRate: Double(model.sampleRateValue()))
+        } catch {
+            self.logger.error(
+                "talk mlx generation failed: \(error.localizedDescription, privacy: .public)")
+            throw SynthesizeError.audioGenerationFailed
+        }
+
+        guard self.currentToken == token else {
+            throw SynthesizeError.canceled
+        }
+        return audioData
+    }
+
+    private func loadModel(
+        modelRepo: String,
+        token: UUID) async throws -> any SpeechGenerationModel {
+        if let model = self.model, self.modelRepo == modelRepo {
+            return model
+        }
+
+        self.logger.info("talk mlx loading modelRepo=\(modelRepo, privacy: .public)")
+        do {
+            let model = try await TTS.loadModel(modelRepo: modelRepo)
+            guard self.currentToken == token else {
+                throw SynthesizeError.canceled
+            }
+            self.model = model
+            self.modelRepo = modelRepo
+            return model
+        } catch is CancellationError {
+            throw SynthesizeError.canceled
+        } catch {
+            self.logger.error(
+                "talk mlx load failed: \(error.localizedDescription, privacy: .public)")
+            throw SynthesizeError.modelLoadFailed(modelRepo)
+        }
+    }
+
+    private static func resolvedModelRepo(_ modelRepo: String?) -> String {
+        let trimmed = modelRepo?.trimmingCharacters(in: .whitespacesAndNewlines) ?? ""
+        return trimmed.isEmpty ? Self.defaultModelRepo : trimmed
+    }
+
+    private static func makeWavData(samples: [Float], sampleRate: Double) -> Data {
+        let channels: UInt16 = 1
+        let bitsPerSample: UInt16 = 16
+        let blockAlign = channels * (bitsPerSample / 8)
+        let sampleRateInt = UInt32(sampleRate.rounded())
+        let byteRate = sampleRateInt * UInt32(blockAlign)
+        let dataSize = UInt32(samples.count) * UInt32(blockAlign)
+
+        var data = Data(capacity: Int(44 + dataSize))
+        data.append(contentsOf: [0x52, 0x49, 0x46, 0x46]) // RIFF
+        data.appendLEUInt32(36 + dataSize)
+        data.append(contentsOf: [0x57, 0x41, 0x56, 0x45]) // WAVE
+
+        data.append(contentsOf: [0x66, 0x6D, 0x74, 0x20]) // fmt
+        data.appendLEUInt32(16)
+        data.appendLEUInt16(1)
+        data.appendLEUInt16(channels)
+        data.appendLEUInt32(sampleRateInt)
+        data.appendLEUInt32(byteRate)
+        data.appendLEUInt16(blockAlign)
+        data.appendLEUInt16(bitsPerSample)
+
+        data.append(contentsOf: [0x64, 0x61, 0x74, 0x61]) // data
+        data.appendLEUInt32(dataSize)
+
+        for sample in samples {
+            let clamped = max(-1.0, min(1.0, sample))
+            let scaled = Int16((clamped * Float(Int16.max)).rounded())
+            data.appendLEInt16(scaled)
+        }
+        return data
+    }
+}
+
+extension TalkMLXSpeechSynthesizer: @unchecked Sendable {}
+
+private struct UncheckedSpeechModel {
+    let raw: any SpeechGenerationModel
+
+    func sampleRateValue() -> Int {
+        raw.sampleRate
+    }
+
+    func generateAudio(
+        text: String,
+        voice: String?,
+        language: String?) async throws -> [Float] {
+        let generatedAudio = try await raw.generate(
+            text: text,
+            voice: voice,
+            refAudio: nil,
+            refText: nil,
+            language: language)
+        return generatedAudio.asArray(Float.self)
+    }
+}
+
+extension UncheckedSpeechModel: @unchecked Sendable {}
+
+extension Data {
+    fileprivate mutating func appendLEUInt16(_ value: UInt16) {
+        var littleEndian = value.littleEndian
+        Swift.withUnsafeBytes(of: &littleEndian) { append(contentsOf: $0) }
+    }
+
+    fileprivate mutating func appendLEUInt32(_ value: UInt32) {
+        var littleEndian = value.littleEndian
+        Swift.withUnsafeBytes(of: &littleEndian) { append(contentsOf: $0) }
+    }
+
+    fileprivate mutating func appendLEInt16(_ value: Int16) {
+        var littleEndian = value.littleEndian
+        Swift.withUnsafeBytes(of: &littleEndian) { append(contentsOf: $0) }
+    }
+}
+
+// swiftformat:enable wrap wrapMultilineStatementBraces trailingCommas redundantSelf extensionAccessControl

apps/macos/Sources/OpenClaw/TalkModeGatewayConfig.swift

@@ -44,7 +44,13 @@ enum TalkModeGatewayConfigParser {
                 acc[key] = value
             } ?? [:]
         let model = activeConfig?["modelId"]?.stringValue?.trimmingCharacters(in: .whitespacesAndNewlines)
-        let resolvedModel = (model?.isEmpty == false) ? model! : defaultModelIdFallback
+        let resolvedModel: String? = if model?.isEmpty == false {
+            model!
+        } else if activeProvider == defaultProvider {
+            defaultModelIdFallback
+        } else {
+            nil
+        }
         let outputFormat = activeConfig?["outputFormat"]?.stringValue
         let interrupt = talk?["interruptOnSpeech"]?.boolValue
         let apiKey = activeConfig?["apiKey"]?.stringValue

apps/macos/Sources/OpenClaw/TalkModeRuntime.swift

@@ -10,6 +10,7 @@ actor TalkModeRuntime {
 
     enum PlaybackPlan: Equatable {
         case elevenLabsThenSystemVoice(apiKey: String, voiceId: String)
+        case mlxThenSystemVoice
         case systemVoiceOnly
     }
 
@@ -17,6 +18,8 @@ actor TalkModeRuntime {
     private let ttsLogger = Logger(subsystem: "ai.openclaw", category: "talk.tts")
     private static let defaultModelIdFallback = "eleven_v3"
     private static let defaultTalkProvider = "elevenlabs"
+    private static let mlxTalkProvider = "mlx"
+    private static let systemTalkProvider = "system"
     private static let defaultSilenceTimeoutMs = TalkDefaults.silenceTimeoutMs
 
     private final class RMSMeter: @unchecked Sendable {
@@ -65,6 +68,7 @@ actor TalkModeRuntime {
     private var modelOverrideActive = false
     private var defaultOutputFormat: String?
     private var interruptOnSpeech: Bool = true
+    private var activeTalkProvider = TalkModeRuntime.defaultTalkProvider
     private var lastInterruptedAtSeconds: Double?
     private var voiceAliases: [String: String] = [:]
     private var lastSpokenText: String?
@@ -462,7 +466,7 @@ actor TalkModeRuntime {
     private func playAssistant(text: String) async {
         guard let input = await self.preparePlaybackInput(text: text) else { return }
 
-        switch Self.playbackPlan(apiKey: input.apiKey, voiceId: input.voiceId) {
+        switch Self.playbackPlan(provider: input.provider, apiKey: input.apiKey, voiceId: input.voiceId) {
         case let .elevenLabsThenSystemVoice(apiKey, voiceId):
             do {
                 try await self.playElevenLabs(input: input, apiKey: apiKey, voiceId: voiceId)
@@ -477,6 +481,23 @@ actor TalkModeRuntime {
                     self.ttsLogger.error("talk system voice failed: \(error.localizedDescription, privacy: .public)")
                 }
             }
+        case .mlxThenSystemVoice:
+            do {
+                try await self.playMLX(input: input)
+            } catch TalkMLXSpeechSynthesizer.SynthesizeError.canceled {
+                self.ttsLogger.info("talk mlx canceled")
+                return
+            } catch {
+                self.ttsLogger
+                    .error(
+                        "talk MLX failed: \(error.localizedDescription, privacy: .public); " +
+                            "falling back to system voice")
+                do {
+                    try await self.playSystemVoice(input: input)
+                } catch {
+                    self.ttsLogger.error("talk system voice failed: \(error.localizedDescription, privacy: .public)")
+                }
+            }
         case .systemVoiceOnly:
             do {
                 try await self.playSystemVoice(input: input)
@@ -491,19 +512,30 @@ actor TalkModeRuntime {
         }
     }
 
-    static func playbackPlan(apiKey: String?, voiceId: String?) -> PlaybackPlan {
-        guard let apiKey, !apiKey.isEmpty, let voiceId else {
+    static func playbackPlan(provider: String, apiKey: String?, voiceId: String?) -> PlaybackPlan {
+        switch provider {
+        case self.defaultTalkProvider:
+            guard let apiKey, !apiKey.isEmpty, let voiceId else {
+                return .systemVoiceOnly
+            }
+            return .elevenLabsThenSystemVoice(apiKey: apiKey, voiceId: voiceId)
+        case self.mlxTalkProvider:
+            return .mlxThenSystemVoice
+        case self.systemTalkProvider:
+            return .systemVoiceOnly
+        default:
             return .systemVoiceOnly
         }
-        return .elevenLabsThenSystemVoice(apiKey: apiKey, voiceId: voiceId)
     }
 
     private struct TalkPlaybackInput {
         let generation: Int
+        let provider: String
         let cleanedText: String
         let directive: TalkDirective?
         let apiKey: String?
         let voiceId: String?
+        let voicePreset: String?
         let language: String?
         let synthTimeoutSeconds: Double
     }
@@ -552,18 +584,20 @@ actor TalkModeRuntime {
             resolvedVoice ??
             self.currentVoiceId ??
             self.defaultVoiceId
+        let voicePreset = preferredVoice
+        let provider = self.activeTalkProvider
 
         let language = ElevenLabsTTSClient.validatedLanguage(directive?.language)
 
-        let voiceId: String? = if let apiKey, !apiKey.isEmpty {
+        let voiceId: String? = if provider == Self.defaultTalkProvider, let apiKey, !apiKey.isEmpty {
             await self.resolveVoiceId(preferred: preferredVoice, apiKey: apiKey)
         } else {
             nil
         }
 
-        if apiKey?.isEmpty != false {
+        if provider == Self.defaultTalkProvider, apiKey?.isEmpty != false {
             self.ttsLogger.warning("talk missing ELEVENLABS_API_KEY; falling back to system voice")
-        } else if voiceId == nil {
+        } else if provider == Self.defaultTalkProvider, voiceId == nil {
             self.ttsLogger.warning("talk missing voiceId; falling back to system voice")
         } else if let voiceId {
             self.ttsLogger
@@ -579,15 +613,21 @@ actor TalkModeRuntime {
 
         return TalkPlaybackInput(
             generation: gen,
+            provider: provider,
             cleanedText: cleaned,
             directive: directive,
             apiKey: apiKey,
             voiceId: voiceId,
+            voicePreset: voicePreset,
             language: language,
             synthTimeoutSeconds: synthTimeoutSeconds)
     }
 
-    private func playElevenLabs(input: TalkPlaybackInput, apiKey: String, voiceId: String) async throws {
+    private func playElevenLabs(
+        input: TalkPlaybackInput,
+        apiKey: String,
+        voiceId: String) async throws
+    {
         let desiredOutputFormat = input.directive?.outputFormat ?? self.defaultOutputFormat ?? "pcm_44100"
         let outputFormat = ElevenLabsTTSClient.validatedOutputFormat(desiredOutputFormat)
         if outputFormat == nil, !desiredOutputFormat.isEmpty {
@@ -696,6 +736,39 @@ actor TalkModeRuntime {
         self.ttsLogger.info("talk system voice done")
     }
 
+    private func playMLX(input: TalkPlaybackInput) async throws {
+        self.ttsLogger.info("talk mlx start chars=\(input.cleanedText.count, privacy: .public)")
+        if self.interruptOnSpeech {
+            guard await self.prepareForPlayback(generation: input.generation) else { return }
+        }
+        await MainActor.run { TalkModeController.shared.updatePhase(.speaking) }
+        self.phase = .speaking
+        let modelRepo = input.directive?.modelId ?? self.currentModelId
+        let audioData: Data
+        do {
+            audioData = try await AsyncTimeout.withTimeout(
+                seconds: input.synthTimeoutSeconds,
+                onTimeout: {
+                    TalkMLXSpeechSynthesizer.SynthesizeError.timedOut
+                },
+                operation: { [self] in
+                    try await self.synthesizeMLXVoice(
+                        text: input.cleanedText,
+                        modelRepo: modelRepo,
+                        language: input.language,
+                        voicePreset: input.voicePreset)
+                })
+        } catch TalkMLXSpeechSynthesizer.SynthesizeError.timedOut {
+            self.stopMLXVoice()
+            throw TalkMLXSpeechSynthesizer.SynthesizeError.timedOut
+        }
+        let result = await self.playTalkAudio(data: audioData)
+        if !result.finished, result.interruptedAt == nil {
+            throw TalkMLXSpeechSynthesizer.SynthesizeError.audioPlaybackFailed
+        }
+        self.ttsLogger.info("talk mlx done")
+    }
+
     private func prepareForPlayback(generation: Int) async -> Bool {
         await self.startRecognition()
         return self.isCurrent(generation)
@@ -750,10 +823,13 @@ actor TalkModeRuntime {
 
     func stopSpeaking(reason: TalkStopReason) async {
         let usePCM = self.lastPlaybackWasPCM
-        let interruptedAt = usePCM ? await self.stopPCM() : await self.stopMP3()
+        let remoteInterruptedAt = usePCM ? await self.stopPCM() : await self.stopMP3()
         _ = usePCM ? await self.stopMP3() : await self.stopPCM()
+        let localInterruptedAt = await self.stopTalkAudio()
         await TalkSystemSpeechSynthesizer.shared.stop()
+        self.stopMLXVoice()
         guard self.phase == .speaking else { return }
+        let interruptedAt = remoteInterruptedAt ?? localInterruptedAt
         if reason == .speech, let interruptedAt {
             self.lastInterruptedAtSeconds = interruptedAt
         }
@@ -795,6 +871,33 @@ extension TalkModeRuntime {
         StreamingAudioPlayer.shared.stop()
     }
 
+    @MainActor
+    private func playTalkAudio(data: Data) async -> TalkPlaybackResult {
+        await TalkAudioPlayer.shared.play(data: data)
+    }
+
+    @MainActor
+    private func stopTalkAudio() -> Double? {
+        TalkAudioPlayer.shared.stop()
+    }
+
+    private func synthesizeMLXVoice(
+        text: String,
+        modelRepo: String?,
+        language: String?,
+        voicePreset: String?) async throws -> Data
+    {
+        try await TalkMLXSpeechSynthesizer.shared.synthesize(
+            text: text,
+            modelRepo: modelRepo,
+            language: language,
+            voicePreset: voicePreset)
+    }
+
+    private func stopMLXVoice() {
+        TalkMLXSpeechSynthesizer.shared.stop()
+    }
+
     // MARK: - Config
 
     private func reloadConfig() async {
@@ -810,6 +913,7 @@ extension TalkModeRuntime {
         }
         self.defaultOutputFormat = cfg.outputFormat
         self.interruptOnSpeech = cfg.interruptOnSpeech
+        self.activeTalkProvider = cfg.activeProvider
         self.silenceWindow = TimeInterval(cfg.silenceTimeoutMs) / 1000
         self.apiKey = cfg.apiKey
         let hasApiKey = (cfg.apiKey?.isEmpty == false)
@@ -817,7 +921,8 @@ extension TalkModeRuntime {
         let modelLabel = (cfg.modelId?.isEmpty == false) ? cfg.modelId! : "none"
         self.logger
             .info(
-                "talk config voiceId=\(voiceLabel, privacy: .public) " +
+                "talk config provider=\(cfg.activeProvider, privacy: .public) " +
+                    "talk config voiceId=\(voiceLabel, privacy: .public) " +
                     "modelId=\(modelLabel, privacy: .public) " +
                     "apiKey=\(hasApiKey, privacy: .public) " +
                     "interrupt=\(cfg.interruptOnSpeech, privacy: .public) " +
@@ -859,11 +964,17 @@ extension TalkModeRuntime {
             await MainActor.run {
                 AppStateStore.shared.seamColorHex = parsed.seamColorHex
             }
-            if parsed.activeProvider != Self.defaultTalkProvider {
-                self.ttsLogger
-                    .info("talk provider \(parsed.activeProvider, privacy: .public) unsupported; using system voice")
-            } else if parsed.normalizedPayload {
+            if parsed.activeProvider == Self.defaultTalkProvider {
                 self.ttsLogger.info("talk config provider from talk.resolved")
+            } else if parsed.activeProvider == Self.mlxTalkProvider ||
+                parsed.activeProvider == Self.systemTalkProvider
+            {
+                self.ttsLogger.info(
+                    "talk provider \(parsed.activeProvider, privacy: .public) active")
+            } else {
+                self.ttsLogger
+                    .info(
+                        "talk provider \(parsed.activeProvider, privacy: .public) unsupported; using system voice")
             }
             return parsed
         } catch {

apps/macos/Sources/OpenClawProtocol/GatewayModels.swift

@@ -1893,6 +1893,7 @@ public struct ConfigApplyParams: Codable, Sendable {
     public let raw: String
     public let basehash: String?
     public let sessionkey: String?
+    public let deliverycontext: [String: AnyCodable]?
     public let note: String?
     public let restartdelayms: Int?
 
@@ -1900,12 +1901,14 @@ public struct ConfigApplyParams: Codable, Sendable {
         raw: String,
         basehash: String?,
         sessionkey: String?,
+        deliverycontext: [String: AnyCodable]?,
         note: String?,
         restartdelayms: Int?)
     {
         self.raw = raw
         self.basehash = basehash
         self.sessionkey = sessionkey
+        self.deliverycontext = deliverycontext
         self.note = note
         self.restartdelayms = restartdelayms
     }
@@ -1914,6 +1917,7 @@ public struct ConfigApplyParams: Codable, Sendable {
         case raw
         case basehash = "baseHash"
         case sessionkey = "sessionKey"
+        case deliverycontext = "deliveryContext"
         case note
         case restartdelayms = "restartDelayMs"
     }
@@ -1923,6 +1927,7 @@ public struct ConfigPatchParams: Codable, Sendable {
     public let raw: String
     public let basehash: String?
     public let sessionkey: String?
+    public let deliverycontext: [String: AnyCodable]?
     public let note: String?
     public let restartdelayms: Int?
 
@@ -1930,12 +1935,14 @@ public struct ConfigPatchParams: Codable, Sendable {
         raw: String,
         basehash: String?,
         sessionkey: String?,
+        deliverycontext: [String: AnyCodable]?,
         note: String?,
         restartdelayms: Int?)
     {
         self.raw = raw
         self.basehash = basehash
         self.sessionkey = sessionkey
+        self.deliverycontext = deliverycontext
         self.note = note
         self.restartdelayms = restartdelayms
     }
@@ -1944,6 +1951,7 @@ public struct ConfigPatchParams: Codable, Sendable {
         case raw
         case basehash = "baseHash"
         case sessionkey = "sessionKey"
+        case deliverycontext = "deliveryContext"
         case note
         case restartdelayms = "restartDelayMs"
     }
@@ -2510,17 +2518,20 @@ public struct AgentSummary: Codable, Sendable {
 public struct AgentsCreateParams: Codable, Sendable {
     public let name: String
     public let workspace: String
+    public let model: String?
     public let emoji: String?
     public let avatar: String?
 
     public init(
         name: String,
         workspace: String,
+        model: String?,
         emoji: String?,
         avatar: String?)
     {
         self.name = name
         self.workspace = workspace
+        self.model = model
         self.emoji = emoji
         self.avatar = avatar
     }
@@ -2528,6 +2539,7 @@ public struct AgentsCreateParams: Codable, Sendable {
     private enum CodingKeys: String, CodingKey {
         case name
         case workspace
+        case model
         case emoji
         case avatar
     }
@@ -2538,17 +2550,20 @@ public struct AgentsCreateResult: Codable, Sendable {
     public let agentid: String
     public let name: String
     public let workspace: String
+    public let model: String?
 
     public init(
         ok: Bool,
         agentid: String,
         name: String,
-        workspace: String)
+        workspace: String,
+        model: String?)
     {
         self.ok = ok
         self.agentid = agentid
         self.name = name
         self.workspace = workspace
+        self.model = model
     }
 
     private enum CodingKeys: String, CodingKey {
@@ -2556,6 +2571,7 @@ public struct AgentsCreateResult: Codable, Sendable {
         case agentid = "agentId"
         case name
         case workspace
+        case model
     }
 }
 
@@ -2564,6 +2580,7 @@ public struct AgentsUpdateParams: Codable, Sendable {
     public let name: String?
     public let workspace: String?
     public let model: String?
+    public let emoji: String?
     public let avatar: String?
 
     public init(
@@ -2571,12 +2588,14 @@ public struct AgentsUpdateParams: Codable, Sendable {
         name: String?,
         workspace: String?,
         model: String?,
+        emoji: String?,
         avatar: String?)
     {
         self.agentid = agentid
         self.name = name
         self.workspace = workspace
         self.model = model
+        self.emoji = emoji
         self.avatar = avatar
     }
 
@@ -2585,6 +2604,7 @@ public struct AgentsUpdateParams: Codable, Sendable {
         case name
         case workspace
         case model
+        case emoji
         case avatar
     }
 }
@@ -2837,6 +2857,7 @@ public struct ModelChoice: Codable, Sendable {
     public let id: String
     public let name: String
     public let provider: String
+    public let alias: String?
     public let contextwindow: Int?
     public let reasoning: Bool?
 
@@ -2844,12 +2865,14 @@ public struct ModelChoice: Codable, Sendable {
         id: String,
         name: String,
         provider: String,
+        alias: String?,
         contextwindow: Int?,
         reasoning: Bool?)
     {
         self.id = id
         self.name = name
         self.provider = provider
+        self.alias = alias
         self.contextwindow = contextwindow
         self.reasoning = reasoning
     }
@@ -2858,6 +2881,7 @@ public struct ModelChoice: Codable, Sendable {
         case id
         case name
         case provider
+        case alias
         case contextwindow = "contextWindow"
         case reasoning
     }
@@ -2879,6 +2903,92 @@ public struct ModelsListResult: Codable, Sendable {
     }
 }
 
+public struct CommandEntry: Codable, Sendable {
+    public let name: String
+    public let nativename: String?
+    public let textaliases: [String]?
+    public let description: String
+    public let category: AnyCodable?
+    public let source: AnyCodable
+    public let scope: AnyCodable
+    public let acceptsargs: Bool
+    public let args: [[String: AnyCodable]]?
+
+    public init(
+        name: String,
+        nativename: String?,
+        textaliases: [String]?,
+        description: String,
+        category: AnyCodable?,
+        source: AnyCodable,
+        scope: AnyCodable,
+        acceptsargs: Bool,
+        args: [[String: AnyCodable]]?)
+    {
+        self.name = name
+        self.nativename = nativename
+        self.textaliases = textaliases
+        self.description = description
+        self.category = category
+        self.source = source
+        self.scope = scope
+        self.acceptsargs = acceptsargs
+        self.args = args
+    }
+
+    private enum CodingKeys: String, CodingKey {
+        case name
+        case nativename = "nativeName"
+        case textaliases = "textAliases"
+        case description
+        case category
+        case source
+        case scope
+        case acceptsargs = "acceptsArgs"
+        case args
+    }
+}
+
+public struct CommandsListParams: Codable, Sendable {
+    public let agentid: String?
+    public let provider: String?
+    public let scope: AnyCodable?
+    public let includeargs: Bool?
+
+    public init(
+        agentid: String?,
+        provider: String?,
+        scope: AnyCodable?,
+        includeargs: Bool?)
+    {
+        self.agentid = agentid
+        self.provider = provider
+        self.scope = scope
+        self.includeargs = includeargs
+    }
+
+    private enum CodingKeys: String, CodingKey {
+        case agentid = "agentId"
+        case provider
+        case scope
+        case includeargs = "includeArgs"
+    }
+}
+
+public struct CommandsListResult: Codable, Sendable {
+    public let commands: [CommandEntry]
+
+    public init(
+        commands: [CommandEntry])
+    {
+        self.commands = commands
+    }
+
+    private enum CodingKeys: String, CodingKey {
+        case commands
+    }
+}
+
 public struct SkillsStatusParams: Codable, Sendable {
     public let agentid: String?
 
@@ -4170,6 +4280,7 @@ public struct ChatEvent: Codable, Sendable {
     public let state: AnyCodable
     public let message: AnyCodable?
     public let errormessage: String?
+    public let errorkind: AnyCodable?
     public let usage: AnyCodable?
     public let stopreason: String?
 
