Fix Codex image generation tool timeout (#84369)

Summary:
- The branch gives Codex `image_generate` dynamic-tool calls a 120s default watchdog in main and side-thread paths and updates docs, tests, and changelog.
- Reproducibility: yes. Source inspection on current main shows unconfigured Codex `image_generate` calls fall ... -tool default, and the linked source PR includes live Gateway before/after output for the timeout behavior.

Automerge notes:
- PR branch already contained follow-up commit before automerge: fix(clawsweeper): address review for automerge-openclaw-openclaw-8425…
- PR branch already contained follow-up commit before automerge: Fix Codex image generation tool timeout

Validation:
- ClawSweeper review passed for head 10c7f87023.
- Required merge gates passed before the squash merge.

Prepared head SHA: 10c7f87023
Review: https://github.com/openclaw/openclaw/pull/84369#issuecomment-4493288493

Co-authored-by: clawsweeper <274271284+clawsweeper[bot]@users.noreply.github.com>
Co-authored-by: moritzmmayerhofer <254141390+moritzmmayerhofer@users.noreply.github.com>
Co-authored-by: clawsweeper[bot] <274271284+clawsweeper[bot]@users.noreply.github.com>
Approved-by: takhoffman
Co-authored-by: takhoffman <781889+takhoffman@users.noreply.github.com>

CHANGELOG.md

@@ -12,6 +12,7 @@ Docs: https://docs.openclaw.ai
 ### Fixes
 
 - Agents/code mode: spell out the `exec` tool's JavaScript/TypeScript, no Node module, and catalog-bridge constraints in model-visible schema text so agents can use enabled tools without trial-and-error. (#84269) Thanks @Kaspre.
+- Codex: give `image_generate` dynamic-tool calls a 120s default watchdog when no per-call or configured image timeout is set, so image generation no longer falls back to the generic 30s bridge timeout. (#84254) Thanks @moritzmmayerhofer.
 - CLI/message: include a stable top-level `messageId` in `openclaw message --json` output when channel sends return one. (#84191) Thanks @100menotu001.
 - Gateway/agents: use an agent's `identity.name` in Gateway agent summaries when `agents.list[].name` is unset, so configured agent labels remain visible in clients. (#84355; refs #57835) Thanks @luoyanglang.
 - Plugins/hooks: apply a default 30-second timeout to `before_compaction` and `after_compaction` hooks so a hung plugin handler no longer blocks compaction completion. (#84153)

docs/plugins/codex-harness-reference.md

@@ -256,6 +256,8 @@ available timeout in this order:
 
 - A positive per-call `timeoutMs` argument.
 - For `image_generate`, `agents.defaults.imageGenerationModel.timeoutMs`.
+- For `image_generate` without a configured timeout, the 120 second
+  image-generation default.
 - For the media-understanding `image` tool, `tools.media.image.timeoutSeconds`
   converted to milliseconds, or the 60 second media default.
 - The 30 second dynamic-tool default.

docs/plugins/codex-harness.md

@@ -532,9 +532,10 @@ Supported `appServer` fields:
 OpenClaw-owned dynamic tool calls are bounded independently from
 `appServer.requestTimeoutMs`: Codex `item/tool/call` requests use a 30 second
 OpenClaw watchdog by default. A positive per-call `timeoutMs` argument extends
-or shortens that specific tool budget. The `image_generate` tool also uses
+or shortens that specific tool budget. The `image_generate` tool uses
 `agents.defaults.imageGenerationModel.timeoutMs` when the tool call does not
-provide its own timeout, and the media-understanding `image` tool uses
+provide its own timeout, or a 120 second image-generation default otherwise.
+The media-understanding `image` tool uses
 `tools.media.image.timeoutSeconds` or its 60 second media default. Dynamic tool
 budgets are capped at 600000 ms. On timeout, OpenClaw aborts the tool signal
 where supported and returns a failed dynamic-tool response to Codex so the turn

docs/tools/image-generation.md

@@ -237,8 +237,9 @@ from each attempt.
     backends. A per-call `timeoutMs` tool parameter overrides the configured
     default. Google, OpenRouter, and xAI hosted image providers use 180 second
     defaults; Azure OpenAI image generation uses 600 seconds. Codex dynamic-tool
-    calls honor the same timeout budget, bounded by OpenClaw's 600000 ms
-    dynamic-tool bridge maximum.
+    calls use a 120 second `image_generate` bridge default and honor the same
+    timeout budget when configured, bounded by OpenClaw's 600000 ms dynamic-tool
+    bridge maximum.
   </Accordion>
   <Accordion title="Inspect at runtime">
     Use `action: "list"` to inspect the currently registered providers,

extensions/codex/src/app-server/run-attempt.test.ts

@@ -1903,6 +1903,22 @@ describe("runCodexAppServerAttempt", () => {
     ).toBe(180_000);
   });
 
