Compare commits

..

266 Commits

Author SHA1 Message Date
Vincent Koc
c03f97f954 test(plugins): break google contract helper cycles 2026-04-17 14:25:21 -07:00
Vincent Koc
8b5030447a test(plugins): trim contract helper runtime boot 2026-04-17 14:25:21 -07:00
Vincent Koc
48c4a026dd test(plugins): fast-path bundled provider contract loads 2026-04-17 14:25:21 -07:00
Vincent Koc
420b1da82f test(plugins): trim tts summarization contract boot 2026-04-17 14:25:21 -07:00
Vincent Koc
afdbf48914 test(plugins): fast-path bundled setup web providers 2026-04-17 14:25:21 -07:00
Vincent Koc
c0b8250f4f test(plugins): trim contract registry runtime fanout 2026-04-17 14:25:21 -07:00
Vincent Koc
d89cee8787 test(plugins): avoid runtime loads for id-only registry checks 2026-04-17 14:25:21 -07:00
Vincent Koc
815e2fc529 test(plugins): trim tts contract mock startup 2026-04-17 14:25:21 -07:00
Vincent Koc
18b45e63f2 test(plugins): speed up tts contract helper boot 2026-04-17 14:25:21 -07:00
Vincent Koc
855c7cf989 test(plugins): keep loader contracts inventory-backed 2026-04-17 14:25:21 -07:00
Vincent Koc
78f0fb660c test(plugins): avoid per-test discovery reloads 2026-04-17 14:25:21 -07:00
Vincent Koc
30895f7135 fix(auth): restore cli bootstrap split on rebase 2026-04-17 14:19:45 -07:00
Vincent Koc
76812401ca test(auth): align cli overlay coverage after rebase 2026-04-17 14:14:03 -07:00
Vincent Koc
5edf876a5e test(auth): add codex oauth red-blue coverage 2026-04-17 14:14:03 -07:00
Vincent Koc
1e7c7dd02f refactor(auth): polish external oauth bootstrap flow 2026-04-17 14:11:41 -07:00
Vincent Koc
f61712437f refactor(auth): tighten external oauth bootstrap policy 2026-04-17 14:05:26 -07:00
Agustin Rivera
99ef3a63c5 fix(gateway): require read scope for assistant media (#68175)
* fix(gateway): enforce assistant media scopes

* changelog: require read scope for assistant media (#68175)

* skip scope enforcement for auth.mode=none

Exclude method "none" from the identity-bearing scope gate so
gateway.auth.mode=none deployments are not regressed by the new
operator.read check.

---------

Co-authored-by: Devin Robison <drobison@nvidia.com>
2026-04-17 15:03:53 -06:00
Peter Steinberger
af0f7e1bc7 test: type runtime auth overlay mock 2026-04-17 21:56:25 +01:00
Peter Steinberger
8742e8fae3 test: stub channel migration setup surfaces 2026-04-17 21:53:25 +01:00
Peter Steinberger
8dde0acbae test: trim agent test hot spots 2026-04-17 21:53:08 +01:00
Vincent Koc
ff55cd5c16 refactor(auth): drop legacy external cli oauth sync path 2026-04-17 13:52:37 -07:00
Devin Robison
0e7a992d3f fix(agents): filter bundled tools through final policy (#68195)
* fix(agents): filter bundled tools through final policy

* changelog: filter bundled tools through final policy (#68195)

* forward agentId into compaction tool-policy filter

Pass effectiveSkillAgentId to applyFinalEffectiveToolPolicy in the
compaction path so per-agent tool policies apply to bundled tools
during compaction the same way they do during normal runs.

* scope final tool-policy filter to bundled tools only

Running the full tool-policy pipeline on the merged core + bundled tool list
re-filters core tools whose plugin WeakMap metadata no longer survives the
normalize/hook wrappers applied by createOpenClawCodingTools(). Narrow the
helper to only the newly-appended bundled MCP/LSP tools so plugin-provided
core tools keep matching group:plugins and plugin-id allowlist entries.

* harden authorization signals on final tool policy

- message.action gateway handler now server-derives senderIsOwner from the
  authenticated gateway client scopes (ADMIN_SCOPE on client.connect.scopes)
  and ignores any senderIsOwner value on the wire, so a non-admin scoped
  caller cannot spoof owner status to unlock owner-only channel actions or
  owner-only tool policy. Schema keeps the field optional for wire compat
  but documents that it is ignored.

- applyFinalEffectiveToolPolicy now cross-checks caller-provided groupId
  against the session-derived group context resolved from sessionKey (and
  spawnedBy). When they disagree, the caller groupId plus its adjacent
  groupChannel/groupSpace are dropped and a warn is emitted, so a caller
  that fabricates a different group id cannot reach a more permissive
  group-scoped tool policy during the final bundled-tool filter. Added a
  JSDoc trust invariant on the helper input describing the required
  server-verified identity contract.

* align compact agentId resolution with core tools

Drop the explicit agentId on applyFinalEffectiveToolPolicy during
compaction. The core tool set produced just above via
createOpenClawCodingTools(...) also omits agentId, so resolveEffectiveToolPolicy
falls back to resolveAgentIdFromSessionKey(sessionKey) in both places.
Passing effectiveSkillAgentId only to the final filter made the two
policy lookups diverge on legacy/non-agent session keys where the
sessionKey path resolves to main but effectiveSkillAgentId follows the
configured default-agent path, which could deny or allow bundled tools
under a different per-agent policy than the already-created core tools.

* tighten trusted propagation for owner and group signals

- message.action gateway handler: full-operator callers (shared-secret
  bearer or operator.admin scope) now propagate the request-provided
  senderIsOwner through to channel action handlers instead of having it
  hard-coded off. Previously the hardened path force-derived ownership
  from ADMIN_SCOPE alone, which broke owner-gated actions when the
  trusted runtime forwards them via the least-privilege gateway path
  (callGatewayLeastPrivilege requests only the method scope, so even
  legitimate owner senders were downgraded to senderIsOwner=false).
  Narrowly-scoped callers (e.g. operator.write-only) still have the wire
  value forced to false so a non-admin caller cannot assert ownership.