@@ -4180,6 +4291,7 @@ public struct ChatEvent: Codable, Sendable {
         state: AnyCodable,
         message: AnyCodable?,
         errormessage: String?,
+        errorkind: AnyCodable?,
         usage: AnyCodable?,
         stopreason: String?)
     {
@@ -4189,6 +4301,7 @@ public struct ChatEvent: Codable, Sendable {
         self.state = state
         self.message = message
         self.errormessage = errormessage
+        self.errorkind = errorkind
         self.usage = usage
         self.stopreason = stopreason
     }
@@ -4200,6 +4313,7 @@ public struct ChatEvent: Codable, Sendable {
         case state
         case message
         case errormessage = "errorMessage"
+        case errorkind = "errorKind"
         case usage
         case stopreason = "stopReason"
     }
@@ -4207,17 +4321,20 @@ public struct ChatEvent: Codable, Sendable {
 
 public struct UpdateRunParams: Codable, Sendable {
     public let sessionkey: String?
+    public let deliverycontext: [String: AnyCodable]?
     public let note: String?
     public let restartdelayms: Int?
     public let timeoutms: Int?
 
     public init(
         sessionkey: String?,
+        deliverycontext: [String: AnyCodable]?,
         note: String?,
         restartdelayms: Int?,
         timeoutms: Int?)
     {
         self.sessionkey = sessionkey
+        self.deliverycontext = deliverycontext
         self.note = note
         self.restartdelayms = restartdelayms
         self.timeoutms = timeoutms
@@ -4225,6 +4342,7 @@ public struct UpdateRunParams: Codable, Sendable {
 
     private enum CodingKeys: String, CodingKey {
         case sessionkey = "sessionKey"
+        case deliverycontext = "deliveryContext"
         case note
         case restartdelayms = "restartDelayMs"
         case timeoutms = "timeoutMs"

apps/macos/Tests/OpenClawIPCTests/CommandResolverTests.swift

@@ -17,7 +17,6 @@ import Testing
 
     private func makeProjectRootWithPnpm() throws -> (tmp: URL, pnpmPath: URL) {
         let tmp = try makeTempDirForTests()
-        CommandResolver.setProjectRoot(tmp.path)
         let pnpmPath = tmp.appendingPathComponent("node_modules/.bin/pnpm")
         try makeExecutableForTests(at: pnpmPath)
         return (tmp, pnpmPath)
@@ -27,12 +26,17 @@ import Testing
         let defaults = self.makeLocalDefaults()
 
         let tmp = try makeTempDirForTests()
-        CommandResolver.setProjectRoot(tmp.path)
 
         let openclawPath = tmp.appendingPathComponent("node_modules/.bin/openclaw")
         try makeExecutableForTests(at: openclawPath)
 
-        let cmd = CommandResolver.openclawCommand(subcommand: "gateway", defaults: defaults, configRoot: [:])
+        let searchPaths = [tmp.appendingPathComponent("node_modules/.bin").path]
+        let cmd = CommandResolver.openclawCommand(
+            subcommand: "gateway",
+            defaults: defaults,
+            configRoot: [:],
+            searchPaths: searchPaths,
+            projectRoot: tmp)
         #expect(cmd.prefix(2).elementsEqual([openclawPath.path, "gateway"]))
     }
 
@@ -40,7 +44,6 @@ import Testing
         let defaults = self.makeLocalDefaults()
 
         let tmp = try makeTempDirForTests()
-        CommandResolver.setProjectRoot(tmp.path)
 
         let nodePath = tmp.appendingPathComponent("node_modules/.bin/node")
         let scriptPath = tmp.appendingPathComponent("bin/openclaw.js")
@@ -53,7 +56,8 @@ import Testing
             subcommand: "rpc",
             defaults: defaults,
             configRoot: [:],
-            searchPaths: [tmp.appendingPathComponent("node_modules/.bin").path])
+            searchPaths: [tmp.appendingPathComponent("node_modules/.bin").path],
+            projectRoot: tmp)
 
         #expect(cmd.count >= 3)
         if cmd.count >= 3 {
@@ -67,7 +71,6 @@ import Testing
         let defaults = self.makeLocalDefaults()
 
         let tmp = try makeTempDirForTests()
-        CommandResolver.setProjectRoot(tmp.path)
 
         let binDir = tmp.appendingPathComponent("bin")
         let openclawPath = binDir.appendingPathComponent("openclaw")
@@ -79,7 +82,8 @@ import Testing
             subcommand: "rpc",
             defaults: defaults,
             configRoot: [:],
-            searchPaths: [binDir.path])
+            searchPaths: [binDir.path],
+            projectRoot: tmp)
 
         #expect(cmd.prefix(2).elementsEqual([openclawPath.path, "rpc"]))
     }
@@ -88,7 +92,6 @@ import Testing
         let defaults = self.makeLocalDefaults()
 
         let tmp = try makeTempDirForTests()
-        CommandResolver.setProjectRoot(tmp.path)
 
         let binDir = tmp.appendingPathComponent("bin")
         let openclawPath = binDir.appendingPathComponent("openclaw")
@@ -98,7 +101,8 @@ import Testing
             subcommand: "gateway",
             defaults: defaults,
             configRoot: [:],
-            searchPaths: [binDir.path])
+            searchPaths: [binDir.path],
+            projectRoot: tmp)
 
         #expect(cmd.prefix(2).elementsEqual([openclawPath.path, "gateway"]))
     }
@@ -133,9 +137,11 @@ import Testing
 
     @Test func `preferred paths start with project node bins`() throws {
         let tmp = try makeTempDirForTests()
-        CommandResolver.setProjectRoot(tmp.path)
 
-        let first = CommandResolver.preferredPaths().first
+        let first = CommandResolver.preferredPaths(
+            home: FileManager().homeDirectoryForCurrentUser,
+            current: [],
+            projectRoot: tmp).first
         #expect(first == tmp.appendingPathComponent("node_modules/.bin").path)
     }
 
@@ -182,7 +188,6 @@ import Testing
         defaults.set("openclaw@example.com:2222", forKey: remoteTargetKey)
 
         let tmp = try makeTempDirForTests()
-        CommandResolver.setProjectRoot(tmp.path)
 
         let openclawPath = tmp.appendingPathComponent("node_modules/.bin/openclaw")
         try makeExecutableForTests(at: openclawPath)
@@ -190,7 +195,9 @@ import Testing
         let cmd = CommandResolver.openclawCommand(
             subcommand: "daemon",
             defaults: defaults,
-            configRoot: ["gateway": ["mode": "local"]])
+            configRoot: ["gateway": ["mode": "local"]],
+            searchPaths: [tmp.appendingPathComponent("node_modules/.bin").path],
+            projectRoot: tmp)
 
         #expect(cmd.first == openclawPath.path)
         #expect(cmd.count >= 2)

apps/macos/Tests/OpenClawIPCTests/TalkModeGatewayConfigTests.swift

@@ -0,0 +1,48 @@
+import OpenClawProtocol
+import Testing
+@testable import OpenClaw
+
+struct TalkModeGatewayConfigTests {
+    @Test func `mlx provider does not inherit elevenlabs defaults`() {
+        let snapshot = ConfigSnapshot(
+            path: nil,
+            exists: true,
+            raw: nil,
+            hash: nil,
+            parsed: nil,
+            valid: true,
+            config: [
+                "talk": AnyCodable([
+                    "provider": "mlx",
+                    "providers": [
+                        "mlx": [
+                            "voiceId": "unused-voice",
+                        ],
+                    ],
+                    "resolved": [
+                        "provider": "mlx",
+                        "config": [
+                            "voiceId": "unused-voice",
+                        ],
+                    ],
+                ]),
+            ],
+            issues: nil
+        )
+
+        let parsed = TalkModeGatewayConfigParser.parse(
+            snapshot: snapshot,
+            defaultProvider: "elevenlabs",
+            defaultModelIdFallback: "eleven_v3",
+            defaultSilenceTimeoutMs: TalkDefaults.silenceTimeoutMs,
+            envVoice: "env-voice",
+            sagVoice: "sag-voice",
+            envApiKey: "env-key"
+        )
+
+        #expect(parsed.activeProvider == "mlx")
+        #expect(parsed.modelId == nil)
+        #expect(parsed.apiKey == nil)
+        #expect(parsed.voiceId == "unused-voice")
+    }
+}

apps/macos/Tests/OpenClawIPCTests/TalkModeRuntimeSpeechTests.swift

@@ -13,11 +13,34 @@ struct TalkModeRuntimeSpeechTests {
     }
 
     @Test func `playback plan falls back only from elevenlabs`() {
-        #expect(
-            TalkModeRuntime.playbackPlan(apiKey: "key", voiceId: "voice")
-                == .elevenLabsThenSystemVoice(apiKey: "key", voiceId: "voice"))
-        #expect(TalkModeRuntime.playbackPlan(apiKey: nil, voiceId: "voice") == .systemVoiceOnly)
-        #expect(TalkModeRuntime.playbackPlan(apiKey: "key", voiceId: nil) == .systemVoiceOnly)
-        #expect(TalkModeRuntime.playbackPlan(apiKey: "", voiceId: "voice") == .systemVoiceOnly)
+        let elevenLabsPlan = TalkModeRuntime.playbackPlan(
+            provider: "elevenlabs",
+            apiKey: "key",
+            voiceId: "voice"
+        )
+        let missingKeyPlan = TalkModeRuntime.playbackPlan(
+            provider: "elevenlabs",
+            apiKey: nil,
+            voiceId: "voice"
+        )
+        let missingVoicePlan = TalkModeRuntime.playbackPlan(
+            provider: "elevenlabs",
+            apiKey: "key",
+            voiceId: nil
+        )
+        let blankKeyPlan = TalkModeRuntime.playbackPlan(
+            provider: "elevenlabs",
+            apiKey: "",
+            voiceId: "voice"
+        )
+        let mlxPlan = TalkModeRuntime.playbackPlan(provider: "mlx", apiKey: nil, voiceId: nil)
+        let systemPlan = TalkModeRuntime.playbackPlan(provider: "system", apiKey: nil, voiceId: nil)
+
+        #expect(elevenLabsPlan == .elevenLabsThenSystemVoice(apiKey: "key", voiceId: "voice"))
+        #expect(missingKeyPlan == .systemVoiceOnly)
+        #expect(missingVoicePlan == .systemVoiceOnly)
+        #expect(blankKeyPlan == .systemVoiceOnly)
+        #expect(mlxPlan == .mlxThenSystemVoice)
+        #expect(systemPlan == .systemVoiceOnly)
     }
 }

apps/shared/OpenClawKit/Sources/OpenClawProtocol/GatewayModels.swift

@@ -1893,6 +1893,7 @@ public struct ConfigApplyParams: Codable, Sendable {
     public let raw: String
     public let basehash: String?
     public let sessionkey: String?
+    public let deliverycontext: [String: AnyCodable]?
     public let note: String?
     public let restartdelayms: Int?
 
@@ -1900,12 +1901,14 @@ public struct ConfigApplyParams: Codable, Sendable {
         raw: String,
         basehash: String?,
         sessionkey: String?,
+        deliverycontext: [String: AnyCodable]?,
         note: String?,
         restartdelayms: Int?)
     {
         self.raw = raw
         self.basehash = basehash
         self.sessionkey = sessionkey
+        self.deliverycontext = deliverycontext
         self.note = note
         self.restartdelayms = restartdelayms
     }
@@ -1914,6 +1917,7 @@ public struct ConfigApplyParams: Codable, Sendable {
         case raw
         case basehash = "baseHash"
         case sessionkey = "sessionKey"
+        case deliverycontext = "deliveryContext"
         case note
         case restartdelayms = "restartDelayMs"
     }
@@ -1923,6 +1927,7 @@ public struct ConfigPatchParams: Codable, Sendable {
     public let raw: String
     public let basehash: String?
     public let sessionkey: String?
+    public let deliverycontext: [String: AnyCodable]?
     public let note: String?
     public let restartdelayms: Int?
 
@@ -1930,12 +1935,14 @@ public struct ConfigPatchParams: Codable, Sendable {
         raw: String,
         basehash: String?,
         sessionkey: String?,
+        deliverycontext: [String: AnyCodable]?,
         note: String?,
         restartdelayms: Int?)
     {
         self.raw = raw
         self.basehash = basehash
         self.sessionkey = sessionkey
+        self.deliverycontext = deliverycontext
         self.note = note
         self.restartdelayms = restartdelayms
     }
@@ -1944,6 +1951,7 @@ public struct ConfigPatchParams: Codable, Sendable {
         case raw
         case basehash = "baseHash"
         case sessionkey = "sessionKey"
+        case deliverycontext = "deliveryContext"
         case note
         case restartdelayms = "restartDelayMs"
     }
@@ -2510,17 +2518,20 @@ public struct AgentSummary: Codable, Sendable {
 public struct AgentsCreateParams: Codable, Sendable {
     public let name: String
     public let workspace: String
+    public let model: String?
     public let emoji: String?
     public let avatar: String?
 
     public init(
         name: String,
         workspace: String,
+        model: String?,
         emoji: String?,
         avatar: String?)
     {
         self.name = name
         self.workspace = workspace
+        self.model = model
         self.emoji = emoji
         self.avatar = avatar
     }
@@ -2528,6 +2539,7 @@ public struct AgentsCreateParams: Codable, Sendable {
     private enum CodingKeys: String, CodingKey {
         case name
         case workspace
+        case model
         case emoji
         case avatar
     }
@@ -2538,17 +2550,20 @@ public struct AgentsCreateResult: Codable, Sendable {
     public let agentid: String
     public let name: String
     public let workspace: String
+    public let model: String?
 
     public init(
         ok: Bool,
         agentid: String,
         name: String,
-        workspace: String)
+        workspace: String,
+        model: String?)
     {
         self.ok = ok
         self.agentid = agentid
         self.name = name
         self.workspace = workspace
+        self.model = model
     }
 
     private enum CodingKeys: String, CodingKey {
@@ -2556,6 +2571,7 @@ public struct AgentsCreateResult: Codable, Sendable {
         case agentid = "agentId"
         case name
         case workspace
+        case model
     }
 }
 
@@ -2564,6 +2580,7 @@ public struct AgentsUpdateParams: Codable, Sendable {
     public let name: String?
     public let workspace: String?
     public let model: String?
+    public let emoji: String?
     public let avatar: String?
 
     public init(
@@ -2571,12 +2588,14 @@ public struct AgentsUpdateParams: Codable, Sendable {
         name: String?,
         workspace: String?,
         model: String?,
+        emoji: String?,
         avatar: String?)
     {
         self.agentid = agentid
         self.name = name
         self.workspace = workspace
         self.model = model
+        self.emoji = emoji
         self.avatar = avatar
     }
 
@@ -2585,6 +2604,7 @@ public struct AgentsUpdateParams: Codable, Sendable {
         case name
         case workspace
         case model
+        case emoji
         case avatar
     }
 }
@@ -2837,6 +2857,7 @@ public struct ModelChoice: Codable, Sendable {
     public let id: String
     public let name: String
     public let provider: String
+    public let alias: String?
     public let contextwindow: Int?
     public let reasoning: Bool?
 
@@ -2844,12 +2865,14 @@ public struct ModelChoice: Codable, Sendable {
         id: String,
         name: String,
         provider: String,
+        alias: String?,
         contextwindow: Int?,
         reasoning: Bool?)
     {
         self.id = id
         self.name = name
         self.provider = provider
+        self.alias = alias
         self.contextwindow = contextwindow
         self.reasoning = reasoning
     }
@@ -2858,6 +2881,7 @@ public struct ModelChoice: Codable, Sendable {
         case id
         case name
         case provider
+        case alias
         case contextwindow = "contextWindow"
         case reasoning
     }
@@ -2879,6 +2903,92 @@ public struct ModelsListResult: Codable, Sendable {
     }
 }
 
+public struct CommandEntry: Codable, Sendable {
+    public let name: String
+    public let nativename: String?
+    public let textaliases: [String]?
+    public let description: String
+    public let category: AnyCodable?
+    public let source: AnyCodable
+    public let scope: AnyCodable
+    public let acceptsargs: Bool
+    public let args: [[String: AnyCodable]]?
+
+    public init(
+        name: String,
+        nativename: String?,
+        textaliases: [String]?,
+        description: String,
+        category: AnyCodable?,
+        source: AnyCodable,
+        scope: AnyCodable,
+        acceptsargs: Bool,
+        args: [[String: AnyCodable]]?)
+    {
+        self.name = name
+        self.nativename = nativename
+        self.textaliases = textaliases
+        self.description = description
+        self.category = category
+        self.source = source
+        self.scope = scope
+        self.acceptsargs = acceptsargs
+        self.args = args
+    }
+
+    private enum CodingKeys: String, CodingKey {
+        case name
+        case nativename = "nativeName"
+        case textaliases = "textAliases"
+        case description
+        case category
+        case source
+        case scope
+        case acceptsargs = "acceptsArgs"
+        case args
+    }
+}
+
+public struct CommandsListParams: Codable, Sendable {
+    public let agentid: String?
+    public let provider: String?
+    public let scope: AnyCodable?
+    public let includeargs: Bool?
+
+    public init(
+        agentid: String?,
+        provider: String?,
+        scope: AnyCodable?,
+        includeargs: Bool?)
+    {
+        self.agentid = agentid
+        self.provider = provider
+        self.scope = scope
+        self.includeargs = includeargs
+    }
+
+    private enum CodingKeys: String, CodingKey {
+        case agentid = "agentId"
+        case provider
+        case scope
+        case includeargs = "includeArgs"
+    }
+}
+
+public struct CommandsListResult: Codable, Sendable {
+    public let commands: [CommandEntry]
+
+    public init(
+        commands: [CommandEntry])
+    {
+        self.commands = commands
+    }
+
+    private enum CodingKeys: String, CodingKey {
+        case commands
+    }
+}
+
 public struct SkillsStatusParams: Codable, Sendable {
     public let agentid: String?
 
@@ -4170,6 +4280,7 @@ public struct ChatEvent: Codable, Sendable {
     public let state: AnyCodable
     public let message: AnyCodable?
     public let errormessage: String?
+    public let errorkind: AnyCodable?
     public let usage: AnyCodable?
     public let stopreason: String?
 
@@ -4180,6 +4291,7 @@ public struct ChatEvent: Codable, Sendable {
         state: AnyCodable,
         message: AnyCodable?,
         errormessage: String?,
+        errorkind: AnyCodable?,
         usage: AnyCodable?,
         stopreason: String?)
     {
@@ -4189,6 +4301,7 @@ public struct ChatEvent: Codable, Sendable {
         self.state = state
         self.message = message
         self.errormessage = errormessage
+        self.errorkind = errorkind
         self.usage = usage
         self.stopreason = stopreason
     }
@@ -4200,6 +4313,7 @@ public struct ChatEvent: Codable, Sendable {
         case state
         case message
         case errormessage = "errorMessage"
+        case errorkind = "errorKind"
         case usage
         case stopreason = "stopReason"
     }
@@ -4207,17 +4321,20 @@ public struct ChatEvent: Codable, Sendable {
 
 public struct UpdateRunParams: Codable, Sendable {
     public let sessionkey: String?
+    public let deliverycontext: [String: AnyCodable]?
     public let note: String?
     public let restartdelayms: Int?
     public let timeoutms: Int?
 
     public init(
         sessionkey: String?,
+        deliverycontext: [String: AnyCodable]?,
         note: String?,
         restartdelayms: Int?,
         timeoutms: Int?)
     {
         self.sessionkey = sessionkey
+        self.deliverycontext = deliverycontext
         self.note = note
         self.restartdelayms = restartdelayms
         self.timeoutms = timeoutms
@@ -4225,6 +4342,7 @@ public struct UpdateRunParams: Codable, Sendable {
 
     private enum CodingKeys: String, CodingKey {
         case sessionkey = "sessionKey"
+        case deliverycontext = "deliveryContext"
         case note
         case restartdelayms = "restartDelayMs"
         case timeoutms = "timeoutMs"

apps/shared/OpenClawKit/Tools/CanvasA2UI/bootstrap.js

@@ -466,8 +466,10 @@ class OpenClawA2UIHost extends LitElement {
       try {
         // WebKit message handlers support structured objects; Android's JS interface expects strings.
         if (handler === globalThis.openclawCanvasA2UIAction) {
+          // oxlint-disable-next-line unicorn/require-post-message-target-origin -- Native app message handler, not Window.postMessage.
           handler.postMessage(JSON.stringify({ userAction }));
         } else {
+          // oxlint-disable-next-line unicorn/require-post-message-target-origin -- WebKit message handler, not Window.postMessage.
           handler.postMessage({ userAction });
         }
       } catch (e) {

docs/.generated/config-baseline.sha256

@@ -1,4 +1,4 @@
-0a75b57f5dbb0bb1488eacb47111ee22ff42dd3747bfe07bb69c9445d5e55c3e  config-baseline.json
-ff15bb8b4231fc80174249ae89bcb61439d7adda5ee6be95e4d304680253a59f  config-baseline.core.json
-7f42b22b46c487d64aaac46001ba9d9096cf7bf0b1c263a54d39946303ff5018  config-baseline.channel.json
-483d4f3c1d516719870ad6f2aba6779b9950f85471ee77b9994a077a7574a892  config-baseline.plugin.json
+228031f16ad06580bfd137f092d70d03f2796515e723b8b6618ed69d285465fa  config-baseline.json
+bad0a5bb247a62b8fb9ed9fc2b2720eacf3e0913077ac351b5d26ae2723335ad  config-baseline.core.json
+e1f94346a8507ce3dec763b598e79f3bb89ff2e33189ce977cc87d3b05e71c1d  config-baseline.channel.json
+6c19997f1fb2aff4315f2cb9c7d9e299b403fbc0f9e78e3412cc7fe1c655f222  config-baseline.plugin.json

docs/.generated/plugin-sdk-api-baseline.sha256

@@ -1,2 +1,2 @@
-763d2709dd26f4ec7d5807b2f1781b7f58cb115d2b0a9c9235a6c2c7b3788c1f  plugin-sdk-api-baseline.json
-87ab9ec219f037b13a8f42378d1fed02701d4035da0e5eca8a091626e8426523  plugin-sdk-api-baseline.jsonl
+ee16273fa5ad8c5408e9dad8d96fde86dfa666ef8eb44840b78135814ff97173  plugin-sdk-api-baseline.json
+2bd0d5edf23e6a889d6bedb74d0d06411dd7750dac6ebf24971c789f8a69253a  plugin-sdk-api-baseline.jsonl

docs/automation/cron-jobs.md

@@ -43,6 +43,8 @@ together`, and similar hints) and no descendant subagent run is still
   responsible for the final answer, OpenClaw re-prompts once for the actual
   result before delivery.
 
+<a id="maintenance"></a>
+
 Task reconciliation for cron is runtime-owned: an active cron task stays live while the
 cron runtime still tracks that job as running, even if an old child session row still exists.
 Once the runtime stops owning the job and the 5-minute grace window expires, maintenance can

docs/automation/hooks.md

@@ -164,10 +164,14 @@ Enable any bundled hook:
 openclaw hooks enable <hook-name>
 ```
 
+<a id="session-memory"></a>
+
 ### session-memory details
 
 Extracts the last 15 user/assistant messages, generates a descriptive filename slug via LLM, and saves to `<workspace>/memory/YYYY-MM-DD-slug.md`. Requires `workspace.dir` to be configured.
 
+<a id="bootstrap-extra-files"></a>
+
 ### bootstrap-extra-files config
 
 ```json
@@ -187,6 +191,18 @@ Extracts the last 15 user/assistant messages, generates a descriptive filename s
 
 Paths resolve relative to workspace. Only recognized bootstrap basenames are loaded (`AGENTS.md`, `SOUL.md`, `TOOLS.md`, `IDENTITY.md`, `USER.md`, `HEARTBEAT.md`, `BOOTSTRAP.md`, `MEMORY.md`).
 
+<a id="command-logger"></a>
+
+### command-logger details
+
+Logs every slash command to `~/.openclaw/logs/commands.log`.
+
+<a id="boot-md"></a>
+
+### boot-md details
+
+Runs `BOOT.md` from the active workspace when the gateway starts.
+
 ## Plugin hooks
 
 Plugins can register hooks through the Plugin SDK for deeper integration: intercepting tool calls, modifying prompts, controlling message flow, and more. The Plugin SDK exposes 28 hooks covering model resolution, agent lifecycle, message flow, tool execution, subagent coordination, and gateway lifecycle.

docs/automation/tasks.md

@@ -180,7 +180,7 @@ The lookup token accepts a task ID, run ID, or session key. Shows the full recor
 openclaw tasks cancel <lookup>
 ```
 
-For ACP and subagent tasks, this kills the child session. Status transitions to `cancelled` and a delivery notification is sent.
+For ACP and subagent tasks, this kills the child session. For CLI-tracked tasks, cancellation is recorded in the task registry (there is no separate child runtime handle). Status transitions to `cancelled` and a delivery notification is sent when applicable.
 
 ### `tasks notify`
 

docs/ci.md

@@ -12,24 +12,25 @@ The CI runs on every push to `main` and every pull request. It uses smart scopin
 
 ## Job Overview
 
-| Job                      | Purpose                                                                                  | When it runs                        |
-| ------------------------ | ---------------------------------------------------------------------------------------- | ----------------------------------- |
-| `preflight`              | Detect docs-only changes, changed scopes, changed extensions, and build the CI manifest  | Always on non-draft pushes and PRs  |
-| `security-fast`          | Private key detection, workflow audit via `zizmor`, production dependency audit          | Always on non-draft pushes and PRs  |
-| `build-artifacts`        | Build `dist/` and the Control UI once, upload reusable artifacts for downstream jobs     | Node-relevant changes               |
-| `checks-fast-core`       | Fast Linux correctness lanes such as bundled/plugin-contract/protocol checks             | Node-relevant changes               |
-| `checks-fast-extensions` | Aggregate the extension shard lanes after `checks-fast-extensions-shard` completes       | Node-relevant changes               |
-| `extension-fast`         | Focused tests for only the changed bundled plugins                                       | When extension changes are detected |
-| `check`                  | Main local gate in CI: `pnpm check` plus `pnpm build:strict-smoke`                       | Node-relevant changes               |
-| `check-additional`       | Architecture and boundary guards plus the gateway watch regression harness               | Node-relevant changes               |
-| `build-smoke`            | Built-CLI smoke tests and startup-memory smoke                                           | Node-relevant changes               |
-| `checks`                 | Heavier Linux Node lanes: full tests, channel tests, and push-only Node 22 compatibility | Node-relevant changes               |
-| `check-docs`             | Docs formatting, lint, and broken-link checks                                            | Docs changed                        |
-| `skills-python`          | Ruff + pytest for Python-backed skills                                                   | Python-skill-relevant changes       |
-| `checks-windows`         | Windows-specific test lanes                                                              | Windows-relevant changes            |
-| `macos-node`             | macOS TypeScript test lane using the shared built artifacts                              | macOS-relevant changes              |
-| `macos-swift`            | Swift lint, build, and tests for the macOS app                                           | macOS-relevant changes              |
-| `android`                | Android build and test matrix                                                            | Android-relevant changes            |
+| Job                      | Purpose                                                                                 | When it runs                        |
+| ------------------------ | --------------------------------------------------------------------------------------- | ----------------------------------- |
+| `preflight`              | Detect docs-only changes, changed scopes, changed extensions, and build the CI manifest | Always on non-draft pushes and PRs  |
+| `security-fast`          | Private key detection, workflow audit via `zizmor`, production dependency audit         | Always on non-draft pushes and PRs  |
+| `build-artifacts`        | Build `dist/` and the Control UI once, upload reusable artifacts for downstream jobs    | Node-relevant changes               |
+| `checks-fast-core`       | Fast Linux correctness lanes such as bundled/plugin-contract/protocol checks            | Node-relevant changes               |
+| `checks-node-extensions` | Full bundled-plugin test shards across the extension suite                              | Node-relevant changes               |
+| `checks-node-core-test`  | Core Node test shards, excluding channel, bundled, contract, and extension lanes        | Node-relevant changes               |
+| `extension-fast`         | Focused tests for only the changed bundled plugins                                      | When extension changes are detected |
+| `check`                  | Main local gate in CI: `pnpm check` plus `pnpm build:strict-smoke`                      | Node-relevant changes               |
+| `check-additional`       | Architecture, boundary, import-cycle guards plus the gateway watch regression harness   | Node-relevant changes               |
+| `build-smoke`            | Built-CLI smoke tests and startup-memory smoke                                          | Node-relevant changes               |
+| `checks`                 | Remaining Linux Node lanes: channel tests and push-only Node 22 compatibility           | Node-relevant changes               |
+| `check-docs`             | Docs formatting, lint, and broken-link checks                                           | Docs changed                        |
+| `skills-python`          | Ruff + pytest for Python-backed skills                                                  | Python-skill-relevant changes       |
+| `checks-windows`         | Windows-specific test lanes                                                             | Windows-relevant changes            |
+| `macos-node`             | macOS TypeScript test lane using the shared built artifacts                             | macOS-relevant changes              |
+| `macos-swift`            | Swift lint, build, and tests for the macOS app                                          | macOS-relevant changes              |
+| `android`                | Android build and test matrix                                                           | Android-relevant changes            |
 
 ## Fail-Fast Order
 
@@ -38,7 +39,7 @@ Jobs are ordered so cheap checks fail before expensive ones run:
 1. `preflight` decides which lanes exist at all. The `docs-scope` and `changed-scope` logic are steps inside this job, not standalone jobs.
 2. `security-fast`, `check`, `check-additional`, `check-docs`, and `skills-python` fail quickly without waiting on the heavier artifact and platform matrix jobs.
 3. `build-artifacts` overlaps with the fast Linux lanes so downstream consumers can start as soon as the shared build is ready.
-4. Heavier platform and runtime lanes fan out after that: `checks-fast-core`, `checks-fast-extensions`, `extension-fast`, `checks`, `checks-windows`, `macos-node`, `macos-swift`, and `android`.
+4. Heavier platform and runtime lanes fan out after that: `checks-fast-core`, `checks-node-extensions`, `checks-node-core-test`, `extension-fast`, `checks`, `checks-windows`, `macos-node`, `macos-swift`, and `android`.
 
 Scope logic lives in `scripts/ci-changed-scope.mjs` and is covered by unit tests in `src/scripts/ci-changed-scope.test.ts`.
 The separate `install-smoke` workflow reuses the same scope script through its own `preflight` job. It computes `run_install_smoke` from the narrower changed-smoke signal, so Docker/install smoke only runs for install, packaging, and container-relevant changes.
@@ -58,6 +59,7 @@ On pushes, the `checks` matrix adds the push-only `compat-node22` lane. On pull
 ```bash
 pnpm check          # types + lint + format
 pnpm build:strict-smoke
+pnpm check:import-cycles
 pnpm test:gateway:watch-regression
 pnpm test           # vitest tests
 pnpm test:channels

docs/cli/agents.md

@@ -37,7 +37,7 @@ Use routing bindings to pin inbound channel traffic to a specific agent.
 If you also want different visible skills per agent, configure
 `agents.defaults.skills` and `agents.list[].skills` in `openclaw.json`. See
 [Skills config](/tools/skills-config) and
-[Configuration Reference](/gateway/configuration-reference#agentsdefaultsskills).
+[Configuration Reference](/gateway/configuration-reference#agents-defaults-skills).
 