+  it("uses a 120 second default for Codex image generation dynamic tool calls", () => {
+    expect(
+      testing.resolveDynamicToolCallTimeoutMs({
+        call: {
+          threadId: "thread-1",
+          turnId: "turn-1",
+          callId: "call-image-generate-default",
+          namespace: null,
+          tool: "image_generate",
+          arguments: { prompt: "cat" },
+        },
+        config: undefined,
+      }),
+    ).toBe(120_000);
+  });
+
   it("uses the media image timeout for Codex image dynamic tool calls", () => {
     expect(
       testing.resolveDynamicToolCallTimeoutMs({

extensions/codex/src/app-server/run-attempt.ts

@@ -181,6 +181,7 @@ import { filterToolsForVisionInputs } from "./vision-tools.js";
 
 const CODEX_DYNAMIC_TOOL_TIMEOUT_MS = 30_000;
 const CODEX_DYNAMIC_TOOL_MAX_TIMEOUT_MS = 600_000;
+const CODEX_DYNAMIC_IMAGE_GENERATION_TOOL_TIMEOUT_MS = 120_000;
 const CODEX_DYNAMIC_IMAGE_TOOL_TIMEOUT_MS = 60_000;
 const CODEX_APP_SERVER_STARTUP_CONNECTION_CLOSE_MAX_ATTEMPTS = 3;
 const CODEX_APP_SERVER_STARTUP_TIMEOUT_FLOOR_MS = 100;
@@ -3020,9 +3021,12 @@ function readConfiguredDynamicToolTimeoutMs(
   if (toolName === "image_generate") {
     const imageGenerationModel = config?.agents?.defaults?.imageGenerationModel;
     if (!imageGenerationModel || typeof imageGenerationModel !== "object") {
-      return undefined;
+      return CODEX_DYNAMIC_IMAGE_GENERATION_TOOL_TIMEOUT_MS;
     }
-    return readPositiveFiniteTimeoutMs(imageGenerationModel.timeoutMs);
+    return (
+      readPositiveFiniteTimeoutMs(imageGenerationModel.timeoutMs) ??
+      CODEX_DYNAMIC_IMAGE_GENERATION_TOOL_TIMEOUT_MS
+    );
   }
 
   if (toolName === "image") {

extensions/codex/src/app-server/side-question.test.ts

@@ -1072,6 +1072,20 @@ describe("runCodexAppServerSideQuestion", () => {
     expect(timeoutMs).toBe(123_456);
   });
 
+  it("uses a 120 second default for side-thread image_generate calls", () => {
+    const timeoutMs = testing.resolveSideDynamicToolCallTimeoutMs({
+      call: {
+        threadId: "side-thread",
+        turnId: "turn-1",
+        callId: "tool-1",
+        tool: "image_generate",
+      },
+      config: {} as never,
+    });
+
+    expect(timeoutMs).toBe(120_000);
+  });
+
   it("cleans up notification handlers when side tool setup fails", async () => {
     const client = createFakeClient();
     createOpenClawCodingToolsMock.mockImplementation(() => {

extensions/codex/src/app-server/side-question.ts

@@ -66,6 +66,7 @@ import { filterToolsForVisionInputs } from "./vision-tools.js";
 
 const CODEX_SIDE_DYNAMIC_TOOL_TIMEOUT_MS = 30_000;
 const CODEX_SIDE_DYNAMIC_TOOL_MAX_TIMEOUT_MS = 600_000;
+const CODEX_SIDE_DYNAMIC_IMAGE_GENERATION_TOOL_TIMEOUT_MS = 120_000;
 const CODEX_SIDE_DYNAMIC_IMAGE_TOOL_TIMEOUT_MS = 60_000;
 const SIDE_QUESTION_COMPLETION_TIMEOUT_MS = 600_000;
 const CODEX_SIDE_NATIVE_HOOK_RELAY_MIN_TTL_MS = 30 * 60_000;
@@ -671,7 +672,8 @@ function resolveSideDynamicToolCallTimeoutMs(params: {
   const configured =
     readSideDynamicToolCallTimeoutMs(params.call.arguments) ??
     (params.call.tool === "image_generate"
-      ? readSideImageGenerationModelTimeoutMs(params.config)
+      ? (readSideImageGenerationModelTimeoutMs(params.config) ??
+        CODEX_SIDE_DYNAMIC_IMAGE_GENERATION_TOOL_TIMEOUT_MS)
       : undefined) ??
     (params.call.tool === "image"
       ? (readSideTimeoutSecondsAsMs(params.config?.tools?.media?.image?.timeoutSeconds) ??