- applyFinalEffectiveToolPolicy: fail-closed when the session key and
  spawnedBy encode no group context. Previously the helper only dropped
  a caller-provided groupId that conflicted with a non-empty set of
  session-derived group ids, which left an accept-caller fallback open
  when the session had no group context at all (direct/cron/subagent
  session keys). An attacker who could run without a group-bound session
  could then supply an arbitrary groupId and reach a more permissive
  group-scoped tool policy. Now: no session-derived group context plus
  any caller-provided groupId drops the caller value and warns.

* suppress unavailable-core-tool warnings in bundled-only pass

applyToolPolicyPipeline infers its coreToolNames reference set from the
tools array it is filtering. The bundled-only second pass only sees the
MCP/LSP subset, so normal core allowlist entries (for example
tools.allow: ['read', 'exec']) would look "unknown" during this pass
and emit misleading warnings even when the config is valid for the full
effective tool set — polluting logs and potentially evicting real
diagnostics from the shared warning cache. Set
suppressUnavailableCoreToolWarning on every step of this pass so known
core-tool allowlist entries stay silent; genuinely unknown entries
still surface through the otherEntries warning path.
2026-04-17 14:45:12 -06:00
Gustavo Madeira Santana
77e588ebc3 test: avoid bundled session normalizer fallback
Keep explicit session-key normalization on loaded channel plugins so
unknown provider contexts pass through without cold-loading bundled channel
runtimes. This preserves active plugin behavior and removes the slow
unknown-provider test path.
2026-04-17 16:41:46 -04:00
Gustavo Madeira Santana
5ae059db16 test: speed legacy state migration discovery
Keep bundled legacy migration discovery on narrow setup-entry surfaces so
state-migration tests and doctor cold paths avoid unrelated channel runtime
loads. Add targeted setup feature metadata, narrow Telegram/WhatsApp legacy
contracts, and a path-only pairing SDK helper.
2026-04-17 16:41:43 -04:00
Vincent Koc
a8a701291b refactor(auth): drop persisted external oauth ownership metadata 2026-04-17 13:28:54 -07:00
Vincent Koc
2c7c06c9b3 docs(changelog): note runtime-only external oauth import 2026-04-17 13:28:54 -07:00
Altay
d0cf6731aa fix(failover): classify INTERNAL 500 responses as retryable timeouts (#68238)
* Agents: treat Google INTERNAL 500 as timeout failover

(cherry picked from commit c2538523a22d39b65c6b4056ab4857ee84f06887)

* test(failover): narrow INTERNAL timeout patterns

* fix: document INTERNAL timeout retry guard

* fix: ignore plain status prose in server error classification

* fix(failover): preserve mixed server-error retry signals

* test(failover): dedupe internal status samples

* fix(failover): retry status prose with code 500

* fix: classify INTERNAL 500 responses as retryable timeouts

* fix: classify INTERNAL 500 responses as retryable timeouts

---------

Co-authored-by: Kosbling <github@kosbling.com>
Co-authored-by: Openbling <github@openbling.ai>
2026-04-17 23:24:26 +03:00
Vincent Koc
a001b5343f refactor(auth): make external cli oauth runtime-only 2026-04-17 13:14:17 -07:00
Gustavo Madeira Santana
50e71daaa0 test: keep inbound group policy tests hermetic 2026-04-17 15:50:41 -04:00
Gustavo Madeira Santana
8b76bcba90 test: avoid real Telegram config writes in retry tests 2026-04-17 15:50:41 -04:00
Gustavo Madeira Santana
c550642cde test: keep command registry native overrides hermetic 2026-04-17 15:50:39 -04:00
Peter Steinberger
fde25bfb8c test: isolate browser facade cache tests 2026-04-17 20:35:23 +01:00
Peter Steinberger
08e1eb7a9f test: narrow system run dispatch matrix 2026-04-17 20:27:52 +01:00
Peter Steinberger
c408bbe9c9 perf: cache browser plugin sdk facades 2026-04-17 20:26:14 +01:00
Peter Steinberger
809f42eeea test: trim UI and entry test overhead 2026-04-17 20:23:07 +01:00
Peter Steinberger
087f1584df test: streamline system run hotspot coverage 2026-04-17 20:18:01 +01:00
bwjoke
f7422e1fbc fix(failover): detect bare leading 402 assistant errors (#47579)
Merged via squash.

Prepared head SHA: ff336a0d97
Co-authored-by: bwjoke <1284814+bwjoke@users.noreply.github.com>
Co-authored-by: altaywtf <9790196+altaywtf@users.noreply.github.com>
Reviewed-by: @altaywtf
2026-04-17 22:06:55 +03:00
Peter Steinberger
169b68d709 test: narrow chat avatar fallback 2026-04-17 20:04:30 +01:00
Peter Steinberger
014eaa8492 test: merge env rejection invoke cases 2026-04-17 20:03:35 +01:00
Peter Steinberger
e9d052d728 test: merge shell payload plan checks 2026-04-17 20:01:58 +01:00
Peter Steinberger
f897025d9b test: narrow chat attachment rendering 2026-04-17 20:00:46 +01:00
Peter Steinberger
5f075d3d49 test: reuse session file fixture root 2026-04-17 19:57:19 +01:00
Gustavo Madeira Santana
8747351383 Media: keep inbound roots on media contracts 2026-04-17 14:56:47 -04:00
Peter Steinberger
bb5d9948c2 test: mock side result markdown 2026-04-17 19:56:17 +01:00
Peter Steinberger
be6dbd4084 test: mock chat sidebar markdown 2026-04-17 19:55:39 +01:00
Peter Steinberger
bbbb57f7f8 test: source install version helper only 2026-04-17 19:52:33 +01:00
Peter Steinberger
4dd999274b test: merge chat helper render tests 2026-04-17 19:51:43 +01:00
Peter Steinberger
7c862da6a1 test: split chat helper coverage 2026-04-17 19:50:39 +01:00
Peter Steinberger
125b1e0e20 test: reuse node-host runtime bins 2026-04-17 19:47:43 +01:00
Peter Steinberger
55c7776364 test: simplify acp and install test seams 2026-04-17 19:46:40 +01:00
Peter Steinberger
16e7f04a43 test: avoid login shell in install version test 2026-04-17 19:45:51 +01:00
Peter Steinberger
2e2f927d5d test: mock proxy capture store 2026-04-17 19:45:06 +01:00
Peter Steinberger
0a38098248 test: mock tts facade explicitly 2026-04-17 19:44:02 +01:00
Devin Robison
f61896b03c fix(cron): preserve untrusted awareness event labels (#68210)
* fix(cron): preserve untrusted awareness event labels

Keep isolated cron awareness summaries untrusted when they are promoted into the main session, and forward explicit trust downgrades through the gateway cron wrapper. Add focused regression coverage for both paths.