 List bindings:
 

docs/cli/approvals.md

@@ -1,5 +1,5 @@
 ---
-summary: "CLI reference for `openclaw approvals` (exec approvals for gateway or node hosts)"
+summary: "CLI reference for `openclaw approvals` and `openclaw exec-policy`"
 read_when:
   - You want to edit exec approvals from the CLI
   - You need to manage allowlists on gateway or node hosts
@@ -18,6 +18,45 @@ Related:
 - Exec approvals: [Exec approvals](/tools/exec-approvals)
 - Nodes: [Nodes](/nodes)
 
+## `openclaw exec-policy`
+
+`openclaw exec-policy` is the local convenience command for keeping the requested
+`tools.exec.*` config and the local host approvals file aligned in one step.
+
+Use it when you want to:
+
+- inspect the local requested policy, host approvals file, and effective merge
+- apply a local preset such as YOLO or deny-all
+- synchronize local `tools.exec.*` and local `~/.openclaw/exec-approvals.json`
+
+Examples:
+
+```bash
+openclaw exec-policy show
+openclaw exec-policy show --json
+
+openclaw exec-policy preset yolo
+openclaw exec-policy preset cautious --json
+
+openclaw exec-policy set --host gateway --security full --ask off --ask-fallback full
+```
+
+Output modes:
+
+- no `--json`: prints the human-readable table view
+- `--json`: prints machine-readable structured output
+
+Current scope:
+
+- `exec-policy` is **local-only**
+- it updates the local config file and the local approvals file together
+- it does **not** push policy to the gateway host or a node host
+- `--host node` is rejected in this command because node exec approvals are fetched from the node at runtime and must be managed through node-targeted approvals commands instead
+- `openclaw exec-policy show` marks `host=node` scopes as node-managed at runtime instead of deriving an effective policy from the local approvals file
+
+If you need to edit remote host approvals directly, keep using `openclaw approvals set --gateway`
+or `openclaw approvals set --node <id|name|ip>`.
+
 ## Common commands
 
 ```bash
@@ -100,6 +139,16 @@ Why `tools.exec.host=gateway` in this example:
 
 This matches the current host-default YOLO behavior. Tighten it if you want approvals.
 
+Local shortcut:
+
+```bash
+openclaw exec-policy preset yolo
+```
+
+That local shortcut updates both the requested local `tools.exec.*` config and the
+local approvals defaults together. It is equivalent in intent to the manual two-step
+setup above, but only for the local machine.
+
 ## Allowlist helpers
 
 ```bash

docs/cli/devices.md

@@ -49,8 +49,10 @@ openclaw devices clear --yes --pending --json
 
 ### `openclaw devices approve [requestId] [--latest]`
 
-Approve a pending device pairing request. If `requestId` is omitted, OpenClaw
-automatically approves the most recent pending request.
+Approve a pending device pairing request by exact `requestId`. If `requestId`
+is omitted or `--latest` is passed, OpenClaw only prints the selected pending
+request and exits; rerun approval with the exact request ID after verifying
+the details.
 
 Note: if a device retries pairing with changed auth details (role/scopes/public
 key), OpenClaw supersedes the previous pending entry and issues a new
@@ -126,7 +128,7 @@ Pass `--token` or `--password` explicitly. Missing explicit credentials is an er
   `operator.admin`.
 - `devices clear` is intentionally gated by `--yes`.
 - If pairing scope is unavailable on local loopback (and no explicit `--url` is passed), list/approve can use a local pairing fallback.
-- `devices approve` picks the newest pending request automatically when you omit `requestId` or pass `--latest`.
+- `devices approve` requires an explicit request ID before minting tokens; omitting `requestId` or passing `--latest` only previews the newest pending request.
 
 ## Token drift recovery checklist
 

docs/cli/index.md

@@ -852,7 +852,7 @@ Subcommands:
 Notes:
 
 - `devices list` and `devices approve` can fall back to local pairing files on local loopback when direct pairing scope is unavailable.
-- `devices approve` auto-selects the newest pending request when no `requestId` is passed or `--latest` is set.
+- `devices approve` requires an explicit request ID before minting tokens; omitting `requestId` or passing `--latest` only previews the newest pending request.
 - Stored-token reconnects reuse the token's cached approved scopes; explicit
   `devices rotate --scope ...` updates that stored scope set for future
   cached-token reconnects.

docs/cli/memory.md

@@ -167,4 +167,8 @@ Notes:
 - If effectively active memory remote API key fields are configured as SecretRefs, the command resolves those values from the active gateway snapshot. If gateway is unavailable, the command fails fast.
 - Gateway version skew note: this command path requires a gateway that supports `secrets.resolve`; older gateways return an unknown-method error.
 - Tune scheduled sweep cadence with `dreaming.frequency`. Deep promotion policy is otherwise internal; use CLI flags on `memory promote` when you need one-off manual overrides.
+- `memory rem-harness --path <file-or-dir> --grounded` previews grounded `What Happened`, `Reflections`, and `Possible Lasting Updates` from historical daily notes without writing anything.
+- `memory rem-backfill --path <file-or-dir>` writes reversible grounded diary entries into `DREAMS.md` for UI review.
+- `memory rem-backfill --path <file-or-dir> --stage-short-term` also seeds grounded durable candidates into the live short-term promotion store so the normal deep phase can rank them.
+- `memory rem-backfill --rollback` removes previously written grounded diary entries, and `memory rem-backfill --rollback-short-term` removes previously staged grounded short-term candidates.
 - See [Dreaming](/concepts/dreaming) for full phase descriptions and configuration reference.

docs/concepts/active-memory.md

@@ -0,0 +1,608 @@
+---
+title: "Active Memory"
+summary: "A plugin-owned blocking memory sub-agent that injects relevant memory into interactive chat sessions"
+read_when:
+  - You want to understand what active memory is for
+  - You want to turn active memory on for a conversational agent
+  - You want to tune active memory behavior without enabling it everywhere
+---
+
+# Active Memory
+
+Active memory is an optional plugin-owned blocking memory sub-agent that runs
+before the main reply for eligible conversational sessions.
+
+It exists because most memory systems are capable but reactive. They rely on
+the main agent to decide when to search memory, or on the user to say things
+like "remember this" or "search memory." By then, the moment where memory would
+have made the reply feel natural has already passed.
+
+Active memory gives the system one bounded chance to surface relevant memory
+before the main reply is generated.
+
+## Paste This Into Your Agent
+
+Paste this into your agent if you want it to enable Active Memory with a
+self-contained, safe-default setup:
+
+```json5
+{
+  plugins: {
+    entries: {
+      "active-memory": {
+        enabled: true,
+        config: {
+          enabled: true,
+          agents: ["main"],
+          allowedChatTypes: ["direct"],
+          modelFallbackPolicy: "default-remote",
+          queryMode: "recent",
+          promptStyle: "balanced",
+          timeoutMs: 15000,
+          maxSummaryChars: 220,
+          persistTranscripts: false,
+          logging: true,
+        },
+      },
+    },
+  },
+}
+```
+
+This turns the plugin on for the `main` agent, keeps it limited to direct-message
+style sessions by default, lets it inherit the current session model first, and
+still allows the built-in remote fallback if no explicit or inherited model is
+available.
+
+After that, restart the gateway:
+
+```bash
+openclaw gateway
+```
+
+To inspect it live in a conversation:
+
+```text
+/verbose on
+```
+
+## Turn active memory on
+
+The safest setup is:
+
+1. enable the plugin
+2. target one conversational agent
+3. keep logging on only while tuning
+
+Start with this in `openclaw.json`:
+
+```json5
+{
+  plugins: {
+    entries: {
+      "active-memory": {
+        enabled: true,
+        config: {
+          agents: ["main"],
+          allowedChatTypes: ["direct"],
+          modelFallbackPolicy: "default-remote",
+          queryMode: "recent",
+          promptStyle: "balanced",
+          timeoutMs: 15000,
+          maxSummaryChars: 220,
+          persistTranscripts: false,
+          logging: true,
+        },
+      },
+    },
+  },
+}
+```
+
+Then restart the gateway:
+
+```bash
+openclaw gateway
+```
+
+What this means:
+
+- `plugins.entries.active-memory.enabled: true` turns the plugin on
+- `config.agents: ["main"]` opts only the `main` agent into active memory
+- `config.allowedChatTypes: ["direct"]` keeps active memory on for direct-message style sessions only by default
+- if `config.model` is unset, active memory inherits the current session model first
+- `config.modelFallbackPolicy: "default-remote"` keeps the built-in remote fallback as the default when no explicit or inherited model is available
+- `config.promptStyle: "balanced"` uses the default general-purpose prompt style for `recent` mode
+- active memory still runs only on eligible interactive persistent chat sessions
+
+## How to see it
+
+Active memory injects hidden system context for the model. It does not expose
+raw `<active_memory_plugin>...</active_memory_plugin>` tags to the client.
+
+## Session toggle
+
+Use the plugin command when you want to pause or resume active memory for the
+current chat session without editing config:
+
+```text
+/active-memory status
+/active-memory off
+/active-memory on
+```
+
+This is session-scoped. It does not change
+`plugins.entries.active-memory.enabled`, agent targeting, or other global
+configuration.
+
+If you want the command to write config and pause or resume active memory for
+all sessions, use the explicit global form:
+
+```text
+/active-memory status --global
+/active-memory off --global
+/active-memory on --global
+```
+
+The global form writes `plugins.entries.active-memory.config.enabled`. It leaves
+`plugins.entries.active-memory.enabled` on so the command remains available to
+turn active memory back on later.
+
+If you want to see what active memory is doing in a live session, turn verbose
+mode on for that session:
+
+```text
+/verbose on
+```
+
+With verbose enabled, OpenClaw can show:
+
+- an active memory status line such as `Active Memory: ok 842ms recent 34 chars`
+- a readable debug summary such as `Active Memory Debug: Lemon pepper wings with blue cheese.`
+
+Those lines are derived from the same active memory pass that feeds the hidden
+system context, but they are formatted for humans instead of exposing raw prompt
+markup.
+
+By default, the blocking memory sub-agent transcript is temporary and deleted
+after the run completes.
+
+Example flow:
+
+```text
+/verbose on
+what wings should i order?
+```
+
+Expected visible reply shape:
+
+```text
+...normal assistant reply...
+
+🧩 Active Memory: ok 842ms recent 34 chars
+🔎 Active Memory Debug: Lemon pepper wings with blue cheese.
+```
+
+## When it runs
+
+Active memory uses two gates:
+
+1. **Config opt-in**
+   The plugin must be enabled, and the current agent id must appear in
+   `plugins.entries.active-memory.config.agents`.
+2. **Strict runtime eligibility**
+   Even when enabled and targeted, active memory only runs for eligible
+   interactive persistent chat sessions.
+
+The actual rule is:
+
+```text
+plugin enabled
++
+agent id targeted
++
+allowed chat type
++
+eligible interactive persistent chat session
+=
+active memory runs
+```
+
+If any of those fail, active memory does not run.
+
+## Session types
+
+`config.allowedChatTypes` controls which kinds of conversations may run Active
+Memory at all.
+
+The default is:
+
+```json5
+allowedChatTypes: ["direct"]
+```
+
+That means Active Memory runs by default in direct-message style sessions, but
+not in group or channel sessions unless you opt them in explicitly.
+
+Examples:
+
+```json5
+allowedChatTypes: ["direct"]
+```
+
+```json5
+allowedChatTypes: ["direct", "group"]
+```
+
+```json5
+allowedChatTypes: ["direct", "group", "channel"]
+```
+
+## Where it runs
+
+Active memory is a conversational enrichment feature, not a platform-wide
+inference feature.
+
+| Surface                                                             | Runs active memory?                                     |
+| ------------------------------------------------------------------- | ------------------------------------------------------- |
+| Control UI / web chat persistent sessions                           | Yes, if the plugin is enabled and the agent is targeted |
+| Other interactive channel sessions on the same persistent chat path | Yes, if the plugin is enabled and the agent is targeted |
+| Headless one-shot runs                                              | No                                                      |
+| Heartbeat/background runs                                           | No                                                      |
+| Generic internal `agent-command` paths                              | No                                                      |
+| Sub-agent/internal helper execution                                 | No                                                      |
+
+## Why use it
+
+Use active memory when:
+
+- the session is persistent and user-facing
+- the agent has meaningful long-term memory to search
+- continuity and personalization matter more than raw prompt determinism
+
+It works especially well for:
+
+- stable preferences
+- recurring habits
+- long-term user context that should surface naturally
+
+It is a poor fit for:
+
+- automation
+- internal workers
+- one-shot API tasks
+- places where hidden personalization would be surprising
+
+## How it works
+
+The runtime shape is:
+
+```mermaid
+flowchart LR
+  U["User Message"] --> Q["Build Memory Query"]
+  Q --> R["Active Memory Blocking Memory Sub-Agent"]
+  R -->|NONE or empty| M["Main Reply"]
+  R -->|relevant summary| I["Append Hidden active_memory_plugin System Context"]
+  I --> M["Main Reply"]
+```
+
+The blocking memory sub-agent can use only:
+
+- `memory_search`
+- `memory_get`
+
+If the connection is weak, it should return `NONE`.
+
+## Query modes
+
+`config.queryMode` controls how much conversation the blocking memory sub-agent sees.
+
+## Prompt styles
+
+`config.promptStyle` controls how eager or strict the blocking memory sub-agent is
+when deciding whether to return memory.
+
+Available styles:
+
+- `balanced`: general-purpose default for `recent` mode
+- `strict`: least eager; best when you want very little bleed from nearby context
+- `contextual`: most continuity-friendly; best when conversation history should matter more
+- `recall-heavy`: more willing to surface memory on softer but still plausible matches
+- `precision-heavy`: aggressively prefers `NONE` unless the match is obvious
+- `preference-only`: optimized for favorites, habits, routines, taste, and recurring personal facts
+
+Default mapping when `config.promptStyle` is unset:
+
+```text
+message -> strict
+recent -> balanced
+full -> contextual
+```
+
+If you set `config.promptStyle` explicitly, that override wins.
+
+Example:
+
+```json5
+promptStyle: "preference-only"
+```
+
+## Model fallback policy
+
+If `config.model` is unset, Active Memory tries to resolve a model in this order:
+
+```text
+explicit plugin model
+-> current session model
+-> agent primary model
+-> optional built-in remote fallback
+```
+
+`config.modelFallbackPolicy` controls the last step.
+
+Default:
+
+```json5
+modelFallbackPolicy: "default-remote"
+```
+
+Other option:
+
+```json5
+modelFallbackPolicy: "resolved-only"
+```
+
+Use `resolved-only` if you want Active Memory to skip recall instead of falling
+back to the built-in remote default when no explicit or inherited model is
+available.
+
+## Advanced escape hatches
+
+These options are intentionally not part of the recommended setup.
+
+`config.thinking` can override the blocking memory sub-agent thinking level:
+
+```json5
+thinking: "medium"
+```
+
+Default:
+
+```json5
+thinking: "off"
+```
+
+Do not enable this by default. Active Memory runs in the reply path, so extra
+thinking time directly increases user-visible latency.
+
+`config.promptAppend` adds extra operator instructions after the default Active
+Memory prompt and before the conversation context:
+
+```json5
+promptAppend: "Prefer stable long-term preferences over one-off events."
+```
+
+`config.promptOverride` replaces the default Active Memory prompt. OpenClaw
+still appends the conversation context afterward:
+
+```json5
+promptOverride: "You are a memory search agent. Return NONE or one compact user fact."
+```
+
+Prompt customization is not recommended unless you are deliberately testing a
+different recall contract. The default prompt is tuned to return either `NONE`
+or compact user-fact context for the main model.
+
+### `message`
+
+Only the latest user message is sent.
+
+```text
+Latest user message only
+```
+
+Use this when:
+
+- you want the fastest behavior
+- you want the strongest bias toward stable preference recall
+- follow-up turns do not need conversational context
+
+Recommended timeout:
+
+- start around `3000` to `5000` ms
+
+### `recent`
+
+The latest user message plus a small recent conversational tail is sent.
+
+```text
+Recent conversation tail:
+user: ...
+assistant: ...
+user: ...
+
+Latest user message:
+...
+```
+
+Use this when:
+
+- you want a better balance of speed and conversational grounding
+- follow-up questions often depend on the last few turns
+
+Recommended timeout:
+
+- start around `15000` ms
+
+### `full`
+
+The full conversation is sent to the blocking memory sub-agent.
+
+```text
+Full conversation context:
+user: ...
+assistant: ...
+user: ...
+...
+```
+
+Use this when:
+
+- the strongest recall quality matters more than latency
+- the conversation contains important setup far back in the thread
+
+Recommended timeout:
+
+- increase it substantially compared with `message` or `recent`
+- start around `15000` ms or higher depending on thread size
+
+In general, timeout should increase with context size:
+
+```text
+message < recent < full
+```
+
+## Transcript persistence
+
+Active memory blocking memory sub-agent runs create a real `session.jsonl`
+transcript during the blocking memory sub-agent call.
+
+By default, that transcript is temporary:
+
+- it is written to a temp directory
+- it is used only for the blocking memory sub-agent run
+- it is deleted immediately after the run finishes
+
+If you want to keep those blocking memory sub-agent transcripts on disk for debugging or
+inspection, turn persistence on explicitly:
+
+```json5
+{
+  plugins: {
+    entries: {
+      "active-memory": {
+        enabled: true,
+        config: {
+          agents: ["main"],
+          persistTranscripts: true,
+          transcriptDir: "active-memory",
+        },
+      },
+    },
+  },
+}
+```
+
+When enabled, active memory stores transcripts in a separate directory under the
+target agent's sessions folder, not in the main user conversation transcript
+path.
+
+The default layout is conceptually:
+
+```text
+agents/<agent>/sessions/active-memory/<blocking-memory-sub-agent-session-id>.jsonl
+```
+
+You can change the relative subdirectory with `config.transcriptDir`.
+
+Use this carefully:
+
+- blocking memory sub-agent transcripts can accumulate quickly on busy sessions
+- `full` query mode can duplicate a lot of conversation context
+- these transcripts contain hidden prompt context and recalled memories
+
+## Configuration
+
+All active memory configuration lives under:
+
+```text
+plugins.entries.active-memory
+```
+
+The most important fields are:
+
+| Key                         | Type                                                                                                 | Meaning                                                                                                |
+| --------------------------- | ---------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------ |
+| `enabled`                   | `boolean`                                                                                            | Enables the plugin itself                                                                              |
+| `config.agents`             | `string[]`                                                                                           | Agent ids that may use active memory                                                                   |
+| `config.model`              | `string`                                                                                             | Optional blocking memory sub-agent model ref; when unset, active memory uses the current session model |
+| `config.queryMode`          | `"message" \| "recent" \| "full"`                                                                    | Controls how much conversation the blocking memory sub-agent sees                                      |
+| `config.promptStyle`        | `"balanced" \| "strict" \| "contextual" \| "recall-heavy" \| "precision-heavy" \| "preference-only"` | Controls how eager or strict the blocking memory sub-agent is when deciding whether to return memory   |
+| `config.thinking`           | `"off" \| "minimal" \| "low" \| "medium" \| "high" \| "xhigh" \| "adaptive"`                         | Advanced thinking override for the blocking memory sub-agent; default `off` for speed                  |
+| `config.promptOverride`     | `string`                                                                                             | Advanced full prompt replacement; not recommended for normal use                                       |
+| `config.promptAppend`       | `string`                                                                                             | Advanced extra instructions appended to the default or overridden prompt                               |
+| `config.timeoutMs`          | `number`                                                                                             | Hard timeout for the blocking memory sub-agent                                                         |
+| `config.maxSummaryChars`    | `number`                                                                                             | Maximum total characters allowed in the active-memory summary                                          |
+| `config.logging`            | `boolean`                                                                                            | Emits active memory logs while tuning                                                                  |
+| `config.persistTranscripts` | `boolean`                                                                                            | Keeps blocking memory sub-agent transcripts on disk instead of deleting temp files                     |
+| `config.transcriptDir`      | `string`                                                                                             | Relative blocking memory sub-agent transcript directory under the agent sessions folder                |
+
+Useful tuning fields:
+
+| Key                           | Type     | Meaning                                                       |
+| ----------------------------- | -------- | ------------------------------------------------------------- |
+| `config.maxSummaryChars`      | `number` | Maximum total characters allowed in the active-memory summary |
+| `config.recentUserTurns`      | `number` | Prior user turns to include when `queryMode` is `recent`      |
+| `config.recentAssistantTurns` | `number` | Prior assistant turns to include when `queryMode` is `recent` |
+| `config.recentUserChars`      | `number` | Max chars per recent user turn                                |
+| `config.recentAssistantChars` | `number` | Max chars per recent assistant turn                           |
+| `config.cacheTtlMs`           | `number` | Cache reuse for repeated identical queries                    |
+
+## Recommended setup
+
+Start with `recent`.
+
+```json5
+{
+  plugins: {
+    entries: {
+      "active-memory": {
+        enabled: true,
+        config: {
+          agents: ["main"],
+          queryMode: "recent",
+          promptStyle: "balanced",
+          timeoutMs: 15000,
+          maxSummaryChars: 220,
+          logging: true,
+        },
+      },
+    },
+  },
+}
+```
+
+If you want to inspect live behavior while tuning, use `/verbose on` in the
+session instead of looking for a separate active-memory debug command.
+
+Then move to:
+
+- `message` if you want lower latency
+- `full` if you decide extra context is worth the slower blocking memory sub-agent
+
+## Debugging
+
+If active memory is not showing up where you expect:
+
+1. Confirm the plugin is enabled under `plugins.entries.active-memory.enabled`.
+2. Confirm the current agent id is listed in `config.agents`.
+3. Confirm you are testing through an interactive persistent chat session.
+4. Turn on `config.logging: true` and watch the gateway logs.
+5. Verify memory search itself works with `openclaw memory status --deep`.
+
+If memory hits are noisy, tighten:
+
+- `maxSummaryChars`
+
+If active memory is too slow:
+
+- lower `queryMode`
+- lower `timeoutMs`
+- reduce recent turn counts
+- reduce per-turn char caps
+
+## Related pages
+
+- [Memory Search](/concepts/memory-search)
+- [Memory configuration reference](/reference/memory-config)
+- [Plugin SDK setup](/plugins/sdk-setup)

docs/concepts/agent-loop.md

@@ -151,7 +151,7 @@ See [Plugin hooks](/plugins/architecture#provider-runtime-hooks) for the hook AP
 
 - `agent.wait` default: 30s (just the wait). `timeoutMs` param overrides.
 - Agent runtime: `agents.defaults.timeoutSeconds` default 172800s (48 hours); enforced in `runEmbeddedPiAgent` abort timer.
-- LLM idle timeout: `agents.defaults.llm.idleTimeoutSeconds` aborts a model request when no response chunks arrive before the idle window. Set it explicitly for slow local models or reasoning/tool-call providers; set it to 0 to disable. If it is not set, OpenClaw uses `agents.defaults.timeoutSeconds` when configured, otherwise 60s. Cron-triggered runs with no explicit LLM or agent timeout disable the idle watchdog and rely on the cron outer timeout.
+- LLM idle timeout: `agents.defaults.llm.idleTimeoutSeconds` aborts a model request when no response chunks arrive before the idle window. Set it explicitly for slow local models or reasoning/tool-call providers; set it to 0 to disable. If it is not set, OpenClaw uses `agents.defaults.timeoutSeconds` when configured, otherwise 120s. Cron-triggered runs with no explicit LLM or agent timeout disable the idle watchdog and rely on the cron outer timeout.
 
 ## Where things can end early
 

docs/concepts/dreaming.md

@@ -81,6 +81,20 @@ subagent turn (using the default runtime model) and appends a short diary entry.
 
 This diary is for human reading in the Dreams UI, not a promotion source.
 
+There is also a grounded historical backfill lane for review and recovery work:
+
+- `memory rem-harness --path ... --grounded` previews grounded diary output from historical `YYYY-MM-DD.md` notes.
+- `memory rem-backfill --path ...` writes reversible grounded diary entries into `DREAMS.md`.
+- `memory rem-backfill --path ... --stage-short-term` stages grounded durable candidates into the same short-term evidence store the normal deep phase already uses.
+- `memory rem-backfill --rollback` and `--rollback-short-term` remove those staged backfill artifacts without touching ordinary diary entries or live short-term recall.
+
+The Control UI exposes the same diary backfill/reset flow so you can inspect
+results in the Dreams scene before deciding whether the grounded candidates
+deserve promotion. The Scene also shows a distinct grounded lane so you can see
+which staged short-term entries came from historical replay, which promoted
+items were grounded-led, and clear only grounded-only staged entries without
+touching ordinary live short-term state.
+
 ## Deep ranking signals
 
 Deep ranking uses six weighted base signals plus phase reinforcement:
@@ -207,8 +221,9 @@ When enabled, the Gateway **Dreams** tab shows:
 
 - current dreaming enabled state
 - phase-level status and managed-sweep presence
-- short-term, long-term, and promoted-today counts
+- short-term, grounded, signal, and promoted-today counts
 - next scheduled run timing
+- a distinct grounded Scene lane for staged historical replay entries
 - an expandable Dream Diary reader backed by `doctor.memory.dreamDiary`
 
 ## Related

docs/concepts/memory-search.md

@@ -138,5 +138,6 @@ earlier conversations. This is opt-in via
 
 ## Further reading
 
+- [Active Memory](/concepts/active-memory) -- sub-agent memory for interactive chat sessions
 - [Memory](/concepts/memory) -- file layout, backends, tools
 - [Memory configuration reference](/reference/memory-config) -- all config knobs

docs/concepts/memory.md

@@ -21,7 +21,7 @@ Your agent has three memory-related files:
 - **`memory/YYYY-MM-DD.md`** -- daily notes. Running context and observations.
   Today and yesterday's notes are loaded automatically.
 - **`DREAMS.md`** (experimental, optional) -- Dream Diary and dreaming sweep
-  summaries for human review.
+  summaries for human review, including grounded historical backfill entries.
 
 These files live in the agent workspace (default `~/.openclaw/workspace`).
 
@@ -133,6 +133,41 @@ It is designed to keep long-term memory high signal:
 For phase behavior, scoring signals, and Dream Diary details, see
 [Dreaming (experimental)](/concepts/dreaming).
 