* changelog: note cron awareness untrusted-label preservation (#68210)
2026-04-17 12:43:48 -06:00
Peter Steinberger
2745e5b3bd test: narrow canvas and context hotspots 2026-04-17 19:42:59 +01:00
Gustavo Madeira Santana
f70b651b12 Tests: avoid media registry load for duplicate ids 2026-04-17 14:41:18 -04:00
Peter Steinberger
dadcfb574f test: trim surrogate chunk fixtures 2026-04-17 19:38:53 +01:00
Peter Steinberger
729feb4b99 test: reuse exec approval home fixture 2026-04-17 19:37:47 +01:00
Peter Steinberger
8c3a8f0b1b test: shrink context registry chunk coverage 2026-04-17 19:35:55 +01:00
Gustavo Madeira Santana
ee0c8177bf Fix canvas host header test type 2026-04-17 14:35:36 -04:00
Gustavo Madeira Santana
462074c4c2 Fix check type errors 2026-04-17 14:35:36 -04:00
Peter Steinberger
c0a9b694f3 test: reuse node host home fixture 2026-04-17 19:35:19 +01:00
Peter Steinberger
2c43c441b2 test: source minimal install helper fixture 2026-04-17 19:34:01 +01:00
Peter Steinberger
b39f3cf266 test: avoid polling settled acp reconnect 2026-04-17 19:31:40 +01:00
Peter Steinberger
79dfb4db69 test: shorten routing cache scalability case 2026-04-17 19:30:36 +01:00
Peter Steinberger
990bd81726 test: avoid canvas host socket setup 2026-04-17 19:29:42 +01:00
Devin Robison
90979d7c3e fix(feishu): resolve card-action chat type before dispatch (#68201)
* fix(feishu): resolve card-action chat type before dispatch

* changelog: resolve card-action chat type before dispatch (#68201)

* address review: prefer chat_mode over chat_type, add error-path tests

- Swap resolution order to check chat_mode (conversation type) before
  chat_type (privacy classification), since Feishu's chat_type can
  return "private" for private group chats which would be wrongly
  classified as p2p.
- Treat "topic" as group semantics in the normalizer.
- Add comment explaining the field semantics and why "private" maps
  to "p2p" (safe-failure direction).
- Add two error-path tests: API returns non-zero code, and API throws.

* map chat_type=public to group in normalizer

Feishu's chat_type can return "public" for public group chats.
Without this mapping the fallback resolver would miss it and default
to p2p, routing a group card action through DM handling.

* address Aisle: cache chat-type lookups and scrub log output

- Add a 30-minute TTL cache for chatId -> chatType so repeated card
  actions on the same chat skip the Feishu API call.
- Strip chatId, event.token, and raw error strings from log messages;
  use err.message instead of String(err) to avoid leaking stack traces
  or HTTP internals from the Feishu SDK.

* prune expired chat-type cache entries

Add pruneChatTypeCache() called on each lookup so expired entries are
evicted and the cache stays bounded in long-running processes.

* address Aisle: scope cache by account, cap size, sanitize logs

- Key cache by accountId:chatId to prevent cross-account contamination.
- Cap cache at 5000 entries and evict oldest when exceeded.
- Sanitize response.msg and err.message with CR/LF stripping and
  length cap before logging to prevent log injection.
2026-04-17 12:29:04 -06:00
Peter Steinberger
8eb577b361 test: slim routing cache rollover coverage 2026-04-17 19:27:52 +01:00
Peter Steinberger
7edce9c8fa test: reuse inline eval fixtures 2026-04-17 19:25:58 +01:00
Peter Steinberger
e75cd46ba6 test: isolate plugin tools mcp handlers 2026-04-17 19:25:20 +01:00
Peter Steinberger
38923d13a6 test: trim boundary and fixture hotspots 2026-04-17 19:22:38 +01:00
Peter Steinberger
b303b6c492 test: streamline navigation browser checks 2026-04-17 19:17:07 +01:00
Peter Steinberger
b6e55bf819 test: combine config and skill render checks 2026-04-17 19:13:44 +01:00
Peter Steinberger
c47c4b3574 test: trim remaining ui browser cases 2026-04-17 19:11:58 +01:00
Peter Steinberger
d155d578eb test: merge more ui render hotspots 2026-04-17 19:10:22 +01:00
Gustavo Madeira Santana
3a1e469732 QA: track scenario coverage intent 2026-04-17 14:05:49 -04:00
Gustavo Madeira Santana
f334ca2b50 Auto-reply: fast-path sandbox media root resolution 2026-04-17 14:05:49 -04:00
Peter Steinberger
e606656b56 test: merge remaining small render checks 2026-04-17 19:02:05 +01:00
Peter Steinberger
9feeb921f5 test: trim config form search render cases 2026-04-17 19:00:57 +01:00
Peter Steinberger
c050cdaa96 test: merge small view render cases 2026-04-17 18:59:08 +01:00
Peter Steinberger
783bb1f759 test: move query token checks to settings unit 2026-04-17 18:57:15 +01:00
Peter Steinberger
4ba12bd134 test: trim duplicated navigation auth cases 2026-04-17 18:55:35 +01:00
Peter Steinberger
f0c6b102be test: trim duplicate navigation and cron cases 2026-04-17 18:54:46 +01:00
Peter Steinberger
354dbf2161 test: reset config view state directly 2026-04-17 18:50:18 +01:00
Peter Steinberger
1a3a040cc3 test: move chat markdown sidebar to direct render 2026-04-17 18:49:15 +01:00
Gustavo Madeira Santana
6184f17c91 Twitch: add bundled setup entry (#68008)
Merged via squash.