+## Grounded backfill and live promotion
+
+The dreaming system now has two closely related review lanes:
+
+- **Live dreaming** works from the short-term dreaming store under
+  `memory/.dreams/` and is what the normal deep phase uses when deciding what
+  can graduate into `MEMORY.md`.
+- **Grounded backfill** reads historical `memory/YYYY-MM-DD.md` notes as
+  standalone day files and writes structured review output into `DREAMS.md`.
+
+Grounded backfill is useful when you want to replay older notes and inspect what
+the system thinks is durable without manually editing `MEMORY.md`.
+
+When you use:
+
+```bash
+openclaw memory rem-backfill --path ./memory --stage-short-term
+```
+
+the grounded durable candidates are not promoted directly. They are staged into
+the same short-term dreaming store the normal deep phase already uses. That
+means:
+
+- `DREAMS.md` stays the human review surface.
+- the short-term store stays the machine-facing ranking surface.
+- `MEMORY.md` is still only written by deep promotion.
+
+If you decide the replay was not useful, you can remove the staged artifacts
+without touching ordinary diary entries or normal recall state:
+
+```bash
+openclaw memory rem-backfill --rollback
+openclaw memory rem-backfill --rollback-short-term
+```
+
 ## CLI
 
 ```bash

docs/concepts/model-providers.md

@@ -50,6 +50,13 @@ For model selection rules, see [/concepts/models](/concepts/models).
   family, transcript/tooling quirks, transport/cache hints). It is not the
   same as the [public capability model](/plugins/architecture#public-capability-model)
   which describes what a plugin registers (text inference, speech, etc.).
+- The bundled `codex` provider is paired with the bundled Codex agent harness.
+  Use `codex/gpt-*` when you want Codex-owned login, model discovery, native
+  thread resume, and app-server execution. Plain `openai/gpt-*` refs continue
+  to use the OpenAI provider and the normal OpenClaw provider transport.
+  Codex-only deployments can disable automatic PI fallback with
+  `agents.defaults.embeddedHarness.fallback: "none"`; see
+  [Codex Harness](/plugins/codex-harness).
 
 ## Plugin-owned provider behavior
 

docs/concepts/qa-e2e-automation.md

@@ -52,6 +52,66 @@ pnpm qa:lab:watch
 rebuilds that bundle on change, and the browser auto-reloads when the QA Lab
 asset hash changes.
 
+For a transport-real Matrix smoke lane, run:
+
+```bash
+pnpm openclaw qa matrix
+```
+
+That lane provisions a disposable Tuwunel homeserver in Docker, registers
+temporary driver, SUT, and observer users, creates one private room, then runs
+the real Matrix plugin inside a QA gateway child. The live transport lane keeps
+the child config scoped to the transport under test, so Matrix runs without
+`qa-channel` in the child config.
+
+For a transport-real Telegram smoke lane, run:
+
+```bash
+pnpm openclaw qa telegram
+```
+
+That lane targets one real private Telegram group instead of provisioning a
+disposable server. It requires `OPENCLAW_QA_TELEGRAM_GROUP_ID`,
+`OPENCLAW_QA_TELEGRAM_DRIVER_BOT_TOKEN`, and
+`OPENCLAW_QA_TELEGRAM_SUT_BOT_TOKEN`, plus two distinct bots in the same
+private group. The SUT bot must have a Telegram username, and bot-to-bot
+observation works best when both bots have Bot-to-Bot Communication Mode
+enabled in `@BotFather`.
+
+Live transport lanes now share one smaller contract instead of each inventing
+their own scenario list shape:
+
+`qa-channel` remains the broad synthetic product-behavior suite and is not part
+of the live transport coverage matrix.
+
+| Lane     | Canary | Mention gating | Allowlist block | Top-level reply | Restart resume | Thread follow-up | Thread isolation | Reaction observation | Help command |
+| -------- | ------ | -------------- | --------------- | --------------- | -------------- | ---------------- | ---------------- | -------------------- | ------------ |
+| Matrix   | x      | x              | x               | x               | x              | x                | x                | x                    |              |
+| Telegram | x      |                |                 |                 |                |                  |                  |                      | x            |
+
+This keeps `qa-channel` as the broad product-behavior suite while Matrix,
+Telegram, and future live transports share one explicit transport-contract
+checklist.
+
+For a disposable Linux VM lane without bringing Docker into the QA path, run:
+
+```bash
+pnpm openclaw qa suite --runner multipass --scenario channel-chat-baseline
+```
+
+This boots a fresh Multipass guest, installs dependencies, builds OpenClaw
+inside the guest, runs `qa suite`, then copies the normal QA report and
+summary back into `.artifacts/qa-e2e/...` on the host.
+It reuses the same scenario-selection behavior as `qa suite` on the host.
+Host and Multipass suite runs execute multiple selected scenarios in parallel
+with isolated gateway workers by default, up to 64 workers or the selected
+scenario count. Use `--concurrency <count>` to tune the worker count, or
+`--concurrency 1` for serial execution.
+Live runs forward the supported QA auth inputs that are practical for the
+guest: env-based provider keys, the QA live provider config path, and
+`CODEX_HOME` when present. Keep `--output-dir` under the repo root so the guest
+can write back through the mounted workspace.
+
 ## Repo-backed seeds
 
 Seed assets live in `qa/`:
@@ -89,18 +149,17 @@ refs and write a judged Markdown report:
 pnpm openclaw qa character-eval \
   --model openai/gpt-5.4,thinking=xhigh \
   --model openai/gpt-5.2,thinking=xhigh \
+  --model openai/gpt-5,thinking=xhigh \
   --model anthropic/claude-opus-4-6,thinking=high \
   --model anthropic/claude-sonnet-4-6,thinking=high \
-  --model minimax/MiniMax-M2.7,thinking=high \
   --model zai/glm-5.1,thinking=high \
   --model moonshot/kimi-k2.5,thinking=high \
-  --model qwen/qwen3.6-plus,thinking=high \
-  --model xiaomi/mimo-v2-pro,thinking=high \
   --model google/gemini-3.1-pro-preview,thinking=high \
   --judge-model openai/gpt-5.4,thinking=xhigh,fast \
   --judge-model anthropic/claude-opus-4-6,thinking=high \
-  --concurrency 8 \
-  --judge-concurrency 8
+  --blind-judge-models \
+  --concurrency 16 \
+  --judge-concurrency 16
 ```
 
 The command runs local QA gateway child processes, not Docker. Character eval
@@ -109,6 +168,10 @@ such as chat, workspace help, and small file tasks. The candidate model should
 not be told that it is being evaluated. The command preserves each full
 transcript, records basic run stats, then asks the judge models in fast mode with
 `xhigh` reasoning to rank the runs by naturalness, vibe, and humor.
+Use `--blind-judge-models` when comparing providers: the judge prompt still gets
+every transcript and run status, but candidate refs are replaced with neutral
+labels such as `candidate-01`; the report maps rankings back to real refs after
+parsing.
 Candidate runs default to `high` thinking, with `xhigh` for OpenAI models that
 support it. Override a specific candidate inline with
 `--model provider/model,thinking=<level>`. `--thinking <level>` still sets a
@@ -120,14 +183,14 @@ single candidate or judge needs an override. Pass `--fast` only when you want to
 force fast mode on for every candidate model. Candidate and judge durations are
 recorded in the report for benchmark analysis, but judge prompts explicitly say
 not to rank by speed.
-Candidate and judge model runs both default to concurrency 8. Lower
+Candidate and judge model runs both default to concurrency 16. Lower
 `--concurrency` or `--judge-concurrency` when provider limits or local gateway
 pressure make a run too noisy.
 When no candidate `--model` is passed, the character eval defaults to
-`openai/gpt-5.4`, `openai/gpt-5.2`, `anthropic/claude-opus-4-6`,
-`anthropic/claude-sonnet-4-6`, `minimax/MiniMax-M2.7`, `zai/glm-5.1`,
-`moonshot/kimi-k2.5`, `qwen/qwen3.6-plus`, `xiaomi/mimo-v2-pro`, and
-`google/gemini-3.1-pro-preview`.
+`openai/gpt-5.4`, `openai/gpt-5.2`, `openai/gpt-5`, `anthropic/claude-opus-4-6`,
+`anthropic/claude-sonnet-4-6`, `zai/glm-5.1`,
+`moonshot/kimi-k2.5`, and
+`google/gemini-3.1-pro-preview` when no `--model` is passed.
 When no `--judge-model` is passed, the judges default to
 `openai/gpt-5.4,thinking=xhigh,fast` and
 `anthropic/claude-opus-4-6,thinking=high`.

docs/docs.json

@@ -1074,6 +1074,7 @@
                       "concepts/memory-qmd",
                       "concepts/memory-honcho",
                       "concepts/memory-search",
+                      "concepts/active-memory",
                       "concepts/dreaming"
                     ]
                   },
@@ -1112,6 +1113,7 @@
                   "tools/plugin",
                   "plugins/community",
                   "plugins/bundles",
+                  "plugins/codex-harness",
                   "plugins/webhooks",
                   "plugins/voice-call",
                   {
@@ -1129,6 +1131,7 @@
                       "plugins/sdk-overview",
                       "plugins/sdk-entrypoints",
                       "plugins/sdk-runtime",
+                      "plugins/sdk-agent-harness",
                       "plugins/sdk-setup",
                       "plugins/sdk-testing",
                       "plugins/manifest",

docs/gateway/cli-backends.md

@@ -159,6 +159,14 @@ model_instructions_file="..."`). Codex does not expose a Claude-style
 `--append-system-prompt` flag, so OpenClaw writes the assembled prompt to a
 temporary file for each fresh Codex CLI session.
 
+The bundled Anthropic `claude-cli` backend receives the OpenClaw skills snapshot
+two ways: the compact OpenClaw skills catalog in the appended system prompt, and
+a temporary Claude Code plugin passed with `--plugin-dir`. The plugin contains
+only the eligible skills for that agent/session, so Claude Code's native skill
+resolver sees the same filtered set that OpenClaw would otherwise advertise in
+the prompt. Skill env/API key overrides are still applied by OpenClaw to the
+child process environment for the run.
+
 ## Sessions
 
 - If the CLI supports sessions, set `sessionArg` (e.g. `--session-id`) or
@@ -255,6 +263,31 @@ CLI backend defaults are now part of the plugin surface:
 - Backend-specific config cleanup stays plugin-owned through the optional
   `normalizeConfig` hook.
 
+Plugins that need tiny prompt/message compatibility shims can declare
+bidirectional text transforms without replacing a provider or CLI backend:
+
+```typescript
+api.registerTextTransforms({
+  input: [
+    { from: /red basket/g, to: "blue basket" },
+    { from: /paper ticket/g, to: "digital ticket" },
+    { from: /left shelf/g, to: "right shelf" },
+  ],
+  output: [
+    { from: /blue basket/g, to: "red basket" },
+    { from: /digital ticket/g, to: "paper ticket" },
+    { from: /right shelf/g, to: "left shelf" },
+  ],
+});
+```
+
+`input` rewrites the system prompt and user prompt passed to the CLI. `output`
+rewrites streamed assistant deltas and parsed final text before OpenClaw handles
+its own control markers and channel delivery.
+
+For CLIs that emit Claude Code stream-json compatible JSONL, set
+`jsonlDialect: "claude-stream-json"` on that backend's config.
+
 ## Bundle MCP overlays
 
 CLI backends do **not** receive OpenClaw tool calls directly, but a backend can

docs/gateway/configuration-reference.md

@@ -1053,6 +1053,10 @@ Time format in system prompt. Default: `auto` (OS preference).
         fallbacks: ["openai/gpt-5.4-mini"],
       },
       params: { cacheRetention: "long" }, // global default provider params
+      embeddedHarness: {
+        runtime: "auto", // auto | pi | registered harness id, e.g. codex
+        fallback: "pi", // pi | none
+      },
       pdfMaxBytesMb: 10,
       pdfMaxPages: 20,
       thinkingDefault: "low",
@@ -1100,9 +1104,37 @@ Time format in system prompt. Default: `auto` (OS preference).
 - `models`: the configured model catalog and allowlist for `/model`. Each entry can include `alias` (shortcut) and `params` (provider-specific, for example `temperature`, `maxTokens`, `cacheRetention`, `context1m`).
 - `params`: global default provider parameters applied to all models. Set at `agents.defaults.params` (e.g. `{ cacheRetention: "long" }`).
 - `params` merge precedence (config): `agents.defaults.params` (global base) is overridden by `agents.defaults.models["provider/model"].params` (per-model), then `agents.list[].params` (matching agent id) overrides by key. See [Prompt Caching](/reference/prompt-caching) for details.
+- `embeddedHarness`: default low-level embedded agent runtime policy. Use `runtime: "auto"` to let registered plugin harnesses claim supported models, `runtime: "pi"` to force the built-in PI harness, or a registered harness id such as `runtime: "codex"`. Set `fallback: "none"` to disable automatic PI fallback.
 - Config writers that mutate these fields (for example `/models set`, `/models set-image`, and fallback add/remove commands) save canonical object form and preserve existing fallback lists when possible.
 - `maxConcurrent`: max parallel agent runs across sessions (each session still serialized). Default: 4.
 
+### `agents.defaults.embeddedHarness`
+
+`embeddedHarness` controls which low-level executor runs embedded agent turns.
+Most deployments should keep the default `{ runtime: "auto", fallback: "pi" }`.
+Use it when a trusted plugin provides a native harness, such as the bundled
+Codex app-server harness.
+
+```json5
+{
+  agents: {
+    defaults: {
+      model: "codex/gpt-5.4",
+      embeddedHarness: {
+        runtime: "codex",
+        fallback: "none",
+      },
+    },
+  },
+}
+```
+
+- `runtime`: `"auto"`, `"pi"`, or a registered plugin harness id. The bundled Codex plugin registers `codex`.
+- `fallback`: `"pi"` or `"none"`. `"pi"` keeps the built-in PI harness as the compatibility fallback. `"none"` makes missing or unsupported plugin harness selection fail instead of silently using PI.
+- Environment overrides: `OPENCLAW_AGENT_RUNTIME=<id|auto|pi>` overrides `runtime`; `OPENCLAW_AGENT_HARNESS_FALLBACK=none` disables PI fallback for that process.
+- For Codex-only deployments, set `model: "codex/gpt-5.4"`, `embeddedHarness.runtime: "codex"`, and `embeddedHarness.fallback: "none"`.
+- This only controls the embedded chat harness. Media generation, vision, PDF, music, video, and TTS still use their provider/model settings.
+
 **Built-in alias shorthands** (only apply when the model is in `agents.defaults.models`):
 
 | Alias               | Model                                  |
@@ -1192,6 +1224,7 @@ Periodic heartbeat runs.
         prompt: "Read HEARTBEAT.md if it exists...",
         ackMaxChars: 300,
         suppressToolErrorWarnings: false,
+        timeoutSeconds: 45,
       },
     },
   },
@@ -1201,6 +1234,7 @@ Periodic heartbeat runs.
 - `every`: duration string (ms/s/m/h). Default: `30m` (API-key auth) or `1h` (OAuth auth). Set to `0m` to disable.
 - `includeSystemPromptSection`: when false, omits the Heartbeat section from the system prompt and skips `HEARTBEAT.md` injection into bootstrap context. Default: `true`.
 - `suppressToolErrorWarnings`: when true, suppresses tool error warning payloads during heartbeat runs.
+- `timeoutSeconds`: maximum time in seconds allowed for a heartbeat agent turn before it is aborted. Leave unset to use `agents.defaults.timeoutSeconds`.
 - `directPolicy`: direct/DM delivery policy. `allow` (default) permits direct-target delivery. `block` suppresses direct-target delivery and emits `reason=dm-blocked`.
 - `lightContext`: when true, heartbeat runs use lightweight bootstrap context and keep only `HEARTBEAT.md` from workspace bootstrap files.
 - `isolatedSession`: when true, each heartbeat runs in a fresh session with no prior conversation history. Same isolation pattern as cron `sessionTarget: "isolated"`. Reduces per-heartbeat token cost from ~100K to ~2-5K tokens.
@@ -1583,6 +1617,7 @@ scripts/sandbox-browser-setup.sh   # optional browser image
         thinkingDefault: "high", // per-agent thinking level override
         reasoningDefault: "on", // per-agent reasoning visibility override
         fastModeDefault: false, // per-agent fast mode override
+        embeddedHarness: { runtime: "auto", fallback: "pi" },
         params: { cacheRetention: "none" }, // overrides matching defaults.models params by key
         skills: ["docs-search"], // replaces agents.defaults.skills when set
         identity: {
@@ -1623,6 +1658,7 @@ scripts/sandbox-browser-setup.sh   # optional browser image
 - `thinkingDefault`: optional per-agent default thinking level (`off | minimal | low | medium | high | xhigh | adaptive`). Overrides `agents.defaults.thinkingDefault` for this agent when no per-message or session override is set.
 - `reasoningDefault`: optional per-agent default reasoning visibility (`on | off | stream`). Applies when no per-message or session reasoning override is set.
 - `fastModeDefault`: optional per-agent default for fast mode (`true | false`). Applies when no per-message or session fast-mode override is set.
+- `embeddedHarness`: optional per-agent low-level harness policy override. Use `{ runtime: "codex", fallback: "none" }` to make one agent Codex-only while other agents keep the default PI fallback.
 - `runtime`: optional per-agent runtime descriptor. Use `type: "acp"` with `runtime.acp` defaults (`agent`, `backend`, `mode`, `cwd`) when the agent should default to ACP harness sessions.
 - `identity.avatar`: workspace-relative path, `http(s)` URL, or `data:` URI.
 - `identity` derives defaults: `ackReaction` from `emoji`, `mentionPatterns` from `name`/`emoji`.
@@ -2299,7 +2335,7 @@ Notes:
 
 ### `tools.experimental`
 
-Experimental built-in tool flags. Default off unless a runtime-specific auto-enable rule applies.
+Experimental built-in tool flags. Default off unless a strict-agentic GPT-5 auto-enable rule applies.
 
 ```json5
 {
@@ -2314,7 +2350,7 @@ Experimental built-in tool flags. Default off unless a runtime-specific auto-ena
 Notes:
 
 - `planTool`: enables the structured `update_plan` tool for non-trivial multi-step work tracking.
-- Default: `false` for non-OpenAI providers. OpenAI and OpenAI Codex runs auto-enable it when unset; set `false` to disable that auto-enable.
+- Default: `false` unless `agents.defaults.embeddedPi.executionContract` (or a per-agent override) is set to `"strict-agentic"` for an OpenAI or OpenAI Codex GPT-5-family run. Set `true` to force the tool on outside that scope, or `false` to keep it off even for strict-agentic GPT-5 runs.
 - When enabled, the system prompt also adds usage guidance so the model only uses it for substantial work and keeps at most one step `in_progress`.
 
 ### `agents.defaults.subagents`
@@ -2402,6 +2438,7 @@ OpenClaw uses the built-in model catalog. Add custom providers via `models.provi
   - `request.auth`: auth strategy override. Modes: `"provider-default"` (use provider's built-in auth), `"authorization-bearer"` (with `token`), `"header"` (with `headerName`, `value`, optional `prefix`).
   - `request.proxy`: HTTP proxy override. Modes: `"env-proxy"` (use `HTTP_PROXY`/`HTTPS_PROXY` env vars), `"explicit-proxy"` (with `url`). Both modes accept an optional `tls` sub-object.
   - `request.tls`: TLS override for direct connections. Fields: `ca`, `cert`, `key`, `passphrase` (all accept SecretRef), `serverName`, `insecureSkipVerify`.
+  - `request.allowPrivateNetwork`: when `true`, allow HTTPS to `baseUrl` when DNS resolves to private, CGNAT, or similar ranges, via the provider HTTP fetch guard (operator opt-in for trusted self-hosted OpenAI-compatible endpoints). WebSocket uses the same `request` for headers/TLS but not that fetch SSRF gate. Default `false`.
 - `models.providers.*.models`: explicit provider model catalog entries.
 - `models.providers.*.models.*.contextWindow`: native model context window metadata.
 - `models.providers.*.models.*.contextTokens`: optional runtime context cap. Use this when you want a smaller effective context budget than the model's native `contextWindow`.
@@ -2757,7 +2794,7 @@ See [Plugins](/tools/plugin).
     evaluateEnabled: true,
     defaultProfile: "user",
     ssrfPolicy: {
-      dangerouslyAllowPrivateNetwork: true, // default trusted-network mode
+      // dangerouslyAllowPrivateNetwork: true, // opt in only for trusted private-network access
       // allowPrivateNetwork: true, // legacy alias
       // hostnameAllowlist: ["*.example.com", "example.com"],
       // allowedHostnames: ["localhost"],
@@ -2785,8 +2822,8 @@ See [Plugins](/tools/plugin).
 ```
 
 - `evaluateEnabled: false` disables `act:evaluate` and `wait --fn`.
-- `ssrfPolicy.dangerouslyAllowPrivateNetwork` defaults to `true` when unset (trusted-network model).
-- Set `ssrfPolicy.dangerouslyAllowPrivateNetwork: false` for strict public-only browser navigation.
+- `ssrfPolicy.dangerouslyAllowPrivateNetwork` is disabled when unset, so browser navigation stays strict by default.
+- Set `ssrfPolicy.dangerouslyAllowPrivateNetwork: true` only when you intentionally trust private-network browser navigation.
 - In strict mode, remote CDP profile endpoints (`profiles.*.cdpUrl`) are subject to the same private-network blocking during reachability/discovery checks.
 - `ssrfPolicy.allowPrivateNetwork` remains supported as a legacy alias.
 - In strict mode, use `ssrfPolicy.hostnameAllowlist` and `ssrfPolicy.allowedHostnames` for explicit exceptions.

docs/gateway/configuration.md

@@ -224,7 +224,7 @@ When validation fails:
     - Omit `agents.list[].skills` to inherit the defaults.
     - Set `agents.list[].skills: []` for no skills.
     - See [Skills](/tools/skills), [Skills config](/tools/skills-config), and
-      the [Configuration Reference](/gateway/configuration-reference#agentsdefaultsskills).
+      the [Configuration Reference](/gateway/configuration-reference#agents-defaults-skills).
 
   </Accordion>
 

docs/gateway/doctor.md

@@ -93,6 +93,40 @@ cat ~/.openclaw/openclaw.json
 - Source install checks (pnpm workspace mismatch, missing UI assets, missing tsx binary).
 - Writes updated config + wizard metadata.
 
+## Dreams UI backfill and reset
+
+The Control UI Dreams scene includes **Backfill**, **Reset**, and **Clear Grounded**
+actions for the grounded dreaming workflow. These actions use gateway
+doctor-style RPC methods, but they are **not** part of `openclaw doctor` CLI
+repair/migration.
+
+What they do:
+
+- **Backfill** scans historical `memory/YYYY-MM-DD.md` files in the active
+  workspace, runs the grounded REM diary pass, and writes reversible backfill
+  entries into `DREAMS.md`.
+- **Reset** removes only those marked backfill diary entries from `DREAMS.md`.
+- **Clear Grounded** removes only staged grounded-only short-term entries that
+  came from historical replay and have not accumulated live recall or daily
+  support yet.
+
+What they do **not** do by themselves:
+
+- they do not edit `MEMORY.md`
+- they do not run full doctor migrations
+- they do not automatically stage grounded candidates into the live short-term
+  promotion store unless you explicitly run the staged CLI path first
+
+If you want grounded historical replay to influence the normal deep promotion
+lane, use the CLI flow instead:
+
+```bash
+openclaw memory rem-backfill --path ./memory --stage-short-term
+```
+
+That stages grounded durable candidates into the short-term dreaming store while
+keeping `DREAMS.md` as the review surface.
+
 ## Detailed behavior and rationale
 
 ### 0) Optional update (git installs)

docs/gateway/heartbeat.md

@@ -146,6 +146,7 @@ Example: two agents, only the second agent runs heartbeats.
           every: "1h",
           target: "whatsapp",
           to: "+15551234567",
+          timeoutSeconds: 45,
           prompt: "Read HEARTBEAT.md if it exists (workspace context). Follow it strictly. Do not infer or repeat old tasks from prior chats. If nothing needs attention, reply HEARTBEAT_OK.",
         },
       },

docs/gateway/protocol.md

@@ -400,7 +400,7 @@ implemented in `src/gateway/server-methods/*.ts`.
   - `wake` schedules an immediate or next-heartbeat wake text injection
   - `cron.list`, `cron.status`, `cron.add`, `cron.update`, `cron.remove`,
     `cron.run`, `cron.runs`
-- skills/tools: `skills.*`, `tools.catalog`, `tools.effective`
+- skills/tools: `commands.list`, `skills.*`, `tools.catalog`, `tools.effective`
 
 ### Common event families
 
@@ -431,6 +431,18 @@ implemented in `src/gateway/server-methods/*.ts`.
 
 ### Operator helper methods
 
+- Operators may call `commands.list` (`operator.read`) to fetch the runtime
+  command inventory for an agent.
+  - `agentId` is optional; omit it to read the default agent workspace.
+  - `scope` controls which surface the primary `name` targets:
+    - `text` returns the primary text command token without the leading `/`
+    - `native` and the default `both` path return provider-aware native names
+      when available
+  - `textAliases` carries exact slash aliases such as `/model` and `/m`.
+  - `nativeName` carries the provider-aware native command name when one exists.
+  - `provider` is optional and only affects native naming plus native plugin
+    command availability.
+  - `includeArgs=false` omits serialized argument metadata from the response.
 - Operators may call `tools.catalog` (`operator.read`) to fetch the runtime tool catalog for an
   agent. The response includes grouped tools and provenance metadata:
   - `source`: `core` or `plugin`

docs/gateway/security/index.md

@@ -13,7 +13,7 @@ OpenClaw is **not** a hostile multi-tenant security boundary for multiple advers
 If you need mixed-trust or adversarial-user operation, split trust boundaries (separate gateway + credentials, ideally separate OS users/hosts).
 </Warning>
 
-**On this page:** [Trust model](#scope-first-personal-assistant-security-model) | [Quick audit](#quick-check-openclaw-security-audit) | [Hardened baseline](#hardened-baseline-in-60-seconds) | [DM access model](#dm-access-model-pairing--allowlist--open--disabled) | [Configuration hardening](#configuration-hardening-examples) | [Incident response](#incident-response)
+**On this page:** [Trust model](#scope-first-personal-assistant-security-model) | [Quick audit](#quick-check-openclaw-security-audit) | [Hardened baseline](#hardened-baseline-in-60-seconds) | [DM access model](#dm-access-model-pairing-allowlist-open-disabled) | [Configuration hardening](#configuration-hardening-examples) | [Incident response](#incident-response)
 
 ## Scope first: personal assistant security model
 
@@ -187,7 +187,7 @@ Allowlists gate triggers and command authorization. The `contextVisibility` sett
 - `contextVisibility: "allowlist"` filters supplemental context to senders allowed by the active allowlist checks.
 - `contextVisibility: "allowlist_quote"` behaves like `allowlist`, but still keeps one explicit quoted reply.
 
-Set `contextVisibility` per channel or per room/conversation. See [Group Chats](/channels/groups#context-visibility) for setup details.
+Set `contextVisibility` per channel or per room/conversation. See [Group Chats](/channels/groups#context-visibility-and-allowlists) for setup details.
 
 Advisory triage guidance:
 
@@ -579,6 +579,8 @@ Plugins run **in-process** with the Gateway. Treat them as trusted code:
 
 Details: [Plugins](/tools/plugin)
 
+<a id="dm-access-model-pairing-allowlist-open-disabled"></a>
+
 ## DM access model (pairing / allowlist / open / disabled)
 
 All current DM-capable channels support a DM policy (`dmPolicy` or `*.dm.policy`) that gates inbound DMs **before** the message is processed:
@@ -1149,13 +1151,13 @@ access those accounts and data. Treat browser profiles as **sensitive state**:
 - Disable browser proxy routing when you don’t need it (`gateway.nodes.browser.mode="off"`).
 - Chrome MCP existing-session mode is **not** “safer”; it can act as you in whatever that host Chrome profile can reach.
 
-### Browser SSRF policy (trusted-network default)
+### Browser SSRF policy (strict by default)
 
-OpenClaw’s browser network policy defaults to the trusted-operator model: private/internal destinations are allowed unless you explicitly disable them.
+OpenClaw’s browser navigation policy is strict by default: private/internal destinations stay blocked unless you explicitly opt in.
 
-- Default: `browser.ssrfPolicy.dangerouslyAllowPrivateNetwork: true` (implicit when unset).
+- Default: `browser.ssrfPolicy.dangerouslyAllowPrivateNetwork` is unset, so browser navigation keeps private/internal/special-use destinations blocked.
 - Legacy alias: `browser.ssrfPolicy.allowPrivateNetwork` is still accepted for compatibility.
-- Strict mode: set `browser.ssrfPolicy.dangerouslyAllowPrivateNetwork: false` to block private/internal/special-use destinations by default.
+- Opt-in mode: set `browser.ssrfPolicy.dangerouslyAllowPrivateNetwork: true` to allow private/internal/special-use destinations.
 - In strict mode, use `hostnameAllowlist` (patterns like `*.example.com`) and `allowedHostnames` (exact host exceptions, including blocked names like `localhost`) for explicit exceptions.
 - Navigation is checked before request and best-effort re-checked on the final `http(s)` URL after navigation to reduce redirect-based pivots.
 

docs/gateway/troubleshooting.md

@@ -111,7 +111,7 @@ Fix options:
 Related:
 
 - [/gateway/local-models](/gateway/local-models)
-- [/gateway/configuration#models](/gateway/configuration#models)
+- [/gateway/configuration](/gateway/configuration)
 - [/gateway/configuration-reference#openai-compatible-endpoints](/gateway/configuration-reference#openai-compatible-endpoints)
 
 ## No replies

docs/help/testing.md

@@ -26,7 +26,9 @@ Most days:
 - Faster local full-suite run on a roomy machine: `pnpm test:max`
 - Direct Vitest watch loop: `pnpm test:watch`
 - Direct file targeting now routes extension/channel paths too: `pnpm test extensions/discord/src/monitor/message-handler.preflight.test.ts`
+- Prefer targeted runs first when you are iterating on a single failure.
 - Docker-backed QA site: `pnpm qa:lab:up`
+- Linux VM-backed QA lane: `pnpm openclaw qa suite --runner multipass --scenario channel-chat-baseline`
 
 When you touch tests or want extra confidence:
 
@@ -40,6 +42,51 @@ When debugging real providers/models (requires real creds):
 
 Tip: when you only need one failing case, prefer narrowing live tests via the allowlist env vars described below.
 