Prepared head SHA: 59305356a0
Co-authored-by: gumadeiras <5599352+gumadeiras@users.noreply.github.com>
Reviewed-by: @gumadeiras
2026-04-17 13:49:08 -04:00
Peter Steinberger
bbac7773ff test: fold chat context browser check 2026-04-17 18:47:00 +01:00
Peter Steinberger
aaf9064a75 test: move chat image safety to direct render 2026-04-17 18:46:16 +01:00
Peter Steinberger
36b98f78b2 test: merge navigation singleton browser checks 2026-04-17 18:45:34 +01:00
Peter Steinberger
ddd2c2a602 test: merge chat side-result checks 2026-04-17 18:42:03 +01:00
Peter Steinberger
f7eb746081 test: merge cron history checks 2026-04-17 18:41:16 +01:00
Peter Steinberger
c2e4b47d7b test: merge responsive navigation shell checks 2026-04-17 18:40:07 +01:00
Vincent Koc
628e6cd446 docs(changelog): add codex oauth fixes 2026-04-17 10:39:21 -07:00
Peter Steinberger
5d8cecbe7d test: merge navigation routing cases 2026-04-17 18:39:00 +01:00
Gustavo Madeira Santana
2b08233a3e Tests: mock channel registry bundled fallback
Keep the registry fallback unit test on a minimal bundled fixture instead of loading the real Google Chat plugin. Doctor capability metadata remains covered by the doctor channel capability tests.
2026-04-17 13:38:24 -04:00
Gustavo Madeira Santana
a464f5926b Secrets: avoid broad web search discovery for single plugin config
Add an Exa web-search contract artifact and use single bundled plugin-scoped webSearch config as a provider hint. This keeps runtime secret resolution on metadata-only surfaces instead of importing full provider tool implementations.
2026-04-17 13:38:24 -04:00
Peter Steinberger
20cf51169b test: merge config view browser checks 2026-04-17 18:37:53 +01:00
Vincent Koc
eed71160ae fix(status): align oauth health with runtime 2026-04-17 10:36:51 -07:00
Peter Steinberger
5c2f4afcce test: merge chat context notice checks 2026-04-17 18:36:33 +01:00
Peter Steinberger
1df50183b2 test: merge chat image safety cases 2026-04-17 18:35:52 +01:00
Peter Steinberger
0747a9c85a test(discord): isolate debug proxy env 2026-04-17 18:35:06 +01:00
Peter Steinberger
7876e3e736 test(cron): remove fast retry timer dependency 2026-04-17 18:35:06 +01:00
Peter Steinberger
1519b006b8 test(auth): isolate provider alias registry mock 2026-04-17 18:35:06 +01:00
Peter Steinberger
f93b7da4c4 test: merge chat attachment cases 2026-04-17 18:34:19 +01:00
Gustavo Madeira Santana
9bcf8f8243 Configure: defer channel status until selection (#68007)
Merged via squash.

Prepared head SHA: 24cafcd5fe
Co-authored-by: gumadeiras <5599352+gumadeiras@users.noreply.github.com>
Co-authored-by: gumadeiras <5599352+gumadeiras@users.noreply.github.com>
Reviewed-by: @gumadeiras
2026-04-17 13:34:13 -04:00
Peter Steinberger
474b08bfbd test: merge dreaming view UI cases 2026-04-17 18:30:08 +01:00
Gustavo Madeira Santana
00fadb978f Tests: narrow bundled shape guard source check 2026-04-17 13:28:49 -04:00
Peter Steinberger
f665c767e6 test: drop duplicate markdown cases 2026-04-17 18:28:40 +01:00
Peter Steinberger
81d6cf9c82 test: merge cron view UI cases 2026-04-17 18:25:20 +01:00
Vincent Koc
f513bae67e fix(oauth): make codex tls preflight advisory 2026-04-17 10:24:00 -07:00
Peter Steinberger
76c8db3766 test: merge chat view UI cases 2026-04-17 18:23:24 +01:00
Vincent Koc
3ed0995fa9 fix(auth): keep codex oauth canonical in openclaw 2026-04-17 10:20:43 -07:00
Peter Steinberger
df06343dfa test: merge mobile navigation UI checks 2026-04-17 18:19:37 +01:00
Peter Steinberger
8448569aca test: narrow skills config imports 2026-04-17 18:16:43 +01:00
Devin Robison
114b87caf2 fix(macos): require trusted SSH host keys (#68199)
* fix(macos): require trusted SSH host keys