+## QA-specific runners
+
+These commands sit beside the main test suites when you need QA-lab realism:
+
+- `pnpm openclaw qa suite`
+  - Runs repo-backed QA scenarios directly on the host.
+  - Runs multiple selected scenarios in parallel by default with isolated
+    gateway workers, up to 64 workers or the selected scenario count. Use
+    `--concurrency <count>` to tune the worker count, or `--concurrency 1` for
+    the older serial lane.
+- `pnpm openclaw qa suite --runner multipass`
+  - Runs the same QA suite inside a disposable Multipass Linux VM.
+  - Keeps the same scenario-selection behavior as `qa suite` on the host.
+  - Reuses the same provider/model selection flags as `qa suite`.
+  - Live runs forward the supported QA auth inputs that are practical for the guest:
+    env-based provider keys, the QA live provider config path, and `CODEX_HOME`
+    when present.
+  - Output dirs must stay under the repo root so the guest can write back through
+    the mounted workspace.
+  - Writes the normal QA report + summary plus Multipass logs under
+    `.artifacts/qa-e2e/...`.
+- `pnpm qa:lab:up`
+  - Starts the Docker-backed QA site for operator-style QA work.
+- `pnpm openclaw qa matrix`
+  - Runs the Matrix live QA lane against a disposable Docker-backed Tuwunel homeserver.
+  - Provisions three temporary Matrix users (`driver`, `sut`, `observer`) plus one private room, then starts a QA gateway child with the real Matrix plugin as the SUT transport.
+  - Uses the pinned stable Tuwunel image `ghcr.io/matrix-construct/tuwunel:v1.5.1` by default. Override with `OPENCLAW_QA_MATRIX_TUWUNEL_IMAGE` when you need to test a different image.
+  - Writes a Matrix QA report, summary, and observed-events artifact under `.artifacts/qa-e2e/...`.
+- `pnpm openclaw qa telegram`
+  - Runs the Telegram live QA lane against a real private group using the driver and SUT bot tokens from env.
+  - Requires `OPENCLAW_QA_TELEGRAM_GROUP_ID`, `OPENCLAW_QA_TELEGRAM_DRIVER_BOT_TOKEN`, and `OPENCLAW_QA_TELEGRAM_SUT_BOT_TOKEN`. The group id must be the numeric Telegram chat id.
+  - Requires two distinct bots in the same private group, with the SUT bot exposing a Telegram username.
+  - For stable bot-to-bot observation, enable Bot-to-Bot Communication Mode in `@BotFather` for both bots and ensure the driver bot can observe group bot traffic.
+  - Writes a Telegram QA report, summary, and observed-messages artifact under `.artifacts/qa-e2e/...`.
+
+Live transport lanes share one standard contract so new transports do not drift:
+
+`qa-channel` remains the broad synthetic QA suite and is not part of the live
+transport coverage matrix.
+
+| Lane     | Canary | Mention gating | Allowlist block | Top-level reply | Restart resume | Thread follow-up | Thread isolation | Reaction observation | Help command |
+| -------- | ------ | -------------- | --------------- | --------------- | -------------- | ---------------- | ---------------- | -------------------- | ------------ |
+| Matrix   | x      | x              | x               | x               | x              | x                | x                | x                    |              |
+| Telegram | x      |                |                 |                 |                |                  |                  |                      | x            |
+
 ## Test suites (what runs where)
 
 Think of the suites as “increasing realism” (and increasing flakiness/cost):
@@ -62,7 +109,7 @@ Think of the suites as “increasing realism” (and increasing flakiness/cost):
   - `pnpm test --watch` still uses the native root `vitest.config.ts` project graph, because a multi-shard watch loop is not practical.
   - `pnpm test`, `pnpm test:watch`, and `pnpm test:perf:imports` route explicit file/directory targets through scoped lanes first, so `pnpm test extensions/discord/src/monitor/message-handler.preflight.test.ts` avoids paying the full root project startup tax.
   - `pnpm test:changed` expands changed git paths into the same scoped lanes when the diff only touches routable source/test files; config/setup edits still fall back to the broad root-project rerun.
-  - Selected `plugin-sdk` and `commands` tests also route through dedicated light lanes that skip `test/setup-openclaw-runtime.ts`; stateful/runtime-heavy files stay on the existing lanes.
+  - Import-light unit tests from agents, commands, plugins, auto-reply helpers, `plugin-sdk`, and similar pure utility areas route through the `unit-fast` lane, which skips `test/setup-openclaw-runtime.ts`; stateful/runtime-heavy files stay on the existing lanes.
   - Selected `plugin-sdk` and `commands` helper source files also map changed-mode runs to explicit sibling tests in those light lanes, so helper edits avoid rerunning the full heavy suite for that directory.
   - `auto-reply` now has three dedicated buckets: top-level core helpers, top-level `reply.*` integration tests, and the `src/auto-reply/reply/**` subtree. This keeps the heaviest reply harness work off the cheap status/chunk/token tests.
 - Embedded runner note:
@@ -203,6 +250,7 @@ Live tests are split into two layers so we can isolate failures:
   - `OPENCLAW_LIVE_MODELS=modern` to run the modern allowlist (Opus/Sonnet 4.6+, GPT-5.x + Codex, Gemini 3, GLM 4.7, MiniMax M2.7, Grok 4)
   - `OPENCLAW_LIVE_MODELS=all` is an alias for the modern allowlist
   - or `OPENCLAW_LIVE_MODELS="openai/gpt-5.4,anthropic/claude-opus-4-6,..."` (comma allowlist)
+  - Modern/all sweeps default to a curated high-signal cap; set `OPENCLAW_LIVE_MAX_MODELS=0` for an exhaustive modern sweep or a positive number for a smaller cap.
 - How to select providers:
   - `OPENCLAW_LIVE_PROVIDERS="google,google-antigravity,google-gemini-cli"` (comma allowlist)
 - Where keys come from:
@@ -234,6 +282,7 @@ Live tests are split into two layers so we can isolate failures:
   - Default: modern allowlist (Opus/Sonnet 4.6+, GPT-5.x + Codex, Gemini 3, GLM 4.7, MiniMax M2.7, Grok 4)
   - `OPENCLAW_LIVE_GATEWAY_MODELS=all` is an alias for the modern allowlist
   - Or set `OPENCLAW_LIVE_GATEWAY_MODELS="provider/model"` (or comma list) to narrow
+  - Modern/all gateway sweeps default to a curated high-signal cap; set `OPENCLAW_LIVE_GATEWAY_MAX_MODELS=0` for an exhaustive modern sweep or a positive number for a smaller cap.
 - How to select providers (avoid “OpenRouter everything”):
   - `OPENCLAW_LIVE_GATEWAY_PROVIDERS="google,google-antigravity,google-gemini-cli,openai,anthropic,zai,minimax"` (comma allowlist)
 - Tool + image probes are always on in this live test:
@@ -292,6 +341,7 @@ Single-provider Docker recipes:
 
 ```bash
 pnpm test:docker:live-cli-backend:claude
+pnpm test:docker:live-cli-backend:claude-subscription
 pnpm test:docker:live-cli-backend:codex
 pnpm test:docker:live-cli-backend:gemini
 ```
@@ -301,6 +351,7 @@ Notes:
 - The Docker runner lives at `scripts/test-live-cli-backend-docker.sh`.
 - It runs the live CLI-backend smoke inside the repo Docker image as the non-root `node` user.
 - It resolves CLI smoke metadata from the owning extension, then installs the matching Linux CLI package (`@anthropic-ai/claude-code`, `@openai/codex`, or `@google/gemini-cli`) into a cached writable prefix at `OPENCLAW_DOCKER_CLI_TOOLS_DIR` (default: `~/.cache/openclaw/docker-cli-tools`).
+- `pnpm test:docker:live-cli-backend:claude-subscription` requires portable Claude Code subscription OAuth through either `~/.claude/.credentials.json` with `claudeAiOauth.subscriptionType` or `CLAUDE_CODE_OAUTH_TOKEN` from `claude setup-token`. It first proves direct `claude -p` in Docker, then runs two Gateway CLI-backend turns without preserving Anthropic API-key env vars. This subscription lane disables the Claude MCP/tool and image probes by default because Claude currently routes third-party app usage through extra-usage billing instead of normal subscription plan limits.
 - The live CLI-backend smoke now exercises the same end-to-end flow for Claude, Codex, and Gemini: text turn, image classification turn, then MCP `cron` tool call verified through the gateway CLI.
 - Claude's default smoke also patches the session from Sonnet to Opus and verifies the resumed session still remembers an earlier note.
 
@@ -360,6 +411,58 @@ Docker notes:
 - It sources `~/.profile`, stages the matching CLI auth material into the container, installs `acpx` into a writable npm prefix, then installs the requested live CLI (`@anthropic-ai/claude-code`, `@openai/codex`, or `@google/gemini-cli`) if missing.
 - Inside Docker, the runner sets `OPENCLAW_LIVE_ACP_BIND_ACPX_COMMAND=$HOME/.npm-global/bin/acpx` so acpx keeps provider env vars from the sourced profile available to the child harness CLI.
 
+## Live: Codex app-server harness smoke
+
+- Goal: validate the plugin-owned Codex harness through the normal gateway
+  `agent` method:
+  - load the bundled `codex` plugin
+  - select `OPENCLAW_AGENT_RUNTIME=codex`
+  - send a first gateway agent turn to `codex/gpt-5.4`
+  - send a second turn to the same OpenClaw session and verify the app-server
+    thread can resume
+  - run `/codex status` and `/codex models` through the same gateway command
+    path
+- Test: `src/gateway/gateway-codex-harness.live.test.ts`
+- Enable: `OPENCLAW_LIVE_CODEX_HARNESS=1`
+- Default model: `codex/gpt-5.4`
+- Optional image probe: `OPENCLAW_LIVE_CODEX_HARNESS_IMAGE_PROBE=1`
+- Optional MCP/tool probe: `OPENCLAW_LIVE_CODEX_HARNESS_MCP_PROBE=1`
+- The smoke sets `OPENCLAW_AGENT_HARNESS_FALLBACK=none` so a broken Codex
+  harness cannot pass by silently falling back to PI.
+- Auth: `OPENAI_API_KEY` from the shell/profile, plus optional copied
+  `~/.codex/auth.json` and `~/.codex/config.toml`
+
+Local recipe:
+
+```bash
+source ~/.profile
+OPENCLAW_LIVE_CODEX_HARNESS=1 \
+  OPENCLAW_LIVE_CODEX_HARNESS_IMAGE_PROBE=1 \
+  OPENCLAW_LIVE_CODEX_HARNESS_MCP_PROBE=1 \
+  OPENCLAW_LIVE_CODEX_HARNESS_MODEL=codex/gpt-5.4 \
+  pnpm test:live -- src/gateway/gateway-codex-harness.live.test.ts
+```
+
+Docker recipe:
+
+```bash
+source ~/.profile
+pnpm test:docker:live-codex-harness
+```
+
+Docker notes:
+
+- The Docker runner lives at `scripts/test-live-codex-harness-docker.sh`.
+- It sources the mounted `~/.profile`, passes `OPENAI_API_KEY`, copies Codex CLI
+  auth files when present, installs `@openai/codex` into a writable mounted npm
+  prefix, stages the source tree, then runs only the Codex-harness live test.
+- Docker enables the image and MCP/tool probes by default. Set
+  `OPENCLAW_LIVE_CODEX_HARNESS_IMAGE_PROBE=0` or
+  `OPENCLAW_LIVE_CODEX_HARNESS_MCP_PROBE=0` when you need a narrower debug run.
+- Docker also exports `OPENCLAW_AGENT_HARNESS_FALLBACK=none`, matching the live
+  test config so `openai-codex/*` or PI fallback cannot hide a Codex harness
+  regression.
+
 ### Recommended live recipes
 
 Narrow, explicit allowlists are fastest and least flaky:
@@ -588,6 +691,7 @@ The live-model Docker runners also bind-mount only the needed CLI auth homes (or
 - Direct models: `pnpm test:docker:live-models` (script: `scripts/test-live-models-docker.sh`)
 - ACP bind smoke: `pnpm test:docker:live-acp-bind` (script: `scripts/test-live-acp-bind-docker.sh`)
 - CLI backend smoke: `pnpm test:docker:live-cli-backend` (script: `scripts/test-live-cli-backend-docker.sh`)
+- Codex app-server harness smoke: `pnpm test:docker:live-codex-harness` (script: `scripts/test-live-codex-harness-docker.sh`)
 - Gateway + dev agent: `pnpm test:docker:live-gateway` (script: `scripts/test-live-gateway-models-docker.sh`)
 - Open WebUI live smoke: `pnpm test:docker:openwebui` (script: `scripts/e2e/openwebui-docker.sh`)
 - Onboarding wizard (TTY, full scaffolding): `pnpm test:docker:onboard` (script: `scripts/e2e/onboard-docker.sh`)
@@ -645,6 +749,7 @@ Useful env vars:
   - Override manually with `OPENCLAW_DOCKER_AUTH_DIRS=all`, `OPENCLAW_DOCKER_AUTH_DIRS=none`, or a comma list like `OPENCLAW_DOCKER_AUTH_DIRS=.claude,.codex`
 - `OPENCLAW_LIVE_GATEWAY_MODELS=...` / `OPENCLAW_LIVE_MODELS=...` to narrow the run
 - `OPENCLAW_LIVE_GATEWAY_PROVIDERS=...` / `OPENCLAW_LIVE_PROVIDERS=...` to filter providers in-container
+- `OPENCLAW_SKIP_DOCKER_BUILD=1` to reuse an existing `openclaw:local-live` image for reruns that do not need a rebuild
 - `OPENCLAW_LIVE_REQUIRE_PROFILE_KEYS=1` to ensure creds come from the profile store (not env)
 - `OPENCLAW_OPENWEBUI_MODEL=...` to choose the model exposed by the gateway for the Open WebUI smoke
 - `OPENCLAW_OPENWEBUI_PROMPT=...` to override the nonce-check prompt used by the Open WebUI smoke

docs/help/troubleshooting.md

@@ -251,18 +251,19 @@ flowchart TD
 
     Common log signatures:
 
-- `cron: scheduler disabled; jobs will not run automatically` → cron is disabled.
-- `heartbeat skipped` with `reason=quiet-hours` → outside configured active hours.
-- `heartbeat skipped` with `reason=empty-heartbeat-file` → `HEARTBEAT.md` exists but only contains blank/header-only scaffolding.
-- `heartbeat skipped` with `reason=no-tasks-due` → `HEARTBEAT.md` task mode is active but none of the task intervals are due yet.
-- `heartbeat skipped` with `reason=alerts-disabled` → all heartbeat visibility is disabled (`showOk`, `showAlerts`, and `useIndicator` are all off).
-- `requests-in-flight` → main lane busy; heartbeat wake was deferred. - `unknown accountId` → heartbeat delivery target account does not exist.
+    - `cron: scheduler disabled; jobs will not run automatically` → cron is disabled.
+    - `heartbeat skipped` with `reason=quiet-hours` → outside configured active hours.
+    - `heartbeat skipped` with `reason=empty-heartbeat-file` → `HEARTBEAT.md` exists but only contains blank/header-only scaffolding.
+    - `heartbeat skipped` with `reason=no-tasks-due` → `HEARTBEAT.md` task mode is active but none of the task intervals are due yet.
+    - `heartbeat skipped` with `reason=alerts-disabled` → all heartbeat visibility is disabled (`showOk`, `showAlerts`, and `useIndicator` are all off).
+    - `requests-in-flight` → main lane busy; heartbeat wake was deferred.
+    - `unknown accountId` → heartbeat delivery target account does not exist.
 
-      Deep pages:
+    Deep pages:
 
-      - [/gateway/troubleshooting#cron-and-heartbeat-delivery](/gateway/troubleshooting#cron-and-heartbeat-delivery)
-      - [/automation/cron-jobs#troubleshooting](/automation/cron-jobs#troubleshooting)
-      - [/gateway/heartbeat](/gateway/heartbeat)
+    - [/gateway/troubleshooting#cron-and-heartbeat-delivery](/gateway/troubleshooting#cron-and-heartbeat-delivery)
+    - [/automation/cron-jobs#troubleshooting](/automation/cron-jobs#troubleshooting)
+    - [/gateway/heartbeat](/gateway/heartbeat)
 
     </Accordion>
 
@@ -338,7 +339,7 @@ flowchart TD
 
       - [/tools/exec](/tools/exec)
       - [/tools/exec-approvals](/tools/exec-approvals)
-      - [/gateway/security#runtime-expectation-drift](/gateway/security#runtime-expectation-drift)
+      - [/gateway/security#what-the-audit-checks-high-level](/gateway/security#what-the-audit-checks-high-level)
 
     </Accordion>
 
@@ -376,6 +377,7 @@ flowchart TD
       - [/tools/browser-wsl2-windows-remote-cdp-troubleshooting](/tools/browser-wsl2-windows-remote-cdp-troubleshooting)
 