* chore(changelog): add macOS SSH strict host-key entry
2026-04-17 11:11:10 -06:00
Peter Steinberger
dfca5bd0fe test: isolate oauth refresh queue mocks 2026-04-17 18:10:07 +01:00
Peter Steinberger
89d3117ad0 test: narrow auth profile runtime mocks 2026-04-17 18:06:01 +01:00
Gustavo Madeira Santana
42817a1707 Tests: isolate OAuth mirror external auth lookup
Use the existing external auth test hook and a lightweight OAuth package mock so mirror-refresh coverage does not load provider runtime work while seeding test stores.
2026-04-17 12:50:52 -04:00
Peter Steinberger
8c249a8cca fix(matrix): keep guarded transport mockable 2026-04-17 17:44:11 +01:00
Gustavo Madeira Santana
8d7a722487 Tests: register models command text surfaces
Keep models command tests inside the in-memory channel registry for Discord and WhatsApp so text-surface assertions do not load bundled channel runtimes.
2026-04-17 12:40:04 -04:00
Peter Steinberger
f8a0ae0b08 test: merge redundant navigation shell assertions 2026-04-17 17:36:29 +01:00
Gustavo Madeira Santana
06e3d53c8a Tests: avoid bundled channel fallback in adapter test
Register a lightweight Telegram test plugin so the default-adapter assertion stays inside the in-memory registry instead of loading the real bundled channel runtime.
2026-04-17 12:34:38 -04:00
Peter Steinberger
7815d25eef fix: keep Matrix transport tests on mocked fetch 2026-04-17 17:33:34 +01:00
Peter Steinberger
8caad53f57 test(matrix): mock runtime fetch seam 2026-04-17 17:33:01 +01:00
Peter Steinberger
769198e67e perf: skip Matrix secret resolver for plain credentials 2026-04-17 17:31:00 +01:00
Peter Steinberger
41ef752dd8 fix(extensions): guard channel runtime fetches 2026-04-17 17:28:21 +01:00
Peter Steinberger
c580933623 test: shorten Matrix thread binding waits 2026-04-17 17:26:46 +01:00
Peter Steinberger
b9d5c1a58b test: narrow Matrix storage path test imports 2026-04-17 17:24:01 +01:00
Peter Steinberger
1d26f0cc6e test: flush navigation scroll frames directly 2026-04-17 17:21:49 +01:00
Peter Steinberger
75e09e21f2 test: remove gateway handshake waits 2026-04-17 17:20:26 +01:00
Peter Steinberger
a027a40c90 test(plugins): allow secret input runtime sdk subpath 2026-04-17 17:18:12 +01:00
Peter Steinberger
97f713f459 test(agents): isolate compaction token estimator mocks 2026-04-17 17:18:12 +01:00
Peter Steinberger
c0a16650d5 test(commands): fix command fixture typing 2026-04-17 17:18:12 +01:00
Peter Steinberger
a71b810e43 fix(plugin-sdk): expose session store runtime helpers 2026-04-17 17:18:12 +01:00
Peter Steinberger
ccc23f6cb6 test: trim navigation scroll fixture 2026-04-17 17:18:02 +01:00
Gustavo Madeira Santana
c66703300a Tests: narrow bootstrap routing coverage 2026-04-17 12:17:41 -04:00
Peter Steinberger
79cd5ed368 test: split Matrix client config tests 2026-04-17 17:16:21 +01:00
Peter Steinberger
54d9a09912 perf: narrow Matrix monitor reply imports 2026-04-17 17:14:44 +01:00
Peter Steinberger
24f8d6470e test: split chat session control tests 2026-04-17 17:11:28 +01:00
Peter Steinberger
73d8d3b2eb test: remove duplicate Matrix runtime API check 2026-04-17 17:08:37 +01:00
Peter Steinberger
d851f9e816 perf: narrow Matrix thread binding runtime imports 2026-04-17 17:04:31 +01:00
Peter Steinberger
d7f9f67296 perf: narrow Matrix onboarding resolution test 2026-04-17 16:58:19 +01:00
Peter Steinberger
14c4d6457a perf: narrow Matrix account runtime imports 2026-04-17 16:53:46 +01:00
Peter Steinberger
1fad8efa12 perf: split chat session controls 2026-04-17 16:45:37 +01:00
Peter Steinberger
7b27d08e56 perf: lazy load system run config 2026-04-17 16:39:24 +01:00
Gustavo Madeira Santana
8de7aefe0a Tests: narrow embedded timeout wiring 2026-04-17 11:37:46 -04:00
Gustavo Madeira Santana
d6c90b5af1 Tests: avoid memory-search cold plugin loads 2026-04-17 11:37:46 -04:00
Peter Steinberger
2535331e94 perf: remove Matrix test polling 2026-04-17 16:32:57 +01:00
Peter Steinberger
acace04c35 perf: remove Matrix auth retry waits in tests 2026-04-17 16:30:04 +01:00
Chunyue Wang
0b3d876e74 fix(codex): prevent gateway crash when app-server subprocess terminates abruptly (#67947)
Fixes openclaw#67886. Handles stdin EPIPE in CodexAppServerClient by attaching an error handler, guarding writeMessage against writes after close, and aligning closeWithError cleanup with close.
2026-04-17 23:28:37 +08:00
Peter Steinberger
d565c2cc34 perf: add lightweight secret input runtime 2026-04-17 16:28:15 +01:00
Peter Steinberger
5f3bb53788 perf: skip missing Matrix IDB snapshot locks 2026-04-17 16:24:29 +01:00
Peter Steinberger
a954e2fb46 perf: narrow Matrix thread binding test imports 2026-04-17 16:21:48 +01:00
Peter Steinberger
b2b835fb18 perf: reduce Matrix monitor wait polling 2026-04-17 16:19:29 +01:00
Tak Hoffman
62703d8430 fix(bootstrap): workspace bootstrap prompt routing (#68000)
* fix(bootstrap): workspace bootstrap prompt routing