     </Accordion>
+
   </AccordionGroup>
 
 ## Related

docs/plans/2026-04-05-002-refactor-real-plugin-sdk-workspace-plan.md

@@ -1,580 +0,0 @@
----
-title: "refactor: Make plugin-sdk a real workspace package incrementally"
-type: refactor
-status: active
-date: 2026-04-05
----
-
-# refactor: Make plugin-sdk a real workspace package incrementally
-
-## Overview
-
-This plan introduces a real workspace package for the plugin SDK at
-`packages/plugin-sdk` and uses it to opt in a small first wave of extensions to
-compiler-enforced package boundaries. The goal is to make illegal relative
-imports fail under normal `tsc` for a selected set of bundled provider
-extensions, without forcing a repo-wide migration or a giant merge-conflict
-surface.
-
-The key incremental move is to run two modes in parallel for a while:
-
-| Mode        | Import shape             | Who uses it                          | Enforcement                                  |
-| ----------- | ------------------------ | ------------------------------------ | -------------------------------------------- |
-| Legacy mode | `openclaw/plugin-sdk/*`  | all existing non-opted-in extensions | current permissive behavior remains          |
-| Opt-in mode | `@openclaw/plugin-sdk/*` | first-wave extensions only           | package-local `rootDir` + project references |
-
-## Problem Frame
-
-The current repo exports a large public plugin SDK surface, but it is not a real
-workspace package. Instead:
-
-- root `tsconfig.json` maps `openclaw/plugin-sdk/*` directly to
-  `src/plugin-sdk/*.ts`
-- extensions that were not opted into the previous experiment still share that
-  global source-alias behavior
-- adding `rootDir` only works when allowed SDK imports stop resolving into raw
-  repo source
-
-That means the repo can describe the desired boundary policy, but TypeScript
-does not enforce it cleanly for most extensions.
-
-You want an incremental path that:
-
-- makes `plugin-sdk` real
-- moves the SDK toward a workspace package named `@openclaw/plugin-sdk`
-- changes only about 10 extensions in the first PR
-- leaves the rest of the extension tree on the old scheme until later cleanup
-- avoids the `tsconfig.plugin-sdk.dts.json` + postinstall-generated declaration
-  workflow as the primary mechanism for the first-wave rollout
-
-## Requirements Trace
-
-- R1. Create a real workspace package for the plugin SDK under `packages/`.
-- R2. Name the new package `@openclaw/plugin-sdk`.
-- R3. Give the new SDK package its own `package.json` and `tsconfig.json`.
-- R4. Keep legacy `openclaw/plugin-sdk/*` imports working for non-opted-in
-  extensions during the migration window.
-- R5. Opt in only a small first wave of extensions in the first PR.
-- R6. The first-wave extensions must fail closed for relative imports that leave
-  their package root.
-- R7. The first-wave extensions must consume the SDK through a package
-  dependency and a TS project reference, not through root `paths` aliases.
-- R8. The plan must avoid a repo-wide mandatory postinstall generation step for
-  editor correctness.
-- R9. The first-wave rollout must be reviewable and mergeable as a moderate PR,
-  not a repo-wide 300+ file refactor.
-
-## Scope Boundaries
-
-- No full migration of all bundled extensions in the first PR.
-- No requirement to delete `src/plugin-sdk` in the first PR.
-- No requirement to rewire every root build or test path to use the new package
-  immediately.
-- No attempt to force VS Code squiggles for every non-opted-in extension.
-- No broad lint cleanup for the rest of the extension tree.
-- No large runtime behavior changes beyond import resolution, package ownership,
-  and boundary enforcement for the opted-in extensions.
-
-## Context & Research
-
-### Relevant Code and Patterns
-
-- `pnpm-workspace.yaml` already includes `packages/*` and `extensions/*`, so a
-  new workspace package under `packages/plugin-sdk` fits the existing repo
-  layout.
-- Existing workspace packages such as `packages/memory-host-sdk/package.json`
-  and `packages/plugin-package-contract/package.json` already use package-local
-  `exports` maps rooted in `src/*.ts`.
-- Root `package.json` currently publishes the SDK surface through `./plugin-sdk`
-  and `./plugin-sdk/*` exports backed by `dist/plugin-sdk/*.js` and
-  `dist/plugin-sdk/*.d.ts`.
-- `src/plugin-sdk/entrypoints.ts` and `scripts/lib/plugin-sdk-entrypoints.json`
-  already act as the canonical entrypoint inventory for the SDK surface.
-- Root `tsconfig.json` currently maps:
-  - `openclaw/plugin-sdk` -> `src/plugin-sdk/index.ts`
-  - `openclaw/plugin-sdk/*` -> `src/plugin-sdk/*.ts`
-- The previous boundary experiment showed that package-local `rootDir` works for
-  illegal relative imports only after allowed SDK imports stop resolving to raw
-  source outside the extension package.
-
-### First-Wave Extension Set
-
-This plan assumes the first wave is the provider-heavy set that is least likely
-to drag in complex channel-runtime edge cases:
-
-- `extensions/anthropic`
-- `extensions/exa`
-- `extensions/firecrawl`
-- `extensions/groq`
-- `extensions/mistral`
-- `extensions/openai`
-- `extensions/perplexity`
-- `extensions/tavily`
-- `extensions/together`
-- `extensions/xai`
-
-### First-Wave SDK Surface Inventory
-
-The first-wave extensions currently import a manageable subset of SDK subpaths.
-The initial `@openclaw/plugin-sdk` package only needs to cover these:
-
-- `agent-runtime`
-- `cli-runtime`
-- `config-runtime`
-- `core`
-- `image-generation`
-- `media-runtime`
-- `media-understanding`
-- `plugin-entry`
-- `plugin-runtime`
-- `provider-auth`
-- `provider-auth-api-key`
-- `provider-auth-login`
-- `provider-auth-runtime`
-- `provider-catalog-shared`
-- `provider-entry`
-- `provider-http`
-- `provider-model-shared`
-- `provider-onboard`
-- `provider-stream-family`
-- `provider-stream-shared`
-- `provider-tools`
-- `provider-usage`
-- `provider-web-fetch`
-- `provider-web-search`
-- `realtime-transcription`
-- `realtime-voice`
-- `runtime-env`
-- `secret-input`
-- `security-runtime`
-- `speech`
-- `testing`
-
-### Institutional Learnings
-
-- No relevant `docs/solutions/` entries were present in this worktree.
-
-### External References
-
-- No external research was needed for this plan. The repo already contains the
-  relevant workspace-package and SDK-export patterns.
-
-## Key Technical Decisions
-
-- Introduce `@openclaw/plugin-sdk` as a new workspace package while keeping the
-  legacy root `openclaw/plugin-sdk/*` surface alive during migration.
-  Rationale: this lets a first-wave extension set move onto real package
-  resolution without forcing every extension and every root build path to change
-  at once.
-
-- Use a dedicated opt-in boundary base config such as
-  `extensions/tsconfig.package-boundary.base.json` instead of replacing the
-  existing extension base for everyone.
-  Rationale: the repo needs to support both legacy and opt-in extension modes
-  simultaneously during migration.
-
-- Use TS project references from first-wave extensions to
-  `packages/plugin-sdk/tsconfig.json` and set
-  `disableSourceOfProjectReferenceRedirect` for the opt-in boundary mode.
-  Rationale: this gives `tsc` a real package graph while discouraging editor and
-  compiler fallback to raw source traversal.
-
-- Keep `@openclaw/plugin-sdk` private in the first wave.
-  Rationale: the immediate goal is internal boundary enforcement and migration
-  safety, not publishing a second external SDK contract before the surface is
-  stable.
-
-- Move only the first-wave SDK subpaths in the first implementation slice, and
-  keep compatibility bridges for the rest.
-  Rationale: physically moving all 315 `src/plugin-sdk/*.ts` files in one PR is
-  exactly the merge-conflict surface this plan is trying to avoid.
-
-- Do not rely on `scripts/postinstall-bundled-plugins.mjs` to build SDK
-  declarations for the first wave.
-  Rationale: explicit build/reference flows are easier to reason about and keep
-  repo behavior more predictable.
-
-## Open Questions
-
-### Resolved During Planning
-
-- Which extensions should be in the first wave?
-  Use the 10 provider/web-search extensions listed above because they are more
-  structurally isolated than the heavier channel packages.
-
-- Should the first PR replace the entire extension tree?
-  No. The first PR should support two modes in parallel and only opt in the
-  first wave.
-
-- Should the first wave require a postinstall declaration build?
-  No. The package/reference graph should be explicit, and CI should run the
-  relevant package-local typecheck intentionally.
-
-### Deferred to Implementation
-
-- Whether the first-wave package can point directly at package-local `src/*.ts`
-  via project references alone, or whether a small declaration-emission step is
-  still required for the `@openclaw/plugin-sdk` package.
-  This is an implementation-owned TS graph validation question.
-
-- Whether the root `openclaw` package should proxy first-wave SDK subpaths to
-  `packages/plugin-sdk` outputs immediately or continue using generated
-  compatibility shims under `src/plugin-sdk`.
-  This is a compatibility and build-shape detail that depends on the minimal
-  implementation path that keeps CI green.
-
-## High-Level Technical Design
-
-> This illustrates the intended approach and is directional guidance for review, not implementation specification. The implementing agent should treat it as context, not code to reproduce.
-
-```mermaid
-flowchart TB
-  subgraph Legacy["Legacy extensions (unchanged)"]
-    L1["extensions/*\nopenclaw/plugin-sdk/*"]
-    L2["root tsconfig paths"]
-    L1 --> L2
-    L2 --> L3["src/plugin-sdk/*"]
-  end
-
-  subgraph OptIn["First-wave extensions"]
-    O1["10 opted-in extensions"]
-    O2["extensions/tsconfig.package-boundary.base.json"]
-    O3["rootDir = '.'\nproject reference"]
-    O4["@openclaw/plugin-sdk"]
-    O1 --> O2
-    O2 --> O3
-    O3 --> O4
-  end
-
-  subgraph SDK["New workspace package"]
-    P1["packages/plugin-sdk/package.json"]
-    P2["packages/plugin-sdk/tsconfig.json"]
-    P3["packages/plugin-sdk/src/<first-wave-subpaths>.ts"]
-    P1 --> P2
-    P2 --> P3
-  end
-
-  O4 --> SDK
-```
-
-## Implementation Units
-
-- [ ] **Unit 1: Introduce the real `@openclaw/plugin-sdk` workspace package**
-
-**Goal:** Create a real workspace package for the SDK that can own the
-first-wave subpath surface without forcing a repo-wide migration.
-
-**Requirements:** R1, R2, R3, R8, R9
-
-**Dependencies:** None
-
-**Files:**
-
-- Create: `packages/plugin-sdk/package.json`
-- Create: `packages/plugin-sdk/tsconfig.json`
-- Create: `packages/plugin-sdk/src/index.ts`
-- Create: `packages/plugin-sdk/src/*.ts` for the first-wave SDK subpaths
-- Modify: `pnpm-workspace.yaml` only if package-glob adjustments are needed
-- Modify: `package.json`
-- Modify: `src/plugin-sdk/entrypoints.ts`
-- Modify: `scripts/lib/plugin-sdk-entrypoints.json`
-- Test: `src/plugins/contracts/plugin-sdk-workspace-package.contract.test.ts`
-
-**Approach:**
-
-- Add a new workspace package named `@openclaw/plugin-sdk`.
-- Start with the first-wave SDK subpaths only, not the entire 315-file tree.
-- If directly moving a first-wave entrypoint would create an oversized diff, the
-  first PR may introduce that subpath in `packages/plugin-sdk/src` as a thin
-  package wrapper first and then flip the source of truth to the package in a
-  follow-up PR for that subpath cluster.
-- Reuse the existing entrypoint inventory machinery so the first-wave package
-  surface is declared in one canonical place.
-- Keep the root package exports alive for legacy users while the workspace
-  package becomes the new opt-in contract.
-
-**Patterns to follow:**
-
-- `packages/memory-host-sdk/package.json`
-- `packages/plugin-package-contract/package.json`
-- `src/plugin-sdk/entrypoints.ts`
-
-**Test scenarios:**
-
-- Happy path: the workspace package exports every first-wave subpath listed in
-  the plan and no required first-wave export is missing.
-- Edge case: package export metadata remains stable when the first-wave entry
-  list is re-generated or compared against the canonical inventory.
-- Integration: root package legacy SDK exports remain present after introducing
-  the new workspace package.
-
-**Verification:**
-
-- The repo contains a valid `@openclaw/plugin-sdk` workspace package with a
-  stable first-wave export map and no legacy export regression in root
-  `package.json`.
-
-- [ ] **Unit 2: Add an opt-in TS boundary mode for package-enforced extensions**
-
-**Goal:** Define the TS configuration mode that opted-in extensions will use,
-while leaving the existing extension TS behavior unchanged for everyone else.
-
-**Requirements:** R4, R6, R7, R8, R9
-
-**Dependencies:** Unit 1
-
-**Files:**
-
-- Create: `extensions/tsconfig.package-boundary.base.json`
-- Create: `tsconfig.boundary-optin.json`
-- Modify: `extensions/xai/tsconfig.json`
-- Modify: `extensions/openai/tsconfig.json`
-- Modify: `extensions/anthropic/tsconfig.json`
-- Modify: `extensions/mistral/tsconfig.json`
-- Modify: `extensions/groq/tsconfig.json`
-- Modify: `extensions/together/tsconfig.json`
-- Modify: `extensions/perplexity/tsconfig.json`
-- Modify: `extensions/tavily/tsconfig.json`
-- Modify: `extensions/exa/tsconfig.json`
-- Modify: `extensions/firecrawl/tsconfig.json`
-- Test: `src/plugins/contracts/extension-package-project-boundaries.test.ts`
-- Test: `test/extension-package-tsc-boundary.test.ts`
-
-**Approach:**
-
-- Leave `extensions/tsconfig.base.json` in place for legacy extensions.
-- Add a new opt-in base config that:
-  - sets `rootDir: "."`
-  - references `packages/plugin-sdk`
-  - enables `composite`
-  - disables project-reference source redirect when needed
-- Add a dedicated solution config for the first-wave typecheck graph instead of
-  reshaping the root repo TS project in the same PR.
-
-**Execution note:** Start with a failing package-local canary typecheck for one
-opted-in extension before applying the pattern to all 10.
-
-**Patterns to follow:**
-
-- Existing package-local extension `tsconfig.json` pattern from the prior
-  boundary work
-- Workspace package pattern from `packages/memory-host-sdk`
-
-**Test scenarios:**
-
-- Happy path: each opted-in extension typechecks successfully through the
-  package-boundary TS config.
-- Error path: a canary relative import from `../../src/cli/acp-cli.ts` fails
-  with `TS6059` for an opted-in extension.
-- Integration: non-opted-in extensions remain untouched and do not need to
-  participate in the new solution config.
-
-**Verification:**
-
-- There is a dedicated typecheck graph for the 10 opted-in extensions, and bad
-  relative imports from one of them fail through normal `tsc`.
-
-- [ ] **Unit 3: Migrate the first-wave extensions onto `@openclaw/plugin-sdk`**
-
-**Goal:** Change the first-wave extensions to consume the real SDK package
-through dependency metadata, project references, and package-name imports.
-
-**Requirements:** R5, R6, R7, R9
-
-**Dependencies:** Unit 2
-
-**Files:**
-
-- Modify: `extensions/anthropic/package.json`
-- Modify: `extensions/exa/package.json`
-- Modify: `extensions/firecrawl/package.json`
-- Modify: `extensions/groq/package.json`
-- Modify: `extensions/mistral/package.json`
-- Modify: `extensions/openai/package.json`
-- Modify: `extensions/perplexity/package.json`
-- Modify: `extensions/tavily/package.json`
-- Modify: `extensions/together/package.json`
-- Modify: `extensions/xai/package.json`
-- Modify: production and test imports under each of the 10 extension roots that
-  currently reference `openclaw/plugin-sdk/*`
-
-**Approach:**
-
-- Add `@openclaw/plugin-sdk: workspace:*` to the first-wave extension
-  `devDependencies`.
-- Replace `openclaw/plugin-sdk/*` imports in those packages with
-  `@openclaw/plugin-sdk/*`.
-- Keep local extension-internal imports on local barrels such as `./api.ts` and
-  `./runtime-api.ts`.
-- Do not change non-opted-in extensions in this PR.
-
-**Patterns to follow:**
-
-- Existing extension-local import barrels (`api.ts`, `runtime-api.ts`)
-- Package dependency shape used by other `@openclaw/*` workspace packages
-
-**Test scenarios:**
-
-- Happy path: each migrated extension still registers/loads through its existing
-  plugin tests after the import rewrite.
-- Edge case: test-only SDK imports in the opted-in extension set still resolve
-  correctly through the new package.
-- Integration: migrated extensions do not require root `openclaw/plugin-sdk/*`
-  aliases for typechecking.
-
-**Verification:**
-
-- The first-wave extensions build and test against `@openclaw/plugin-sdk`
-  without needing the legacy root SDK alias path.
-
-- [ ] **Unit 4: Preserve legacy compatibility while the migration is partial**
-
-**Goal:** Keep the rest of the repo working while the SDK exists in both legacy
-and new-package forms during migration.
-
-**Requirements:** R4, R8, R9
-
-**Dependencies:** Units 1-3
-
-**Files:**
-
-- Modify: `src/plugin-sdk/*.ts` for first-wave compatibility shims as needed
-- Modify: `package.json`
-- Modify: build or export plumbing that assembles SDK artifacts
-- Test: `src/plugins/contracts/plugin-sdk-runtime-api-guardrails.test.ts`
-- Test: `src/plugins/contracts/plugin-sdk-index.bundle.test.ts`
-
-**Approach:**
-
-- Keep root `openclaw/plugin-sdk/*` as the compatibility surface for legacy
-  extensions and for external consumers that are not moving yet.
-- Use either generated shims or root-export proxy wiring for the first-wave
-  subpaths that have moved into `packages/plugin-sdk`.
-- Do not attempt to retire the root SDK surface in this phase.
-
-**Patterns to follow:**
-
-- Existing root SDK export generation via `src/plugin-sdk/entrypoints.ts`
-- Existing package export compatibility in root `package.json`
-
-**Test scenarios:**
-
-- Happy path: a legacy root SDK import still resolves for a non-opted-in
-  extension after the new package exists.
-- Edge case: a first-wave subpath works through both the legacy root surface and
-  the new package surface during the migration window.
-- Integration: plugin-sdk index/bundle contract tests continue to see a coherent
-  public surface.
-
-**Verification:**
-
-- The repo supports both legacy and opt-in SDK consumption modes without
-  breaking unchanged extensions.
-
-- [ ] **Unit 5: Add scoped enforcement and document the migration contract**
-
-**Goal:** Land CI and contributor guidance that enforce the new behavior for the
-first wave without pretending the entire extension tree is migrated.
-
-**Requirements:** R5, R6, R8, R9
-
-**Dependencies:** Units 1-4
-
-**Files:**
-
-- Modify: `package.json`
-- Modify: CI workflow files that should run the opt-in boundary typecheck
-- Modify: `AGENTS.md`
-- Modify: `docs/plugins/sdk-overview.md`
-- Modify: `docs/plugins/sdk-entrypoints.md`
-- Modify: `docs/plans/2026-04-05-001-refactor-extension-package-resolution-boundary-plan.md`
-
-**Approach:**
-
-- Add an explicit first-wave gate, such as a dedicated `tsc -b` solution run for
-  `packages/plugin-sdk` plus the 10 opted-in extensions.
-- Document that the repo now supports both legacy and opt-in extension modes,
-  and that new extension boundary work should prefer the new package route.
-- Record the next-wave migration rule so later PRs can add more extensions
-  without re-litigating the architecture.
-
-**Patterns to follow:**
-
-- Existing contract tests under `src/plugins/contracts/`
-- Existing docs updates that explain staged migrations
-
-**Test scenarios:**
-
-- Happy path: the new first-wave typecheck gate passes for the workspace package
-  and the opted-in extensions.
-- Error path: introducing a new illegal relative import in an opted-in
-  extension fails the scoped typecheck gate.
-- Integration: CI does not require non-opted-in extensions to satisfy the new
-  package-boundary mode yet.
-
-**Verification:**
-
-- The first-wave enforcement path is documented, tested, and runnable without
-  forcing the entire extension tree to migrate.
-
-## System-Wide Impact
-
-- **Interaction graph:** this work touches the SDK source-of-truth, root package
-  exports, extension package metadata, TS graph layout, and CI verification.
-- **Error propagation:** the main intended failure mode becomes compile-time TS
-  errors (`TS6059`) in opted-in extensions instead of custom script-only
-  failures.
-- **State lifecycle risks:** dual-surface migration introduces drift risk between
-  root compatibility exports and the new workspace package.
-- **API surface parity:** first-wave subpaths must remain semantically identical
-  through both `openclaw/plugin-sdk/*` and `@openclaw/plugin-sdk/*` during the
-  transition.
-- **Integration coverage:** unit tests are not enough; scoped package-graph
-  typechecks are required to prove the boundary.
-- **Unchanged invariants:** non-opted-in extensions keep their current behavior
-  in PR 1. This plan does not claim repo-wide import-boundary enforcement.
-
-## Risks & Dependencies
-
-| Risk                                                                                                   | Mitigation                                                                                                              |
-| ------------------------------------------------------------------------------------------------------ | ----------------------------------------------------------------------------------------------------------------------- |
-| The first-wave package still resolves back into raw source and `rootDir` does not actually fail closed | Make the first implementation step a package-reference canary on one opted-in extension before widening to the full set |
-| Moving too much SDK source at once recreates the original merge-conflict problem                       | Move only the first-wave subpaths in the first PR and keep root compatibility bridges                                   |
-| Legacy and new SDK surfaces drift semantically                                                         | Keep a single entrypoint inventory, add compatibility contract tests, and make dual-surface parity explicit             |
-| Root repo build/test paths accidentally start depending on the new package in uncontrolled ways        | Use a dedicated opt-in solution config and keep root-wide TS topology changes out of the first PR                       |
-
-## Phased Delivery
-
-### Phase 1
-
-- Introduce `@openclaw/plugin-sdk`
-- Define the first-wave subpath surface
-- Prove one opted-in extension can fail closed through `rootDir`
-
-### Phase 2
-
-- Opt in the 10 first-wave extensions
-- Keep root compatibility alive for everyone else
-
-### Phase 3
-
-- Add more extensions in later PRs
-- Move more SDK subpaths into the workspace package
-- Retire root compatibility only after the legacy extension set is gone
-
-## Documentation / Operational Notes
-
-- The first PR should explicitly describe itself as a dual-mode migration, not a
-  repo-wide enforcement completion.
-- The migration guide should make it easy for later PRs to add more extensions
-  by following the same package/dependency/reference pattern.
-
-## Sources & References
-
-- Prior plan: `docs/plans/2026-04-05-001-refactor-extension-package-resolution-boundary-plan.md`
-- Workspace config: `pnpm-workspace.yaml`
-- Existing SDK entrypoint inventory: `src/plugin-sdk/entrypoints.ts`
-- Existing root SDK exports: `package.json`
-- Existing workspace package patterns:
-  - `packages/memory-host-sdk/package.json`
-  - `packages/plugin-package-contract/package.json`

docs/plugins/codex-harness.md

@@ -0,0 +1,489 @@
+---
+title: "Codex Harness"
+summary: "Run OpenClaw embedded agent turns through the bundled Codex app-server harness"
+read_when:
+  - You want to use the bundled Codex app-server harness
+  - You need Codex model refs and config examples
+  - You want to disable PI fallback for Codex-only deployments
+---
+
+# Codex Harness
+
+The bundled `codex` plugin lets OpenClaw run embedded agent turns through the
+Codex app-server instead of the built-in PI harness.
+
+Use this when you want Codex to own the low-level agent session: model
+discovery, native thread resume, native compaction, and app-server execution.
+OpenClaw still owns chat channels, session files, model selection, tools,
+approvals, media delivery, and the visible transcript mirror.
+
+The harness is off by default. It is selected only when the `codex` plugin is
+enabled and the resolved model is a `codex/*` model, or when you explicitly
+force `embeddedHarness.runtime: "codex"` or `OPENCLAW_AGENT_RUNTIME=codex`.
+If you never configure `codex/*`, existing PI, OpenAI, Anthropic, Gemini, local,
+and custom-provider runs keep their current behavior.
+
+## Pick the right model prefix
+
+OpenClaw has separate routes for OpenAI and Codex-shaped access:
+
+| Model ref              | Runtime path                                 | Use when                                                                |
+| ---------------------- | -------------------------------------------- | ----------------------------------------------------------------------- |
+| `openai/gpt-5.4`       | OpenAI provider through OpenClaw/PI plumbing | You want direct OpenAI Platform API access with `OPENAI_API_KEY`.       |
+| `openai-codex/gpt-5.4` | OpenAI Codex OAuth provider through PI       | You want ChatGPT/Codex OAuth without the Codex app-server harness.      |
+| `codex/gpt-5.4`        | Bundled Codex provider plus Codex harness    | You want native Codex app-server execution for the embedded agent turn. |
+
+The Codex harness only claims `codex/*` model refs. Existing `openai/*`,
+`openai-codex/*`, Anthropic, Gemini, xAI, local, and custom provider refs keep
+their normal paths.
+
+## Requirements
+
+- OpenClaw with the bundled `codex` plugin available.
+- Codex app-server `0.118.0` or newer.
+- Codex auth available to the app-server process.
+
+The plugin blocks older or unversioned app-server handshakes. That keeps
+OpenClaw on the protocol surface it has been tested against.
+
+For live and Docker smoke tests, auth usually comes from `OPENAI_API_KEY`, plus
+optional Codex CLI files such as `~/.codex/auth.json` and
+`~/.codex/config.toml`. Use the same auth material your local Codex app-server
+uses.
+
+## Minimal config
+
+Use `codex/gpt-5.4`, enable the bundled plugin, and force the `codex` harness:
+
+```json5
+{
+  plugins: {
+    entries: {
+      codex: {
+        enabled: true,
+      },
+    },
+  },
+  agents: {
+    defaults: {
+      model: "codex/gpt-5.4",
+      embeddedHarness: {
+        runtime: "codex",
+        fallback: "none",
+      },
+    },
+  },
+}
+```
+
+If your config uses `plugins.allow`, include `codex` there too:
+
+```json5
+{
+  plugins: {
+    allow: ["codex"],
+    entries: {
+      codex: {
+        enabled: true,
+      },
+    },
+  },
+}
+```
+
+Setting `agents.defaults.model` or an agent model to `codex/<model>` also
+auto-enables the bundled `codex` plugin. The explicit plugin entry is still
+useful in shared configs because it makes the deployment intent obvious.
+
+## Add Codex without replacing other models
+
+Keep `runtime: "auto"` when you want Codex for `codex/*` models and PI for
+everything else:
+
+```json5
+{
+  plugins: {
+    entries: {
+      codex: {
+        enabled: true,
+      },
+    },
+  },
+  agents: {
+    defaults: {
+      model: {
+        primary: "codex/gpt-5.4",
+        fallbacks: ["openai/gpt-5.4", "anthropic/claude-opus-4-6"],
+      },
+      models: {
+        "codex/gpt-5.4": { alias: "codex" },
+        "codex/gpt-5.4-mini": { alias: "codex-mini" },
+        "openai/gpt-5.4": { alias: "gpt" },
+        "anthropic/claude-opus-4-6": { alias: "opus" },
+      },
+      embeddedHarness: {
+        runtime: "auto",
+        fallback: "pi",
+      },
+    },
+  },
+}
+```
+
+With this shape:
+
+- `/model codex` or `/model codex/gpt-5.4` uses the Codex app-server harness.
+- `/model gpt` or `/model openai/gpt-5.4` uses the OpenAI provider path.
+- `/model opus` uses the Anthropic provider path.
+- If a non-Codex model is selected, PI remains the compatibility harness.
+
+## Codex-only deployments
+
+Disable PI fallback when you need to prove that every embedded agent turn uses
+the Codex harness:
+
+```json5
+{
+  agents: {
+    defaults: {
+      model: "codex/gpt-5.4",
+      embeddedHarness: {
+        runtime: "codex",
+        fallback: "none",
+      },
+    },
+  },
+}
+```
+
+Environment override:
+
+```bash
+OPENCLAW_AGENT_RUNTIME=codex \
+OPENCLAW_AGENT_HARNESS_FALLBACK=none \
+openclaw gateway run
+```
+
+With fallback disabled, OpenClaw fails early if the Codex plugin is disabled,
+the requested model is not a `codex/*` ref, the app-server is too old, or the
+app-server cannot start.
+
+## Per-agent Codex
+
+You can make one agent Codex-only while the default agent keeps normal
+auto-selection:
+
+```json5
+{
+  agents: {
+    defaults: {
+      embeddedHarness: {
+        runtime: "auto",
+        fallback: "pi",
+      },
+    },
+    list: [
+      {
+        id: "main",
+        default: true,
+        model: "anthropic/claude-opus-4-6",
+      },
+      {
+        id: "codex",
+        name: "Codex",
+        model: "codex/gpt-5.4",
+        embeddedHarness: {
+          runtime: "codex",
+          fallback: "none",
+        },
+      },
+    ],
+  },
+}
+```
+
+Use normal session commands to switch agents and models. `/new` creates a fresh
+OpenClaw session and the Codex harness creates or resumes its sidecar app-server
+thread as needed. `/reset` clears the OpenClaw session binding for that thread.
+
+## Model discovery
+
+By default, the Codex plugin asks the app-server for available models. If
+discovery fails or times out, it uses the bundled fallback catalog:
+
+- `codex/gpt-5.4`
+- `codex/gpt-5.4-mini`
+- `codex/gpt-5.2`
+
+You can tune discovery under `plugins.entries.codex.config.discovery`:
+
+```json5
+{
+  plugins: {
+    entries: {
+      codex: {
+        enabled: true,
+        config: {
+          discovery: {
+            enabled: true,
+            timeoutMs: 2500,
+          },
+        },
+      },
+    },
+  },
+}
+```
+
+Disable discovery when you want startup to avoid probing Codex and stick to the
+fallback catalog:
+
+```json5
+{
+  plugins: {
+    entries: {
+      codex: {
+        enabled: true,
+        config: {
+          discovery: {
+            enabled: false,
+          },
+        },
+      },
+    },
+  },
+}
+```
+
+## App-server connection and policy
+
+By default, the plugin starts Codex locally with:
+
+```bash
+codex app-server --listen stdio://
+```
+
+You can keep that default and only tune Codex native policy:
+
+```json5
+{
+  plugins: {
+    entries: {
+      codex: {
+        enabled: true,
+        config: {
+          appServer: {
+            approvalPolicy: "on-request",
+            sandbox: "workspace-write",
+            serviceTier: "priority",
+          },
+        },
+      },
+    },
+  },
+}
+```
+
+For an already-running app-server, use WebSocket transport:
+
+```json5
+{
+  plugins: {
+    entries: {
+      codex: {
+        enabled: true,
+        config: {
+          appServer: {
+            transport: "websocket",
+            url: "ws://127.0.0.1:39175",
+            authToken: "${CODEX_APP_SERVER_TOKEN}",
+            requestTimeoutMs: 60000,
+          },
+        },
+      },
+    },
+  },
+}
+```
+
+Supported `appServer` fields:
+
+| Field               | Default                                  | Meaning                                                                  |
+| ------------------- | ---------------------------------------- | ------------------------------------------------------------------------ |
+| `transport`         | `"stdio"`                                | `"stdio"` spawns Codex; `"websocket"` connects to `url`.                 |
+| `command`           | `"codex"`                                | Executable for stdio transport.                                          |
+| `args`              | `["app-server", "--listen", "stdio://"]` | Arguments for stdio transport.                                           |
+| `url`               | unset                                    | WebSocket app-server URL.                                                |
+| `authToken`         | unset                                    | Bearer token for WebSocket transport.                                    |
+| `headers`           | `{}`                                     | Extra WebSocket headers.                                                 |
+| `requestTimeoutMs`  | `60000`                                  | Timeout for app-server control-plane calls.                              |
+| `approvalPolicy`    | `"never"`                                | Native Codex approval policy sent to thread start/resume/turn.           |
+| `sandbox`           | `"workspace-write"`                      | Native Codex sandbox mode sent to thread start/resume.                   |
+| `approvalsReviewer` | `"user"`                                 | Use `"guardian_subagent"` to let Codex guardian review native approvals. |
+| `serviceTier`       | unset                                    | Optional Codex service tier, for example `"priority"`.                   |
+
+The older environment variables still work as fallbacks for local testing when
+the matching config field is unset:
+
+- `OPENCLAW_CODEX_APP_SERVER_BIN`
+- `OPENCLAW_CODEX_APP_SERVER_ARGS`
+- `OPENCLAW_CODEX_APP_SERVER_APPROVAL_POLICY`
+- `OPENCLAW_CODEX_APP_SERVER_SANDBOX`
+- `OPENCLAW_CODEX_APP_SERVER_GUARDIAN=1`
+
+Config is preferred for repeatable deployments.
+
+## Common recipes
+
+Local Codex with default stdio transport:
+
+```json5
+{
+  plugins: {
+    entries: {
+      codex: {
+        enabled: true,
+      },
+    },
+  },
+}
+```
+
+Codex-only harness validation, with PI fallback disabled:
+
+```json5
+{
+  embeddedHarness: {
+    fallback: "none",
+  },
+  plugins: {
+    entries: {
+      codex: {
+        enabled: true,
+      },
+    },
+  },
+}
+```
+
+Guardian-reviewed Codex approvals:
+
+```json5
+{
+  plugins: {
+    entries: {
+      codex: {
+        enabled: true,
+        config: {
+          appServer: {
+            approvalPolicy: "on-request",
+            approvalsReviewer: "guardian_subagent",
+            sandbox: "workspace-write",
+          },
+        },
+      },
+    },
+  },
+}
+```
+
+Remote app-server with explicit headers:
+
+```json5
+{
+  plugins: {
+    entries: {
+      codex: {
+        enabled: true,
+        config: {
+          appServer: {
+            transport: "websocket",
+            url: "ws://gateway-host:39175",
+            headers: {
+              "X-OpenClaw-Agent": "main",
+            },
+          },
+        },
+      },
+    },
+  },
+}
+```
+
+Model switching stays OpenClaw-controlled. When an OpenClaw session is attached
+to an existing Codex thread, the next turn sends the currently selected
+`codex/*` model, provider, approval policy, sandbox, and service tier to
+app-server again. Switching from `codex/gpt-5.4` to `codex/gpt-5.2` keeps the
+thread binding but asks Codex to continue with the newly selected model.
+
+## Codex command
+
+The bundled plugin registers `/codex` as an authorized slash command. It is
+generic and works on any channel that supports OpenClaw text commands.
+
+Common forms:
+
+- `/codex status` shows live app-server connectivity, models, account, rate limits, MCP servers, and skills.
+- `/codex models` lists live Codex app-server models.
+- `/codex threads [filter]` lists recent Codex threads.
+- `/codex resume <thread-id>` attaches the current OpenClaw session to an existing Codex thread.
+- `/codex compact` asks Codex app-server to compact the attached thread.
+- `/codex review` starts Codex native review for the attached thread.
+- `/codex account` shows account and rate-limit status.
+- `/codex mcp` lists Codex app-server MCP server status.
+- `/codex skills` lists Codex app-server skills.
+
+`/codex resume` writes the same sidecar binding file that the harness uses for
+normal turns. On the next message, OpenClaw resumes that Codex thread, passes the
+currently selected OpenClaw `codex/*` model into app-server, and keeps extended
+history enabled.
+
+The command surface requires Codex app-server `0.118.0` or newer. Individual
+control methods are reported as `unsupported by this Codex app-server` if a
+future or custom app-server does not expose that JSON-RPC method.
+
+## Tools, media, and compaction
+
+The Codex harness changes the low-level embedded agent executor only.
+
+OpenClaw still builds the tool list and receives dynamic tool results from the
+harness. Text, images, video, music, TTS, approvals, and messaging-tool output
+continue through the normal OpenClaw delivery path.
+
+When the selected model uses the Codex harness, native thread compaction is
+delegated to Codex app-server. OpenClaw keeps a transcript mirror for channel
+history, search, `/new`, `/reset`, and future model or harness switching. The
+mirror includes the user prompt, final assistant text, and lightweight Codex
+reasoning or plan records when the app-server emits them.
+
+Media generation does not require PI. Image, video, music, PDF, TTS, and media
+understanding continue to use the matching provider/model settings such as
+`agents.defaults.imageGenerationModel`, `videoGenerationModel`, `pdfModel`, and
+`messages.tts`.
+
+## Troubleshooting
+
+**Codex does not appear in `/model`:** enable `plugins.entries.codex.enabled`,
+set a `codex/*` model ref, or check whether `plugins.allow` excludes `codex`.
+
+**OpenClaw falls back to PI:** set `embeddedHarness.fallback: "none"` or
+`OPENCLAW_AGENT_HARNESS_FALLBACK=none` while testing.
+
+**The app-server is rejected:** upgrade Codex so the app-server handshake
+reports version `0.118.0` or newer.
+
+**Model discovery is slow:** lower `plugins.entries.codex.config.discovery.timeoutMs`
+or disable discovery.
+
+**WebSocket transport fails immediately:** check `appServer.url`, `authToken`,
+and that the remote app-server speaks the same Codex app-server protocol version.
+
+**A non-Codex model uses PI:** that is expected. The Codex harness only claims
+`codex/*` model refs.
+
+## Related
+
+- [Agent Harness Plugins](/plugins/sdk-agent-harness)
+- [Model Providers](/concepts/model-providers)
+- [Configuration Reference](/gateway/configuration-reference)
+- [Testing](/help/testing#live-codex-app-server-harness-smoke)

docs/plugins/manifest.md

@@ -147,6 +147,7 @@ Those belong in your plugin code and `package.json`.
 | `providers`                         | No       | `string[]`                       | Provider ids owned by this plugin.                                                                                                                                                                           |
 | `modelSupport`                      | No       | `object`                         | Manifest-owned shorthand model-family metadata used to auto-load the plugin before runtime.                                                                                                                  |
 | `cliBackends`                       | No       | `string[]`                       | CLI inference backend ids owned by this plugin. Used for startup auto-activation from explicit config refs.                                                                                                  |
+| `commandAliases`                    | No       | `object[]`                       | Command names owned by this plugin that should produce plugin-aware config and CLI diagnostics before runtime loads.                                                                                         |
 | `providerAuthEnvVars`               | No       | `Record<string, string[]>`       | Cheap provider-auth env metadata that OpenClaw can inspect without loading plugin code.                                                                                                                      |
 | `providerAuthAliases`               | No       | `Record<string, string>`         | Provider ids that should reuse another provider id for auth lookup, for example a coding provider that shares the base provider API key and auth profiles.                                                   |
 | `channelEnvVars`                    | No       | `Record<string, string[]>`       | Cheap channel env metadata that OpenClaw can inspect without loading plugin code. Use this for env-driven channel setup or auth surfaces that generic startup/config helpers should see.                     |
@@ -183,6 +184,30 @@ OpenClaw reads this before provider runtime loads.
 | `cliDescription`      | No       | `string`                                        | Description used in CLI help.                                                                            |
 | `onboardingScopes`    | No       | `Array<"text-inference" \| "image-generation">` | Which onboarding surfaces this choice should appear in. If omitted, it defaults to `["text-inference"]`. |
 