* Fix bootstrap routing edge cases

* Refine bootstrap mode routing and reset prompts

* Fix bootstrap workspace routing for embedded runs

* Fix embedded bootstrap compile follow-up

* Align bare reset bootstrap file access

* Honor reset override model for bootstrap gating

* Align chat reset bootstrap topology
2026-04-17 10:18:50 -05:00
Peter Steinberger
4d7d14cfa7 perf: flush Matrix event tests deterministically 2026-04-17 16:18:27 +01:00
Peter Steinberger
bac3d26fe7 perf: narrow Matrix reaction approval imports 2026-04-17 16:17:01 +01:00
Peter Steinberger
2a0a498b0d perf: speed Matrix handler tests 2026-04-17 16:14:28 +01:00
Peter Steinberger
3b81bf4c7c perf: narrow Matrix handler session store import 2026-04-17 16:09:10 +01:00
Peter Steinberger
40c9da1d57 perf: avoid Matrix entry full-channel load in tests 2026-04-17 16:06:26 +01:00
Peter Steinberger
48aa076d12 perf: optimize remaining core tests 2026-04-17 16:05:10 +01:00
Peter Steinberger
310b5e4f6a test: reduce core command hotspots 2026-04-17 16:05:10 +01:00
Peter Steinberger
418056f7a0 perf: narrow plugin SDK import surfaces 2026-04-17 16:05:09 +01:00
Peter Steinberger
af954a81d1 perf: optimize bundled extension tests 2026-04-17 16:05:09 +01:00
Peter Steinberger
605cb60586 perf: optimize Matrix test boundaries 2026-04-17 16:05:09 +01:00
Peter Steinberger
a861da41b5 test: trim CLI and doctor hotspots 2026-04-17 16:05:09 +01:00
Peter Steinberger
199bb1fe05 test: mock auth alias registry in onboarding auth 2026-04-17 16:05:09 +01:00
Peter Steinberger
d3e12cee7e test: move gateway token cases to unit seam 2026-04-17 16:05:09 +01:00
Peter Steinberger
d3b70f9823 test: tighten message and onboarding hotspots 2026-04-17 16:05:09 +01:00
Peter Steinberger
f7f88e52e4 test: trim doctor and auth choice hotspots 2026-04-17 16:05:09 +01:00
Peter Steinberger
675eb38ad0 test: keep provider auth onboarding config in memory 2026-04-17 16:05:09 +01:00
Peter Steinberger
a90daa5759 test: narrow channel token summary coverage 2026-04-17 16:05:09 +01:00
Peter Steinberger
e477125608 test: merge repeated onboarding auth cases 2026-04-17 16:05:09 +01:00
Peter Steinberger
7995d43625 test: merge provider auth onboarding cases 2026-04-17 16:05:09 +01:00
Peter Steinberger
68cf9e52a2 test: narrow auth credential fixtures 2026-04-17 16:05:09 +01:00
Peter Steinberger
e53a8bd865 test: trim provider auth onboarding fixtures 2026-04-17 16:05:09 +01:00
Peter Steinberger
290371399f test: mock agent command runtime seams 2026-04-17 16:05:09 +01:00
Peter Steinberger
e2099301c5 test: narrow auth choice provider fixtures 2026-04-17 16:05:09 +01:00
Peter Steinberger
f810cc4d58 test: lower matrix bind coverage boundary 2026-04-17 16:05:09 +01:00
Peter Steinberger
efb37f8949 test: narrow channels status command mocks 2026-04-17 16:05:09 +01:00
Peter Steinberger
c93b2540ec test: mock status command scan seam 2026-04-17 16:05:09 +01:00
Peter Steinberger
ab726235bd test: narrow agents bind and provider auth cases 2026-04-17 16:05:09 +01:00
Peter Steinberger
824b5e4d91 test: mock agent runtime secret targets 2026-04-17 16:05:09 +01:00
Peter Steinberger
bb70b41340 test: split lightweight agent session coverage 2026-04-17 16:05:09 +01:00
Peter Steinberger
a9fab78f64 test: trim duplicate auth choice integration cases 2026-04-17 16:05:09 +01:00
Peter Steinberger
e4f04d92a3 test: move custom onboarding edge cases down-stack 2026-04-17 16:05:09 +01:00
Peter Steinberger
24ef516879 test: trim duplicate direct provider token cases 2026-04-17 16:05:09 +01:00
Peter Steinberger
2e3ef1b9e1 fix: pass message routing context to send actions 2026-04-17 16:05:09 +01:00
Peter Steinberger
4ac8b08265 test: mock agent runtime config imports 2026-04-17 16:05:09 +01:00
Peter Steinberger
e00f9c7a9d test: trim custom provider onboarding duplicates 2026-04-17 16:05:09 +01:00
Peter Steinberger
e19e94ef07 test: split channels list auth profile coverage 2026-04-17 16:05:09 +01:00
Peter Steinberger
cfba24fa3c test: move open policy repair cases to unit seam 2026-04-17 16:05:08 +01:00
Peter Steinberger
8ebb3ff0d4 test: narrow message command mocks 2026-04-17 16:05:08 +01:00
Peter Steinberger
4451e8479a test: mock channels status command seam 2026-04-17 16:05:08 +01:00
Peter Steinberger
271fc360e7 test: mock message command config seam 2026-04-17 16:05:08 +01:00
Peter Steinberger
82355d1d9f test: isolate agent runtime config imports 2026-04-17 16:05:08 +01:00
Peter Steinberger
769a09842d test: narrow channels command test imports 2026-04-17 16:05:08 +01:00
Gustavo Madeira Santana
82fe6f50ef QA: organize scenarios by theme 2026-04-17 11:03:47 -04:00
Val Alexander
a45ebf3281 fix(ui): reset settings scroll and align details headers (#68150) thanks @BunsDev
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
2026-04-17 09:55:30 -05:00
Val Alexander
be7a415eb0 fix: preserve hello-ok scopes for reused device tokens (#68039) 2026-04-17 03:20:48 -05:00
Val Alexander
f377db1015 feat: add macOS screen snapshots for monitor preview (#67954) thanks @BunsDev
Co-authored-by: Val Alexander <68980965+BunsDev@users.noreply.github.com>
2026-04-17 02:58:21 -05:00
Val Alexander
0b6c39be18 fix: report shared auth scopes in hello-ok (#67810) thanks @BunsDev
Co-authored-by: Val Alexander <bunsthedev@gmail.com>
2026-04-17 02:48:30 -05:00
Gustavo Madeira Santana
3ea1bf4232 Auto-reply: avoid eager bundled route fallback 2026-04-17 03:36:13 -04:00
Gustavo Madeira Santana
54e4e16844 Tests: narrow session binding contract setup 2026-04-17 03:36:13 -04:00
J. Tyler Bittner
00951dc9f9 fix(macOS): enable undo/redo in webchat composer text input (#34962)
* fix(macOS): enable undo/redo in webchat composer text input

Set `allowsUndo = true` on ChatComposerNSTextView in makeNSView().
NSTextView defaults allowsUndo to false, which prevented Cmd+Z and
the Edit menu Undo/Redo items from functioning.

Fixes #34898

* fix(macos): enable webchat composer undo/redo (#34962) (thanks @tylerbittner)

---------

Co-authored-by: Nimrod Gutman <nimrod.gutman@gmail.com>
2026-04-17 10:07:20 +03:00
Gustavo Madeira Santana
82b529a6d9 Tests: speed up channel setup promotion 2026-04-17 02:58:24 -04:00
Gustavo Madeira Santana
5775fe272a Docs: refresh agent instructions 2026-04-17 02:46:38 -04:00
Viz
8e79080bef fix(auth): serialize OAuth refresh across agents to fix #26322 (#67876) 2026-04-16 23:44:03 -07:00
Peter Steinberger
7d4f1a6777 test: allow ollama public surface boundary test 2026-04-17 07:28:09 +01:00
Gustavo Madeira Santana
89706d323c Docs: add test performance guardrails 2026-04-17 02:23:49 -04:00
Gustavo Madeira Santana
e4c4f955b3 Tests: restore context-engine usage proof 2026-04-17 02:23:49 -04:00
Gustavo Madeira Santana
74c198f2e8 Tests: slim context engine runtime coverage 2026-04-17 02:23:49 -04:00
Peter Steinberger
0ee5baf6c5 ci: retry failed custom checkouts 2026-04-17 07:20:51 +01:00
Peter Steinberger
1ffc02e930 test: trim duplicate provider auth onboarding cases 2026-04-17 07:19:23 +01:00
EE
1ce2596195 matrix: fix sessions_spawn --thread subagent session spawning (#67643)
Merged via squash.

Prepared head SHA: 1e5127e217
Co-authored-by: eejohnso-ops <238848106+eejohnso-ops@users.noreply.github.com>
Co-authored-by: gumadeiras <5599352+gumadeiras@users.noreply.github.com>
Reviewed-by: @gumadeiras
2026-04-17 02:17:56 -04:00
Peter Steinberger
857b9cd326 test: reduce auth choice fixture churn 2026-04-17 07:15:28 +01:00
Peter Steinberger
9d5ab4a54c test: mock health status config boundaries 2026-04-17 07:15:28 +01:00
Peter Steinberger
299694d721 test: mock onboard config io boundary 2026-04-17 07:15:28 +01:00
Peter Steinberger
2713089220 test: mock legacy state plugin boundaries 2026-04-17 07:15:28 +01:00
Peter Steinberger
b945248650 test: mock channel install boundaries 2026-04-17 07:15:28 +01:00
Peter Steinberger
b1a3ad49a4 test: mock doctor preview channel boundaries 2026-04-17 07:15:27 +01:00
Peter Steinberger
c66f16ac55 test: trim doctor command hotspots 2026-04-17 07:15:27 +01:00
Peter Steinberger
92859357bb test: isolate agent auth and spawn hotspots 2026-04-17 07:15:27 +01:00
Peter Steinberger
dd9d2ebd01 test: stabilize MCP startup disposal race 2026-04-17 07:15:27 +01:00
Peter Steinberger
5817a76236 test: merge browser contract server suites 2026-04-17 07:15:27 +01:00
Peter Steinberger
a0d9598425 test: narrow ollama provider discovery setup 2026-04-17 07:15:27 +01:00
Peter Steinberger
daaebb8558 test: split browser snapshot target helper 2026-04-17 07:15:27 +01:00
Peter Steinberger
f57ce21d73 test: trim process-backed agent assertions 2026-04-17 07:15:27 +01:00
Peter Steinberger
b71c91022b test: collapse exec preflight parser cases 2026-04-17 07:15:27 +01:00
Peter Steinberger
d07c921ae3 test: trim provider extra-param imports 2026-04-17 07:15:27 +01:00
Peter Steinberger
132d3c76a0 test: reuse browser server harness imports 2026-04-17 07:15:27 +01:00
Peter Steinberger
35dcd06764 test: trim agent test hotspots 2026-04-17 07:15:27 +01:00
Gustavo Madeira Santana
7ae670e501 Tests: fast-path Slack message tool discovery 2026-04-17 02:00:26 -04:00
Gustavo Madeira Santana
878f2122e5 Tests: fast-path Matrix ACP thread binding 2026-04-17 02:00:26 -04:00
Gustavo Madeira Santana
807c6648f9 Tests: fast-path gateway auth bypass discovery 2026-04-17 02:00:26 -04:00
Gustavo Madeira Santana
178c36532d Tests: isolate perf-sensitive env state 2026-04-17 02:00:26 -04:00
Ayaan Zaidi
26f7198eda fix(memory-core): preserve vector dims on readonly recovery 2026-04-17 11:22:56 +05:30
Ayaan Zaidi
bec52e5f7e fix: clear compaction replay after visible boundaries (#67993) 2026-04-17 11:18:22 +05:30
Ayaan Zaidi
5aad79571e fix(telegram): clear compaction replay after visible boundaries 2026-04-17 11:18:22 +05:30
Ayaan Zaidi
671579663b fix: preserve post-stream error payloads (#67991)
* fix(reply): preserve post-stream error payloads

* fix: preserve post-stream error payloads (#67991)
2026-04-17 11:11:37 +05:30
Rubén Cuevas
7b0e950e09 fix: dedupe degraded sqlite-vec warnings (#67898) (thanks @rubencu)
* Agents: dedupe bootstrap truncation warnings

* Memory: dedupe sqlite-vec degradation warnings

* Memory: align degraded vector warning

* test(memory-core): remove stale vector warning arg

* fix(memory-core): reset degraded warning on vector reset

* fix(memory-core): preserve warning latch across reindex rollback

* fix: dedupe degraded sqlite-vec warnings (#67898) (thanks @rubencu)

---------

Co-authored-by: Ayaan Zaidi <hi@obviy.us>
2026-04-17 11:09:14 +05:30
Chinar Amrutkar
8205de84a9 fix: clear stale telegram ACP bindings on startup (#67822) (thanks @chinar-amrutkar)
* fix(telegram): clean up thread bindings to stale/failed ACP sessions on startup

When loading persisted thread bindings on manager creation, validate each
ACP session against the session store. Remove bindings where:
- Session entry doesn't exist (deleted externally)
- Session status is failed/killed/timeout
- ACP runtime state is 'error'

This addresses issue #60102 where Telegram DMs remained routed to stale
ACP sessions even after restart, because the binding file persisted
across restarts without validating the target session was still valid.

* fix(telegram): guard against null session entry and transient store read failures

Address review comments on PR #67822:

1. Skip bindings when readAcpSessionEntry returns null or when
   session store is temporarily unreadable (storeReadFailed: true).
   Without this, a transient I/O error would mark all ACP bindings
   as stale and delete them on every startup.

2. Only set needsPersist when bindings were actually removed.
   Previously, stale session keys from OTHER accounts could set
   needsPersist=true even when zero bindings were removed for
   the current account — causing spurious disk writes.

Also clean up redundant optional chaining on entry.status now
that we guard against undefined/nullable sessionEntry.