+## commandAliases reference
+
+Use `commandAliases` when a plugin owns a runtime command name that users may
+mistakenly put in `plugins.allow` or try to run as a root CLI command. OpenClaw
+uses this metadata for diagnostics without importing plugin runtime code.
+
+```json
+{
+  "commandAliases": [
+    {
+      "name": "dreaming",
+      "kind": "runtime-slash",
+      "cliCommand": "memory"
+    }
+  ]
+}
+```
+
+| Field        | Required | Type              | What it means                                                           |
+| ------------ | -------- | ----------------- | ----------------------------------------------------------------------- |
+| `name`       | Yes      | `string`          | Command name that belongs to this plugin.                               |
+| `kind`       | No       | `"runtime-slash"` | Marks the alias as a chat slash command rather than a root CLI command. |
+| `cliCommand` | No       | `string`          | Related root CLI command to suggest for CLI operations, if one exists.  |
+
 ## uiHints reference
 
 `uiHints` is a map from config field names to small rendering hints.

docs/plugins/sdk-agent-harness.md

@@ -0,0 +1,264 @@
+---
+title: "Agent Harness Plugins"
+sidebarTitle: "Agent Harness"
+summary: "Experimental SDK surface for plugins that replace the low level embedded agent executor"
+read_when:
+  - You are changing the embedded agent runtime or harness registry
+  - You are registering an agent harness from a bundled or trusted plugin
+  - You need to understand how the Codex plugin relates to model providers
+---
+
+# Agent Harness Plugins
+
+An **agent harness** is the low level executor for one prepared OpenClaw agent
+turn. It is not a model provider, not a channel, and not a tool registry.
+
+Use this surface only for bundled or trusted native plugins. The contract is
+still experimental because the parameter types intentionally mirror the current
+embedded runner.
+
+## When to use a harness
+
+Register an agent harness when a model family has its own native session
+runtime and the normal OpenClaw provider transport is the wrong abstraction.
+
+Examples:
+
+- a native coding-agent server that owns threads and compaction
+- a local CLI or daemon that must stream native plan/reasoning/tool events
+- a model runtime that needs its own resume id in addition to the OpenClaw
+  session transcript
+
+Do **not** register a harness just to add a new LLM API. For normal HTTP or
+WebSocket model APIs, build a [provider plugin](/plugins/sdk-provider-plugins).
+
+## What core still owns
+
+Before a harness is selected, OpenClaw has already resolved:
+
+- provider and model
+- runtime auth state
+- thinking level and context budget
+- the OpenClaw transcript/session file
+- workspace, sandbox, and tool policy
+- channel reply callbacks and streaming callbacks
+- model fallback and live model switching policy
+
+That split is intentional. A harness runs a prepared attempt; it does not pick
+providers, replace channel delivery, or silently switch models.
+
+## Register a harness
+
+**Import:** `openclaw/plugin-sdk/agent-harness`
+
+```typescript
+import type { AgentHarness } from "openclaw/plugin-sdk/agent-harness";
+import { definePluginEntry } from "openclaw/plugin-sdk/plugin-entry";
+
+const myHarness: AgentHarness = {
+  id: "my-harness",
+  label: "My native agent harness",
+
+  supports(ctx) {
+    return ctx.provider === "my-provider"
+      ? { supported: true, priority: 100 }
+      : { supported: false };
+  },
+
+  async runAttempt(params) {
+    // Start or resume your native thread.
+    // Use params.prompt, params.tools, params.images, params.onPartialReply,
+    // params.onAgentEvent, and the other prepared attempt fields.
+    return await runMyNativeTurn(params);
+  },
+};
+
+export default definePluginEntry({
+  id: "my-native-agent",
+  name: "My Native Agent",
+  description: "Runs selected models through a native agent daemon.",
+  register(api) {
+    api.registerAgentHarness(myHarness);
+  },
+});
+```
+
+## Selection policy
+
+OpenClaw chooses a harness after provider/model resolution:
+
+1. `OPENCLAW_AGENT_RUNTIME=<id>` forces a registered harness with that id.
+2. `OPENCLAW_AGENT_RUNTIME=pi` forces the built-in PI harness.
+3. `OPENCLAW_AGENT_RUNTIME=auto` asks registered harnesses if they support the
+   resolved provider/model.
+4. If no registered harness matches, OpenClaw uses PI unless PI fallback is
+   disabled.
+
+Forced plugin harness failures surface as run failures. In `auto` mode,
+OpenClaw may fall back to PI when the selected plugin harness fails before a
+turn has produced side effects. Set `OPENCLAW_AGENT_HARNESS_FALLBACK=none` or
+`embeddedHarness.fallback: "none"` to make that fallback a hard failure instead.
+
+The bundled Codex plugin registers `codex` as its harness id. Core treats that
+as an ordinary plugin harness id; Codex-specific aliases belong in the plugin
+or operator config, not in the shared runtime selector.
+
+## Provider plus harness pairing
+
+Most harnesses should also register a provider. The provider makes model refs,
+auth status, model metadata, and `/model` selection visible to the rest of
+OpenClaw. The harness then claims that provider in `supports(...)`.
+
+The bundled Codex plugin follows this pattern:
+
+- provider id: `codex`
+- user model refs: `codex/gpt-5.4`, `codex/gpt-5.2`, or another model returned
+  by the Codex app server
+- harness id: `codex`
+- auth: synthetic provider availability, because the Codex harness owns the
+  native Codex login/session
+- app-server request: OpenClaw sends the bare model id to Codex and lets the
+  harness talk to the native app-server protocol
+
+The Codex plugin is additive. Plain `openai/gpt-*` refs remain OpenAI provider
+refs and continue to use the normal OpenClaw provider path. Select `codex/gpt-*`
+when you want Codex-managed auth, Codex model discovery, native threads, and
+Codex app-server execution. `/model` can switch among the Codex models returned
+by the Codex app server without requiring OpenAI provider credentials.
+
+For operator setup, model prefix examples, and Codex-only configs, see
+[Codex Harness](/plugins/codex-harness).
+
+OpenClaw requires Codex app-server `0.118.0` or newer. The Codex plugin checks
+the app-server initialize handshake and blocks older or unversioned servers so
+OpenClaw only runs against the protocol surface it has been tested with.
+
+## Disable PI fallback
+
+By default, OpenClaw runs embedded agents with `agents.defaults.embeddedHarness`
+set to `{ runtime: "auto", fallback: "pi" }`. In `auto` mode, registered plugin
+harnesses can claim a provider/model pair. If none match, or if an auto-selected
+plugin harness fails before producing output, OpenClaw falls back to PI.
+
+Set `fallback: "none"` when you need to prove that a plugin harness is the only
+runtime being exercised. This disables automatic PI fallback; it does not block
+an explicit `runtime: "pi"` or `OPENCLAW_AGENT_RUNTIME=pi`.
+
+For Codex-only embedded runs:
+
+```json
+{
+  "agents": {
+    "defaults": {
+      "model": "codex/gpt-5.4",
+      "embeddedHarness": {
+        "runtime": "codex",
+        "fallback": "none"
+      }
+    }
+  }
+}
+```
+
+If you want any registered plugin harness to claim matching models but never
+want OpenClaw to silently fall back to PI, keep `runtime: "auto"` and disable
+the fallback:
+
+```json
+{
+  "agents": {
+    "defaults": {
+      "embeddedHarness": {
+        "runtime": "auto",
+        "fallback": "none"
+      }
+    }
+  }
+}
+```
+
+Per-agent overrides use the same shape:
+
+```json
+{
+  "agents": {
+    "defaults": {
+      "embeddedHarness": {
+        "runtime": "auto",
+        "fallback": "pi"
+      }
+    },
+    "list": [
+      {
+        "id": "codex-only",
+        "model": "codex/gpt-5.4",
+        "embeddedHarness": {
+          "runtime": "codex",
+          "fallback": "none"
+        }
+      }
+    ]
+  }
+}
+```
+
+`OPENCLAW_AGENT_RUNTIME` still overrides the configured runtime. Use
+`OPENCLAW_AGENT_HARNESS_FALLBACK=none` to disable PI fallback from the
+environment.
+
+```bash
+OPENCLAW_AGENT_RUNTIME=codex \
+OPENCLAW_AGENT_HARNESS_FALLBACK=none \
+openclaw gateway run
+```
+
+With fallback disabled, a session fails early when the requested harness is not
+registered, does not support the resolved provider/model, or fails before
+producing turn side effects. That is intentional for Codex-only deployments and
+for live tests that must prove the Codex app-server path is actually in use.
+
+This setting only controls the embedded agent harness. It does not disable
+image, video, music, TTS, PDF, or other provider-specific model routing.
+
+## Native sessions and transcript mirror
+
+A harness may keep a native session id, thread id, or daemon-side resume token.
+Keep that binding explicitly associated with the OpenClaw session, and keep
+mirroring user-visible assistant/tool output into the OpenClaw transcript.
+
+The OpenClaw transcript remains the compatibility layer for:
+
+- channel-visible session history
+- transcript search and indexing
+- switching back to the built-in PI harness on a later turn
+- generic `/new`, `/reset`, and session deletion behavior
+
+If your harness stores a sidecar binding, implement `reset(...)` so OpenClaw can
+clear it when the owning OpenClaw session is reset.
+
+## Tool and media results
+
+Core constructs the OpenClaw tool list and passes it into the prepared attempt.
+When a harness executes a dynamic tool call, return the tool result back through
+the harness result shape instead of sending channel media yourself.
+
+This keeps text, image, video, music, TTS, approval, and messaging-tool outputs
+on the same delivery path as PI-backed runs.
+
+## Current limitations
+
+- The public import path is generic, but some attempt/result type aliases still
+  carry `Pi` names for compatibility.
+- Third-party harness installation is experimental. Prefer provider plugins
+  until you need a native session runtime.
+- Harness switching is supported across turns. Do not switch harnesses in the
+  middle of a turn after native tools, approvals, assistant text, or message
+  sends have started.
+
+## Related
+
+- [SDK Overview](/plugins/sdk-overview)
+- [Runtime Helpers](/plugins/sdk-runtime)
+- [Provider Plugins](/plugins/sdk-provider-plugins)
+- [Codex Harness](/plugins/codex-harness)
+- [Model Providers](/concepts/model-providers)

docs/plugins/sdk-channel-plugins.md

@@ -256,7 +256,7 @@ should use `resolveInboundMentionDecision({ facts, policy })`.
   <Step title="Package and manifest">
     Create the standard plugin files. The `channel` field in `package.json` is
     what makes this a channel plugin. For the full package-metadata surface,
-    see [Plugin Setup and Config](/plugins/sdk-setup#openclawchannel):
+    see [Plugin Setup and Config](/plugins/sdk-setup#openclaw-channel):
 
     <CodeGroup>
     ```json package.json

docs/plugins/sdk-migration.md

@@ -245,6 +245,7 @@ Current bundled provider examples:
   | `plugin-sdk/allow-from` | Allowlist formatting | `formatAllowFromLowercase` |
   | `plugin-sdk/allowlist-resolution` | Allowlist input mapping | `mapAllowlistResolutionInputs` |
   | `plugin-sdk/command-auth` | Command gating and command-surface helpers | `resolveControlCommandGate`, sender-authorization helpers, command registry helpers |
+  | `plugin-sdk/command-status` | Command status/help renderers | `buildCommandsMessage`, `buildCommandsMessagePaginated`, `buildHelpMessage` |
   | `plugin-sdk/secret-input` | Secret input parsing | Secret input helpers |
   | `plugin-sdk/webhook-ingress` | Webhook request helpers | Webhook target utilities |
   | `plugin-sdk/webhook-request-guards` | Webhook body guard helpers | Request body read/limit helpers |

docs/plugins/sdk-overview.md

@@ -149,6 +149,7 @@ explicitly promotes one as public.
     | Subpath | Key exports |
     | --- | --- |
     | `plugin-sdk/command-auth` | `resolveControlCommandGate`, command registry helpers, sender-authorization helpers |
+    | `plugin-sdk/command-status` | Command/help message builders such as `buildCommandsMessagePaginated` and `buildHelpMessage` |
     | `plugin-sdk/approval-auth-runtime` | Approver resolution and same-chat action-auth helpers |
     | `plugin-sdk/approval-client-runtime` | Native exec approval profile/filter helpers |
     | `plugin-sdk/approval-delivery-runtime` | Native approval capability/delivery adapters |
@@ -218,6 +219,7 @@ explicitly promotes one as public.
     | `plugin-sdk/models-provider-runtime` | `/models` command/provider reply helpers |
     | `plugin-sdk/skill-commands-runtime` | Skill command listing helpers |
     | `plugin-sdk/native-command-registry` | Native command registry/build/serialize helpers |
+    | `plugin-sdk/agent-harness` | Experimental trusted-plugin surface for low-level agent harnesses: harness types, active-run steer/abort helpers, OpenClaw tool bridge helpers, and attempt result utilities |
     | `plugin-sdk/provider-zai-endpoint` | Z.AI endpoint detection helpers |
     | `plugin-sdk/infra-runtime` | System event/heartbeat helpers |
     | `plugin-sdk/collection-runtime` | Small bounded cache helpers |
@@ -301,20 +303,21 @@ methods:
 
 ### Capability registration
 
-| Method                                           | What it registers                |
-| ------------------------------------------------ | -------------------------------- |
-| `api.registerProvider(...)`                      | Text inference (LLM)             |
-| `api.registerCliBackend(...)`                    | Local CLI inference backend      |
-| `api.registerChannel(...)`                       | Messaging channel                |
-| `api.registerSpeechProvider(...)`                | Text-to-speech / STT synthesis   |
-| `api.registerRealtimeTranscriptionProvider(...)` | Streaming realtime transcription |
-| `api.registerRealtimeVoiceProvider(...)`         | Duplex realtime voice sessions   |
-| `api.registerMediaUnderstandingProvider(...)`    | Image/audio/video analysis       |
-| `api.registerImageGenerationProvider(...)`       | Image generation                 |
-| `api.registerMusicGenerationProvider(...)`       | Music generation                 |
-| `api.registerVideoGenerationProvider(...)`       | Video generation                 |
-| `api.registerWebFetchProvider(...)`              | Web fetch / scrape provider      |
-| `api.registerWebSearchProvider(...)`             | Web search                       |
+| Method                                           | What it registers                     |
+| ------------------------------------------------ | ------------------------------------- |
+| `api.registerProvider(...)`                      | Text inference (LLM)                  |
+| `api.registerAgentHarness(...)`                  | Experimental low-level agent executor |
+| `api.registerCliBackend(...)`                    | Local CLI inference backend           |
+| `api.registerChannel(...)`                       | Messaging channel                     |
+| `api.registerSpeechProvider(...)`                | Text-to-speech / STT synthesis        |
+| `api.registerRealtimeTranscriptionProvider(...)` | Streaming realtime transcription      |
+| `api.registerRealtimeVoiceProvider(...)`         | Duplex realtime voice sessions        |
+| `api.registerMediaUnderstandingProvider(...)`    | Image/audio/video analysis            |
+| `api.registerImageGenerationProvider(...)`       | Image generation                      |
+| `api.registerMusicGenerationProvider(...)`       | Music generation                      |
+| `api.registerVideoGenerationProvider(...)`       | Video generation                      |
+| `api.registerWebFetchProvider(...)`              | Web fetch / scrape provider           |
+| `api.registerWebSearchProvider(...)`             | Web search                            |
 
 ### Tools and commands
 

docs/plugins/sdk-provider-plugins.md

@@ -20,6 +20,13 @@ API key auth, and dynamic model resolution.
   structure and manifest setup.
 </Info>
 
+<Tip>
+  Provider plugins add models to OpenClaw's normal inference loop. If the model
+  must run through a native agent daemon that owns threads, compaction, or tool
+  events, pair the provider with an [agent harness](/plugins/sdk-agent-harness)
+  instead of putting daemon protocol details in core.
+</Tip>
+
 ## Walkthrough
 
 <Steps>
@@ -168,6 +175,28 @@ API key auth, and dynamic model resolution.
     `openclaw onboard --acme-ai-api-key <key>` and select
     `acme-ai/acme-large` as their model.
 
+    If the upstream provider uses different control tokens than OpenClaw, add a
+    small bidirectional text transform instead of replacing the stream path:
+
+    ```typescript
+    api.registerTextTransforms({
+      input: [
+        { from: /red basket/g, to: "blue basket" },
+        { from: /paper ticket/g, to: "digital ticket" },
+        { from: /left shelf/g, to: "right shelf" },
+      ],
+      output: [
+        { from: /blue basket/g, to: "red basket" },
+        { from: /digital ticket/g, to: "paper ticket" },
+        { from: /right shelf/g, to: "left shelf" },
+      ],
+    });
+    ```
+
+    `input` rewrites the final system prompt and text message content before
+    transport. `output` rewrites assistant text deltas and final text before
+    OpenClaw parses its own control markers or channel delivery.
+
     For bundled providers that only register one text provider with API-key
     auth plus a single catalog-backed runtime, prefer the narrower
     `defineSingleProviderPluginEntry(...)` helper:

docs/plugins/sdk-runtime.md

@@ -50,9 +50,9 @@ const timeoutMs = api.runtime.agent.resolveAgentTimeoutMs(cfg);
 // Ensure workspace exists
 await api.runtime.agent.ensureAgentWorkspace(cfg);
 
-// Run an embedded Pi agent
+// Run an embedded agent turn
 const agentDir = api.runtime.agent.resolveAgentDir(cfg);
-const result = await api.runtime.agent.runEmbeddedPiAgent({
+const result = await api.runtime.agent.runEmbeddedAgent({
   sessionId: "my-plugin:task-1",
   runId: crypto.randomUUID(),
   sessionFile: path.join(agentDir, "sessions", "my-plugin-task-1.jsonl"),
@@ -62,6 +62,12 @@ const result = await api.runtime.agent.runEmbeddedPiAgent({
 });
 ```
 
+`runEmbeddedAgent(...)` is the neutral helper for starting a normal OpenClaw
+agent turn from plugin code. It uses the same provider/model resolution and
+agent-harness selection as channel-triggered replies.
+
+`runEmbeddedPiAgent(...)` remains as a compatibility alias.
+
 **Session store helpers** are under `api.runtime.agent.session`:
 
 ```typescript

docs/providers/fal.md

@@ -69,15 +69,36 @@ The bundled `fal` video-generation provider defaults to
 
 - Modes: text-to-video and single-image reference flows
 - Runtime: queue-backed submit/status/result flow for long-running jobs
+- HeyGen video-agent model ref:
+  - `fal/fal-ai/heygen/v2/video-agent`
+- Seedance 2.0 model refs:
+  - `fal/bytedance/seedance-2.0/fast/text-to-video`
+  - `fal/bytedance/seedance-2.0/fast/image-to-video`
+  - `fal/bytedance/seedance-2.0/text-to-video`
+  - `fal/bytedance/seedance-2.0/image-to-video`
 
-To use fal as the default video provider:
+To use Seedance 2.0 as the default video model:
 
 ```json5
 {
   agents: {
     defaults: {
       videoGenerationModel: {
-        primary: "fal/fal-ai/minimax/video-01-live",
+        primary: "fal/bytedance/seedance-2.0/fast/text-to-video",
+      },
+    },
+  },
+}
+```
+
+To use HeyGen video-agent as the default video model:
+
+```json5
+{
+  agents: {
+    defaults: {
+      videoGenerationModel: {
+        primary: "fal/fal-ai/heygen/v2/video-agent",
       },
     },
   },

docs/providers/qwen.md

@@ -88,7 +88,9 @@ requiring the built-in `qwen` provider id specifically.
 
 ## Built-in catalog
 
-OpenClaw currently ships this bundled Qwen catalog:
+OpenClaw currently ships this bundled Qwen catalog. The configured catalog is
+endpoint-aware: Coding Plan configs omit models that are only known to work on
+the Standard endpoint.
 
 | Model ref                   | Input       | Context   | Notes                                              |
 | --------------------------- | ----------- | --------- | -------------------------------------------------- |

docs/reference/RELEASING.md

@@ -69,9 +69,9 @@ OpenClaw has three public release lanes:
 - npm release preflight fails closed unless the tarball includes both
   `dist/control-ui/index.html` and a non-empty `dist/control-ui/assets/` payload
   so we do not ship an empty browser dashboard again
-- If the release work touched CI planning, extension timing manifests, or fast
-  test matrices, regenerate and review the planner-owned `checks-fast-extensions`
-  workflow matrix outputs from `.github/workflows/ci.yml`
+- If the release work touched CI planning, extension timing manifests, or
+  extension test matrices, regenerate and review the planner-owned
+  `checks-node-extensions` workflow matrix outputs from `.github/workflows/ci.yml`
   before approval so release notes do not describe a stale CI layout
 - Stable macOS release readiness also includes the updater surfaces:
   - the GitHub release must end up with the packaged `.zip`, `.dmg`, and `.dSYM.zip`

docs/reference/memory-config.md

@@ -17,10 +17,22 @@ conceptual overviews, see:
 - [Builtin Engine](/concepts/memory-builtin) -- default SQLite backend
 - [QMD Engine](/concepts/memory-qmd) -- local-first sidecar
 - [Memory Search](/concepts/memory-search) -- search pipeline and tuning
+- [Active Memory](/concepts/active-memory) -- enabling the memory sub-agent for interactive sessions
 
 All memory search settings live under `agents.defaults.memorySearch` in
 `openclaw.json` unless noted otherwise.
 
+If you are looking for the **active memory** feature toggle and sub-agent config,
+that lives under `plugins.entries.active-memory` instead of `memorySearch`.
+
+Active memory uses a two-gate model:
+
+1. the plugin must be enabled and target the current agent id
+2. the request must be an eligible interactive persistent chat session
+
+See [Active Memory](/concepts/active-memory) for the activation model,
+plugin-owned config, transcript persistence, and safe rollout pattern.
+
 ---
 
 ## Provider selection

docs/reference/templates/AGENTS.md

@@ -136,9 +136,6 @@ Skills provide your tools. When you need one, check its `SKILL.md`. Keep local n
 
 When you receive a heartbeat poll (message matches the configured heartbeat prompt), don't just reply `HEARTBEAT_OK` every time. Use heartbeats productively!
 
-Default heartbeat prompt:
-`Read HEARTBEAT.md if it exists (workspace context). Follow it strictly. Do not infer or repeat old tasks from prior chats. If nothing needs attention, reply HEARTBEAT_OK.`
-
 You are free to edit `HEARTBEAT.md` with a short checklist or reminders. Keep it small to limit token burn.
 
 ### Heartbeat vs Cron: When to Use Each

docs/tools/browser.md

@@ -146,7 +146,7 @@ Browser settings live in `~/.openclaw/openclaw.json`.
   browser: {
     enabled: true, // default: true
     ssrfPolicy: {
-      dangerouslyAllowPrivateNetwork: true, // default trusted-network mode
+      // dangerouslyAllowPrivateNetwork: true, // opt in only for trusted private-network access
       // allowPrivateNetwork: true, // legacy alias
       // hostnameAllowlist: ["*.example.com", "example.com"],
       // allowedHostnames: ["localhost"],
@@ -191,7 +191,7 @@ Notes:
 - `remoteCdpHandshakeTimeoutMs` applies to remote CDP WebSocket reachability checks.
 - Browser navigation/open-tab is SSRF-guarded before navigation and best-effort re-checked on final `http(s)` URL after navigation.
 - In strict SSRF mode, remote CDP endpoint discovery/probes (`cdpUrl`, including `/json/version` lookups) are checked too.
-- `browser.ssrfPolicy.dangerouslyAllowPrivateNetwork` defaults to `true` (trusted-network model). Set it to `false` for strict public-only browsing.
+- `browser.ssrfPolicy.dangerouslyAllowPrivateNetwork` is disabled by default. Set it to `true` only when you intentionally trust private-network browser access.
 - `browser.ssrfPolicy.allowPrivateNetwork` remains supported as a legacy alias for compatibility.
 - `attachOnly: true` means “never launch a local browser; only attach if it is already running.”
 - `color` + per-profile `color` tint the browser UI so you can see which profile is active.
@@ -576,6 +576,27 @@ Notes:
 - If `gateway.auth.mode` is `none` or `trusted-proxy`, these loopback browser
   routes do not inherit those identity-bearing modes; keep them loopback-only.
 
+### `/act` error contract
+
+`POST /act` uses a structured error response for route-level validation and
+policy failures:
+
+```json
+{ "error": "<message>", "code": "ACT_*" }
+```
+
+Current `code` values:
+
+- `ACT_KIND_REQUIRED` (HTTP 400): `kind` is missing or unrecognized.
+- `ACT_INVALID_REQUEST` (HTTP 400): action payload failed normalization or validation.
+- `ACT_SELECTOR_UNSUPPORTED` (HTTP 400): `selector` was used with an unsupported action kind.
+- `ACT_EVALUATE_DISABLED` (HTTP 403): `evaluate` (or `wait --fn`) is disabled by config.
+- `ACT_TARGET_ID_MISMATCH` (HTTP 403): top-level or batched `targetId` conflicts with request target.
+- `ACT_EXISTING_SESSION_UNSUPPORTED` (HTTP 501): action is not supported for existing-session profiles.
+
+Other runtime failures may still return `{ "error": "<message>" }` without a
+`code` field.
+
 ### Playwright requirement
 
 Some features (navigate/act/AI snapshot/role snapshot, element screenshots,

docs/tools/exec-approvals.md

@@ -20,6 +20,11 @@ session or config defaults request `ask: "on-miss"`.
 Use `openclaw approvals get`, `openclaw approvals get --gateway`, or
 `openclaw approvals get --node <id|name|ip>` to inspect the requested policy,
 host policy sources, and the effective result.
+For the local machine, `openclaw exec-policy show` exposes the same merged view and
+`openclaw exec-policy set|preset` can synchronize the local requested policy with the
+local host approvals file in one step. When a local scope requests `host=node`,
+`openclaw exec-policy show` reports that scope as node-managed at runtime instead of
+pretending the local approvals file is the effective source of truth.
 
 If the companion app UI is **not available**, any request that requires a prompt is
 resolved by the **ask fallback** (default: deny).
@@ -143,6 +148,21 @@ openclaw approvals set --stdin <<'EOF'
 EOF
 ```
 
+Local shortcut for the same gateway-host policy on the current machine:
+
+```bash
+openclaw exec-policy preset yolo
+```
+
+That local shortcut updates both:
+
+- local `tools.exec.host/security/ask`
+- local `~/.openclaw/exec-approvals.json` defaults
+
+It is intentionally local-only. If you need to change gateway-host or node-host approvals
+remotely, continue using `openclaw approvals set --gateway` or
+`openclaw approvals set --node <id|name|ip>`.
+
 For a node host, apply the same approvals file on that node instead:
 
 ```bash
@@ -158,6 +178,12 @@ openclaw approvals set --node <id|name|ip> --stdin <<'EOF'
 EOF
 ```
 
+Important local-only limitation:
+
+- `openclaw exec-policy` does not synchronize node approvals
+- `openclaw exec-policy set --host node` is rejected
+- node exec approvals are fetched from the node at runtime, so node-targeted updates must use `openclaw approvals --node ...`
+
 Session-only shortcut:
 
 - `/exec security=full ask=off` changes only the current session.

docs/tools/reactions.md

@@ -68,7 +68,7 @@ tool with the `react` action. Reaction behavior varies by channel.
 Per-channel `reactionLevel` config controls how broadly the agent uses reactions. Values are typically `off`, `ack`, `minimal`, or `extensive`.
 
 - [Telegram reactionLevel](/channels/telegram#reaction-notifications) — `channels.telegram.reactionLevel`
-- [WhatsApp reactionLevel](/channels/whatsapp#reactions) — `channels.whatsapp.reactionLevel`
+- [WhatsApp reactionLevel](/channels/whatsapp#reaction-level) — `channels.whatsapp.reactionLevel`
 
 Set `reactionLevel` on individual channels to tune how actively the agent reacts to messages on each platform.
 

docs/tools/skills.md

@@ -303,6 +303,13 @@ When an agent run starts, OpenClaw:
 
 This is **scoped to the agent run**, not a global shell environment.
 
+For the bundled `claude-cli` backend, OpenClaw also materializes the same
+eligible snapshot as a temporary Claude Code plugin and passes it with
+`--plugin-dir`. Claude Code can then use its native skill resolver while
+OpenClaw still owns precedence, per-agent allowlists, gating, and
+`skills.entries.*` env/API key injection. Other CLI backends use the prompt
+catalog only.
+
 ## Session snapshot (performance)
 
 OpenClaw snapshots the eligible skills **when a session starts** and reuses that list for subsequent turns in the same session. Changes to skills or config take effect on the next new session.

docs/tools/slash-commands.md

@@ -152,6 +152,7 @@ Bundled plugins can add more slash commands. Current bundled commands in this re
 - `/phone status|arm <camera|screen|writes|all> [duration]|disarm` temporarily arms high-risk phone node commands.
 - `/voice status|list [limit]|set <voiceId|name>` manages Talk voice config. On Discord, the native command name is `/talkvoice`.
 - `/card ...` sends LINE rich card presets. See [LINE](/channels/line).
+- `/codex status|models|threads|resume|compact|review|account|mcp|skills` inspects and controls the bundled Codex app-server harness. See [Codex Harness](/plugins/codex-harness).
 - QQBot-only commands:
   - `/bot-ping`
   - `/bot-version`

docs/tools/video-generation.md

@@ -201,22 +201,50 @@ entries.
 }
 ```
 
+HeyGen video-agent on fal can be pinned with:
+
+```json5
+{
+  agents: {
+    defaults: {
+      videoGenerationModel: {
+        primary: "fal/fal-ai/heygen/v2/video-agent",
+      },
+    },
+  },
+}
+```
+
+Seedance 2.0 on fal can be pinned with:
+
+```json5
+{
+  agents: {
+    defaults: {
+      videoGenerationModel: {
+        primary: "fal/bytedance/seedance-2.0/fast/text-to-video",
+      },
+    },
+  },
+}
+```
+
 ## Provider notes
 
-| Provider | Notes                                                                                                                                                       |
-| -------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| Alibaba  | Uses DashScope/Model Studio async endpoint. Reference images and videos must be remote `http(s)` URLs.                                                      |
-| BytePlus | Single image reference only.                                                                                                                                |
-| ComfyUI  | Workflow-driven local or cloud execution. Supports text-to-video and image-to-video through the configured graph.                                           |
-| fal      | Uses queue-backed flow for long-running jobs. Single image reference only.                                                                                  |
-| Google   | Uses Gemini/Veo. Supports one image or one video reference.                                                                                                 |
-| MiniMax  | Single image reference only.                                                                                                                                |
-| OpenAI   | Only `size` override is forwarded. Other style overrides (`aspectRatio`, `resolution`, `audio`, `watermark`) are ignored with a warning.                    |
-| Qwen     | Same DashScope backend as Alibaba. Reference inputs must be remote `http(s)` URLs; local files are rejected upfront.                                        |
-| Runway   | Supports local files via data URIs. Video-to-video requires `runway/gen4_aleph`. Text-only runs expose `16:9` and `9:16` aspect ratios.                     |
-| Together | Single image reference only.                                                                                                                                |
-| Vydra    | Uses `https://www.vydra.ai/api/v1` directly to avoid auth-dropping redirects. `veo3` is bundled as text-to-video only; `kling` requires a remote image URL. |
-| xAI      | Supports text-to-video, image-to-video, and remote video edit/extend flows.                                                                                 |
+| Provider | Notes                                                                                                                                                                |
+| -------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| Alibaba  | Uses DashScope/Model Studio async endpoint. Reference images and videos must be remote `http(s)` URLs.                                                               |
+| BytePlus | Single image reference only.                                                                                                                                         |
+| ComfyUI  | Workflow-driven local or cloud execution. Supports text-to-video and image-to-video through the configured graph.                                                    |
+| fal      | Uses queue-backed flow for long-running jobs. Single image reference only. Includes HeyGen video-agent and Seedance 2.0 text-to-video and image-to-video model refs. |
+| Google   | Uses Gemini/Veo. Supports one image or one video reference.                                                                                                          |
+| MiniMax  | Single image reference only.                                                                                                                                         |
+| OpenAI   | Only `size` override is forwarded. Other style overrides (`aspectRatio`, `resolution`, `audio`, `watermark`) are ignored with a warning.                             |
+| Qwen     | Same DashScope backend as Alibaba. Reference inputs must be remote `http(s)` URLs; local files are rejected upfront.                                                 |
+| Runway   | Supports local files via data URIs. Video-to-video requires `runway/gen4_aleph`. Text-only runs expose `16:9` and `9:16` aspect ratios.                              |
+| Together | Single image reference only.                                                                                                                                         |
+| Vydra    | Uses `https://www.vydra.ai/api/v1` directly to avoid auth-dropping redirects. `veo3` is bundled as text-to-video only; `kling` requires a remote image URL.          |
+| xAI      | Supports text-to-video, image-to-video, and remote video edit/extend flows.                                                                                          |
 
 ## Provider capability modes
 

extensions/acpx/package.json

@@ -1,10 +1,10 @@
 {
   "name": "@openclaw/acpx",
-  "version": "2026.4.9",
+  "version": "2026.4.10",
   "description": "OpenClaw ACP runtime backend",
   "type": "module",
   "dependencies": {
-    "acpx": "0.5.2"
+    "acpx": "0.5.3"
   },
   "devDependencies": {
     "@openclaw/plugin-sdk": "workspace:*"

extensions/acpx/src/runtime-internals/mcp-command-line.test.ts

@@ -2,7 +2,7 @@ import { describe, expect, it } from "vitest";
 
 type SplitCommandLine = (
   value: string,
-  platform?: NodeJS.Platform | string,
+  platform?: string,
 ) => {
   command: string;
   args: string[];

extensions/acpx/src/runtime.test.ts

@@ -1,11 +1,15 @@
-import type { AcpSessionStore } from "acpx/runtime";
 import { beforeEach, describe, expect, it, vi } from "vitest";
 import type { AcpRuntime } from "../runtime-api.js";
 import { AcpxRuntime } from "./runtime.js";
 
-function makeRuntime(baseStore: AcpSessionStore): {
+type TestSessionStore = {
+  load(sessionId: string): Promise<Record<string, unknown> | undefined>;
+  save(record: Record<string, unknown>): Promise<void>;
+};
+
+function makeRuntime(baseStore: TestSessionStore): {
   runtime: AcpxRuntime;
-  wrappedStore: AcpSessionStore & { markFresh: (sessionKey: string) => void };
+  wrappedStore: TestSessionStore & { markFresh: (sessionKey: string) => void };
   delegate: { close: AcpRuntime["close"] };
 } {
   const runtime = new AcpxRuntime({
@@ -22,7 +26,7 @@ function makeRuntime(baseStore: AcpSessionStore): {
     runtime,
     wrappedStore: (
       runtime as unknown as {
-        sessionStore: AcpSessionStore & { markFresh: (sessionKey: string) => void };
+        sessionStore: TestSessionStore & { markFresh: (sessionKey: string) => void };
       }
     ).sessionStore,
     delegate: (runtime as unknown as { delegate: { close: AcpRuntime["close"] } }).delegate,
@@ -35,7 +39,7 @@ describe("AcpxRuntime fresh reset wrapper", () => {
   });
 
   it("keeps stale persistent loads hidden until a fresh record is saved", async () => {
-    const baseStore: AcpSessionStore = {
+    const baseStore: TestSessionStore = {
       load: vi.fn(async () => ({ acpxRecordId: "stale" }) as never),
       save: vi.fn(async () => {}),
     };
@@ -68,7 +72,7 @@ describe("AcpxRuntime fresh reset wrapper", () => {
   });
 
   it("marks the session fresh after discardPersistentState close", async () => {
-    const baseStore: AcpSessionStore = {
+    const baseStore: TestSessionStore = {
       load: vi.fn(async () => ({ acpxRecordId: "stale" }) as never),
       save: vi.fn(async () => {}),
     };

extensions/acpx/src/service.test.ts

@@ -19,7 +19,7 @@ vi.mock("../runtime-api.js", () => ({
 
 vi.mock("./runtime.js", () => ({
   ACPX_BACKEND_ID: "acpx",
-  AcpxRuntime: class {},
+  AcpxRuntime: function AcpxRuntime() {},
   createAgentRegistry: vi.fn(() => ({})),
   createFileSessionStore: vi.fn(() => ({})),
 }));

extensions/active-memory/openclaw.plugin.json

@@ -0,0 +1,120 @@
+{
+  "id": "active-memory",
+  "name": "Active Memory",
+  "description": "Runs a bounded blocking memory sub-agent before eligible conversational replies and injects relevant memory into prompt context.",
+  "configSchema": {
+    "type": "object",
+    "additionalProperties": false,
+    "properties": {
+      "enabled": { "type": "boolean" },
+      "agents": {
+        "type": "array",
+        "items": { "type": "string" }
+      },
+      "model": { "type": "string" },
+      "modelFallbackPolicy": {
+        "type": "string",
+        "enum": ["default-remote", "resolved-only"]
+      },
+      "allowedChatTypes": {
+        "type": "array",
+        "items": {
+          "type": "string",
+          "enum": ["direct", "group", "channel"]
+        }
+      },
+      "thinking": {
+        "type": "string",
+        "enum": ["off", "minimal", "low", "medium", "high", "xhigh", "adaptive"]
+      },
+      "timeoutMs": { "type": "integer", "minimum": 250 },
+      "queryMode": {
+        "type": "string",
+        "enum": ["message", "recent", "full"]
+      },
+      "promptStyle": {
+        "type": "string",
+        "enum": [
+          "balanced",
+          "strict",
+          "contextual",
+          "recall-heavy",
+          "precision-heavy",
+          "preference-only"
+        ]
+      },
+      "promptOverride": { "type": "string" },
+      "promptAppend": { "type": "string" },
+      "maxSummaryChars": { "type": "integer", "minimum": 40, "maximum": 1000 },
+      "recentUserTurns": { "type": "integer", "minimum": 0, "maximum": 4 },
+      "recentAssistantTurns": { "type": "integer", "minimum": 0, "maximum": 3 },
+      "recentUserChars": { "type": "integer", "minimum": 40, "maximum": 1000 },
+      "recentAssistantChars": { "type": "integer", "minimum": 40, "maximum": 1000 },
+      "logging": { "type": "boolean" },
+      "persistTranscripts": { "type": "boolean" },
+      "transcriptDir": { "type": "string" },
+      "cacheTtlMs": { "type": "integer", "minimum": 1000, "maximum": 120000 }
+    }
+  },
+  "uiHints": {
+    "enabled": {
+      "label": "Active Memory Recall",
+      "help": "Globally enable or pause Active Memory recall while keeping the plugin command available."
+    },
+    "agents": {
+      "label": "Target Agents",
+      "help": "Explicit agent ids that may use active memory."
+    },
+    "model": {
+      "label": "Memory Model",
+      "help": "Provider/model used for the blocking memory sub-agent."
+    },
+    "modelFallbackPolicy": {
+      "label": "Model Fallback Policy",
+      "help": "Choose whether Active Memory falls back to the built-in remote default model when no explicit or inherited model is available."
+    },
+    "allowedChatTypes": {
+      "label": "Allowed Chat Types",
+      "help": "Choose which session types may run Active Memory. Defaults to direct-message style sessions only."
+    },
+    "timeoutMs": {
+      "label": "Timeout (ms)"
+    },
+    "queryMode": {
+      "label": "Query Mode",
+      "help": "Choose whether the blocking memory sub-agent sees only the latest user message, a small recent tail, or the full conversation."
+    },
+    "promptStyle": {
+      "label": "Prompt Style",
+      "help": "Choose how eager or strict the blocking memory sub-agent should be when deciding whether to return memory."
+    },
+    "thinking": {
+      "label": "Thinking Override",
+      "help": "Advanced: optional thinking level for the blocking memory sub-agent. Defaults to off for speed."
+    },
+    "promptOverride": {
+      "label": "Prompt Override",
+      "help": "Advanced: replace the default Active Memory sub-agent instructions. Conversation context is still appended."
+    },
+    "promptAppend": {
+      "label": "Prompt Append",
+      "help": "Advanced: append extra operator instructions after the default Active Memory sub-agent instructions."
+    },
+    "maxSummaryChars": {
+      "label": "Max Summary Characters",
+      "help": "Maximum total characters allowed in the active-memory summary."
+    },
+    "logging": {
+      "label": "Enable Logging",
+      "help": "Emit active memory timing and result logs."
+    },
+    "persistTranscripts": {
+      "label": "Persist Transcripts",
+      "help": "Keep blocking memory sub-agent session transcripts on disk in a separate plugin-owned directory."
+    },
+    "transcriptDir": {
+      "label": "Transcript Directory",
+      "help": "Relative directory under the agent sessions folder used when transcript persistence is enabled."
+    }
+  }
+}

extensions/alibaba/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@openclaw/alibaba-provider",
-  "version": "2026.4.9",
+  "version": "2026.4.10",
   "private": true,
   "description": "OpenClaw Alibaba Model Studio video provider plugin",
   "type": "module",

extensions/amazon-bedrock-mantle/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@openclaw/amazon-bedrock-mantle-provider",
-  "version": "2026.4.9",
+  "version": "2026.4.10",
   "private": true,
   "description": "OpenClaw Amazon Bedrock Mantle (OpenAI-compatible) provider plugin",
   "type": "module",

extensions/amazon-bedrock/package.json

@@ -1,11 +1,11 @@
 {
   "name": "@openclaw/amazon-bedrock-provider",
-  "version": "2026.4.9",
+  "version": "2026.4.10",
   "private": true,
   "description": "OpenClaw Amazon Bedrock provider plugin",
   "type": "module",
   "dependencies": {
-    "@aws-sdk/client-bedrock": "3.1024.0"
+    "@aws-sdk/client-bedrock": "3.1028.0"
   },
   "devDependencies": {
     "@openclaw/plugin-sdk": "workspace:*"

extensions/anthropic-vertex/openclaw.plugin.json

@@ -2,6 +2,7 @@
   "id": "anthropic-vertex",
   "enabledByDefault": true,
   "providers": ["anthropic-vertex"],
+  "providerDiscoveryEntry": "./provider-discovery.ts",
   "configSchema": {
     "type": "object",
     "additionalProperties": false,

extensions/anthropic-vertex/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@openclaw/anthropic-vertex-provider",
-  "version": "2026.4.9",
+  "version": "2026.4.10",
   "private": true,
   "description": "OpenClaw Anthropic Vertex provider plugin",
   "type": "module",

extensions/anthropic-vertex/provider-discovery.import-guard.test.ts

@@ -0,0 +1,10 @@
+import { describe, expect, it } from "vitest";
+
+describe("anthropic-vertex provider discovery entry", () => {
+  it("imports without loading the full plugin entry", async () => {
+    const module = await import("./provider-discovery.js");
+
+    expect(module.default.id).toBe("anthropic-vertex");
+    expect(module.default.catalog.order).toBe("simple");
+  });
+});

extensions/anthropic-vertex/provider-discovery.ts

@@ -0,0 +1,215 @@
+import { readFileSync } from "node:fs";
+import { homedir, platform } from "node:os";
+import { join } from "node:path";
+import type { ProviderCatalogContext } from "openclaw/plugin-sdk/provider-catalog-shared";
+import type {
+  ModelDefinitionConfig,
+  ModelProviderConfig,
+} from "openclaw/plugin-sdk/provider-model-shared";
+
+const PROVIDER_ID = "anthropic-vertex";
+const ANTHROPIC_VERTEX_DEFAULT_REGION = "global";
+const ANTHROPIC_VERTEX_REGION_RE = /^[a-z0-9-]+$/;
+const ANTHROPIC_VERTEX_DEFAULT_CONTEXT_WINDOW = 1_000_000;
+const GCP_VERTEX_CREDENTIALS_MARKER = "gcp-vertex-credentials";
+const GCLOUD_DEFAULT_ADC_PATH = join(
+  homedir(),
+  ".config",
+  "gcloud",
+  "application_default_credentials.json",
+);
+
+type AnthropicVertexProviderPlugin = {
+  id: string;
+  label: string;
+  docsPath: string;
+  auth: [];
+  catalog: {
+    order: "simple";
+    run: (ctx: ProviderCatalogContext) => ReturnType<typeof runAnthropicVertexCatalog>;
+  };
+  resolveConfigApiKey: (params: { env: NodeJS.ProcessEnv }) => string | undefined;
+};
+
+type AdcProjectFile = {
+  project_id?: unknown;
+  quota_project_id?: unknown;
+};
+
+function normalizeOptionalString(value: unknown): string | undefined {
+  return typeof value === "string" && value.trim() ? value.trim() : undefined;
+}
+
+function normalizeLowercaseStringOrEmpty(value: unknown): string {
+  return normalizeOptionalString(value)?.toLowerCase() ?? "";
+}
+
+function resolveAnthropicVertexRegion(env: NodeJS.ProcessEnv = process.env): string {
+  const region =
+    normalizeOptionalString(env.GOOGLE_CLOUD_LOCATION) ||
+    normalizeOptionalString(env.CLOUD_ML_REGION);
+
+  return region && ANTHROPIC_VERTEX_REGION_RE.test(region)
+    ? region
+    : ANTHROPIC_VERTEX_DEFAULT_REGION;
+}
+
+function hasAnthropicVertexMetadataServerAdc(env: NodeJS.ProcessEnv = process.env): boolean {
+  const explicitMetadataOptIn = normalizeOptionalString(env.ANTHROPIC_VERTEX_USE_GCP_METADATA);
+  return (
+    explicitMetadataOptIn === "1" ||
+    normalizeLowercaseStringOrEmpty(explicitMetadataOptIn) === "true"
+  );
+}
+
+function resolveAnthropicVertexDefaultAdcPath(env: NodeJS.ProcessEnv = process.env): string {
+  return platform() === "win32"
+    ? join(
+        env.APPDATA ?? join(homedir(), "AppData", "Roaming"),
+        "gcloud",
+        "application_default_credentials.json",
+      )
+    : GCLOUD_DEFAULT_ADC_PATH;
+}
+
+function resolveAnthropicVertexAdcCredentialsPathCandidate(
+  env: NodeJS.ProcessEnv = process.env,
+): string | undefined {
+  const explicit = normalizeOptionalString(env.GOOGLE_APPLICATION_CREDENTIALS);
+  if (explicit) {
+    return explicit;
+  }
+  if (env !== process.env) {
+    return undefined;
+  }
+  return resolveAnthropicVertexDefaultAdcPath(env);
+}
+
+function readAnthropicVertexAdc(env: NodeJS.ProcessEnv = process.env): AdcProjectFile | null {
+  const credentialsPath = resolveAnthropicVertexAdcCredentialsPathCandidate(env);
+  if (!credentialsPath) {
+    return null;
+  }
+  try {
+    return JSON.parse(readFileSync(credentialsPath, "utf8")) as AdcProjectFile;
+  } catch {
+    return null;
+  }
+}
+
+function hasAnthropicVertexAvailableAuth(env: NodeJS.ProcessEnv = process.env): boolean {
+  return hasAnthropicVertexMetadataServerAdc(env) || readAnthropicVertexAdc(env) !== null;
+}
+
+function resolveAnthropicVertexConfigApiKey(
+  env: NodeJS.ProcessEnv = process.env,
+): string | undefined {
+  return hasAnthropicVertexAvailableAuth(env) ? GCP_VERTEX_CREDENTIALS_MARKER : undefined;
+}
+
+function buildAnthropicVertexModel(params: {
+  id: string;
+  name: string;
+  reasoning: boolean;
+  input: ModelDefinitionConfig["input"];
+  cost: ModelDefinitionConfig["cost"];
+  maxTokens: number;
+}): ModelDefinitionConfig {
+  return {
+    id: params.id,
+    name: params.name,
+    reasoning: params.reasoning,
+    input: params.input,
+    cost: params.cost,
+    contextWindow: ANTHROPIC_VERTEX_DEFAULT_CONTEXT_WINDOW,
+    maxTokens: params.maxTokens,
+  };
+}
+
+function buildAnthropicVertexProvider(params?: { env?: NodeJS.ProcessEnv }): ModelProviderConfig {
+  const region = resolveAnthropicVertexRegion(params?.env);
+  const baseUrl =
+    normalizeLowercaseStringOrEmpty(region) === "global"
+      ? "https://aiplatform.googleapis.com"
+      : `https://${region}-aiplatform.googleapis.com`;
+
+  return {
+    baseUrl,
+    api: "anthropic-messages",
+    apiKey: GCP_VERTEX_CREDENTIALS_MARKER,
+    models: [
+      buildAnthropicVertexModel({
+        id: "claude-opus-4-6",
+        name: "Claude Opus 4.6",
+        reasoning: true,
+        input: ["text", "image"],
+        cost: { input: 5, output: 25, cacheRead: 0.5, cacheWrite: 6.25 },
+        maxTokens: 128000,
+      }),
+      buildAnthropicVertexModel({
+        id: "claude-sonnet-4-6",
+        name: "Claude Sonnet 4.6",
+        reasoning: true,
+        input: ["text", "image"],
+        cost: { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 3.75 },
+        maxTokens: 128000,
+      }),
+    ],
+  };
+}
+
+function mergeImplicitAnthropicVertexProvider(params: {
+  existing?: ModelProviderConfig;
+  implicit: ModelProviderConfig;
+}) {
+  const { existing, implicit } = params;
+  if (!existing) {
+    return implicit;
+  }
+  return {
+    ...implicit,
+    ...existing,
+    models:
+      Array.isArray(existing.models) && existing.models.length > 0
+        ? existing.models
+        : implicit.models,
+  };
+}
+
+function resolveImplicitAnthropicVertexProvider(params?: { env?: NodeJS.ProcessEnv }) {
+  const env = params?.env ?? process.env;
+  if (!hasAnthropicVertexAvailableAuth(env)) {
+    return null;
+  }
+
+  return buildAnthropicVertexProvider({ env });
+}
+
+async function runAnthropicVertexCatalog(ctx: ProviderCatalogContext) {
+  const implicit = resolveImplicitAnthropicVertexProvider({
+    env: ctx.env,
+  });
+  if (!implicit) {
+    return null;
+  }
+  return {
+    provider: mergeImplicitAnthropicVertexProvider({
+      existing: ctx.config.models?.providers?.[PROVIDER_ID],
+      implicit,
+    }),
+  };
+}
+
+export const anthropicVertexProviderDiscovery: AnthropicVertexProviderPlugin = {
+  id: PROVIDER_ID,
+  label: "Anthropic Vertex",
+  docsPath: "/providers/models",
+  auth: [],
+  catalog: {
+    order: "simple",
+    run: runAnthropicVertexCatalog,
+  },
+  resolveConfigApiKey: ({ env }) => resolveAnthropicVertexConfigApiKey(env),
+};
+
+export default anthropicVertexProviderDiscovery;

extensions/anthropic/cli-backend.ts

@@ -7,7 +7,6 @@ import {
   CLAUDE_CLI_BACKEND_ID,
   CLAUDE_CLI_DEFAULT_MODEL_REF,
   CLAUDE_CLI_CLEAR_ENV,
-  CLAUDE_CLI_HOST_MANAGED_ENV,
   CLAUDE_CLI_MODEL_ALIASES,
   CLAUDE_CLI_SESSION_ID_FIELDS,
   normalizeClaudeBackendConfig,
@@ -63,7 +62,6 @@ export function buildAnthropicCliBackend(): CliBackendPlugin {
       systemPromptArg: "--append-system-prompt",
       systemPromptMode: "append",
       systemPromptWhen: "first",
-      env: { ...CLAUDE_CLI_HOST_MANAGED_ENV },
       clearEnv: [...CLAUDE_CLI_CLEAR_ENV],
       reliability: {
         watchdog: {

extensions/anthropic/cli-shared.test.ts

@@ -2,7 +2,6 @@ import { describe, expect, it } from "vitest";
 import { buildAnthropicCliBackend } from "./cli-backend.js";
 import {
   CLAUDE_CLI_CLEAR_ENV,
-  CLAUDE_CLI_HOST_MANAGED_ENV,
   normalizeClaudeBackendConfig,
   normalizeClaudePermissionArgs,
   normalizeClaudeSettingSourcesArgs,
@@ -132,16 +131,19 @@ describe("normalizeClaudeBackendConfig", () => {
     expect(normalized?.resumeArgs).toContain("user");
   });
 
-  it("marks claude cli as host-managed, restricts setting sources, and clears inherited env overrides", () => {
+  it("leaves claude cli subscription-managed, restricts setting sources, and clears inherited env overrides", () => {
     const backend = buildAnthropicCliBackend();
 
-    expect(backend.config.env).toEqual(CLAUDE_CLI_HOST_MANAGED_ENV);
+    expect(backend.config.env).toBeUndefined();
     expect(backend.config.args).toContain("--setting-sources");
     expect(backend.config.args).toContain("user");
     expect(backend.config.resumeArgs).toContain("--setting-sources");
     expect(backend.config.resumeArgs).toContain("user");
     expect(backend.config.clearEnv).toEqual([...CLAUDE_CLI_CLEAR_ENV]);
+    expect(backend.config.clearEnv).toContain("ANTHROPIC_API_TOKEN");
     expect(backend.config.clearEnv).toContain("ANTHROPIC_BASE_URL");
+    expect(backend.config.clearEnv).toContain("ANTHROPIC_CUSTOM_HEADERS");
+    expect(backend.config.clearEnv).toContain("ANTHROPIC_OAUTH_TOKEN");
     expect(backend.config.clearEnv).toContain("CLAUDE_CONFIG_DIR");
     expect(backend.config.clearEnv).toContain("CLAUDE_CODE_USE_BEDROCK");
     expect(backend.config.clearEnv).toContain("CLAUDE_CODE_OAUTH_TOKEN");

extensions/anthropic/cli-shared.ts

@@ -40,10 +40,6 @@ export const CLAUDE_CLI_SESSION_ID_FIELDS = [
   "conversationId",
 ] as const;
 
-export const CLAUDE_CLI_HOST_MANAGED_ENV = {
-  CLAUDE_CODE_PROVIDER_MANAGED_BY_HOST: "1",
-} as const;
-
 // Claude Code honors provider-routing, auth, and config-root env before
 // consulting its local login state, so inherited shell overrides must not
 // steer OpenClaw-managed Claude CLI runs toward a different provider,
@@ -51,8 +47,11 @@ export const CLAUDE_CLI_HOST_MANAGED_ENV = {
 export const CLAUDE_CLI_CLEAR_ENV = [
   "ANTHROPIC_API_KEY",
   "ANTHROPIC_API_KEY_OLD",
+  "ANTHROPIC_API_TOKEN",
   "ANTHROPIC_AUTH_TOKEN",
   "ANTHROPIC_BASE_URL",
+  "ANTHROPIC_CUSTOM_HEADERS",
+  "ANTHROPIC_OAUTH_TOKEN",
   "ANTHROPIC_UNIX_SOCKET",
   "CLAUDE_CONFIG_DIR",
   "CLAUDE_CODE_API_KEY_FILE_DESCRIPTOR",

extensions/anthropic/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@openclaw/anthropic-provider",
-  "version": "2026.4.9",
+  "version": "2026.4.10",
   "private": true,
   "description": "OpenClaw Anthropic provider plugin",
   "type": "module",