* perf(telegram): dedupe ACP session reads in startup cleanup

Cache readAcpSessionEntry calls by targetSessionKey. Multiple bindings
to the same ACP session now result in a single session store read instead
of one read per binding.

Addresses chatgpt-codex-connector P2 review comment on PR #67822.

* fix(telegram): skip non-ACP session keys in stale binding cleanup

Address chatgpt-codex-connector P1 review comment on PR #67822:

Plugin-bound Telegram conversations use "plugin-binding:*" keys
with targetKind === "acp", but these are NOT ACP runtime sessions.
readAcpSessionEntry() returns no entry for them, so !sessionEntry.entry
would classify them as stale and delete them on every startup.

Now checks isAcpSessionKey(binding.targetSessionKey) to skip plugin-bound
sessions from the stale session cleanup scan.

Also clarifies the comment to explain why we use targetKind === "acp"
// together with isAcpSessionKey() check.

* fix(telegram): import isAcpSessionKey from sessions/session-key-utils

isAcpSessionKey is not re-exported from openclaw/plugin-sdk/routing.
Fix import to use the correct subpath: openclaw/sessions/session-key-utils.

Addresses chatgpt-codex-connector P1 review comment on PR #67822.

* fix(telegram): import from relative path, remove unused variable

- Import isAcpSessionKey from relative path ../../sessions/session-key-utils.js
  (not openclaw/sessions/session-key-utils which doesn't exist)
- Remove unused 'bindings' variable in for-of loop

Addresses CI failures on PR #67822.

* fix(telegram): export isAcpSessionKey from plugin-sdk/routing

isAcpSessionKey lives in src/routing/session-key.ts, which is already
exported via openclaw/plugin-sdk/routing. Re-export it from routing.ts
so extensions can import via the public plugin-sdk path.

Fixes chatgpt-codex-connector P1: relative path ../../sessions/session-key-utils.js
doesn't exist in the build output, making the Telegram extension fail
module resolution before startup cleanup can run.

* test(telegram): cover startup ACP binding cleanup

* fix: clear stale telegram ACP bindings on startup (#67822) (thanks @chinar-amrutkar)

---------

Co-authored-by: Ayaan Zaidi <hi@obviy.us>
2026-04-17 11:03:36 +05:30
Rubén Cuevas
c65f356ddc fix: keep telegram transient preview across compaction retry (#66939) (thanks @rubencu)
* fix(telegram): keep transient previews across compaction

* test(telegram): cover suppressed approval previews after compaction

* fix(telegram): preserve delayed message-start boundaries

* fix: keep telegram transient preview across compaction retry (#66939) (thanks @rubencu)

---------

Co-authored-by: Ayaan Zaidi <hi@obviy.us>
2026-04-17 10:57:46 +05:30
Rubén Cuevas
7e18c07e41 fix: dedupe repeated bootstrap truncation warnings (#67906) (thanks @rubencu)
* Agents: dedupe bootstrap truncation warnings

* Agents: normalize bootstrap warning cache bookkeeping

* fix(agents): scope bootstrap warning dedupe by workspace

* refactor(agents): simplify bootstrap warning wrapper

---------

Co-authored-by: Ayaan Zaidi <hi@obviy.us>
2026-04-17 10:54:15 +05:30
Ayaan Zaidi
3fe8b24c4e fix: note plugin registration rollback in changelog (#67941) 2026-04-17 10:14:00 +05:30
Ayaan Zaidi
c95507978f fix(plugins): tighten register rollback 2026-04-17 10:14:00 +05:30
Ayaan Zaidi
59d07f0ab4 fix(plugins): roll back failed register globals 2026-04-17 10:14:00 +05:30
Ayaan Zaidi
5c1d6feb33 test(plugins): fix sync register call sites 2026-04-17 10:14:00 +05:30
Ayaan Zaidi
e8fd148437 fix(plugins): roll back failed register side effects 2026-04-17 10:14:00 +05:30
Ayaan Zaidi
2a283e87a7 fix(plugins): enforce synchronous registration 2026-04-17 10:14:00 +05:30
Ayaan Zaidi
15b2827fc1 test(gateway): stabilize canvas auth fetch retries 2026-04-17 09:59:12 +05:30
Ayaan Zaidi
65645ec54f fix(agents): refresh bundle command discovery 2026-04-17 09:59:03 +05:30
Gustavo Madeira Santana
e8ae3901b6 Tests: scope grouped benchmark artifacts 2026-04-16 23:45:57 -04:00
Gustavo Madeira Santana
8e444ac5a6 Tests: add grouped performance report benchmark 2026-04-16 23:43:06 -04:00
Ayaan Zaidi
6b45ba88a1 fix: reuse shared plugin discovery cache across workspaces (#67940) 2026-04-17 08:54:58 +05:30
Ayaan Zaidi
353950894a test(plugins): address discovery review feedback 2026-04-17 08:54:58 +05:30
Ayaan Zaidi
9da4d5f5df fix(plugins): reuse shared discovery cache 2026-04-17 08:54:58 +05:30
Peter Steinberger
c6af0437c9 test: avoid postinstall fixture installs 2026-04-17 04:10:55 +01:00
Peter Steinberger
a2f2e5738e test: trim messaging test hotspots 2026-04-17 04:10:55 +01:00
chaoliang yan
35fb3f7e1c fix: preserve models.json baseUrls on regen (#67893) (thanks @lawrence3699)
* models-config: preserve existing models.json baseUrls

* test: distill models-config baseUrl regression test

* fix: preserve models.json baseUrls on regen (#67893) (thanks @lawrence3699)

---------

Co-authored-by: lawrence3699 <lawrence3699@users.noreply.github.com>
Co-authored-by: Ayaan Zaidi <hi@obviy.us>
2026-04-17 08:32:35 +05:30
chaoliang yan
a189394590 fix: guard WhatsApp setup prompt values (#67895) (thanks @lawrence3699)
* fix(whatsapp): guard setup prompt values

* fix(whatsapp): preserve allowFrom invalid input details

* fix: guard WhatsApp setup prompt values (#67895) (thanks @lawrence3699)

---------

Co-authored-by: lawrence3699 <lawrence3699@users.noreply.github.com>
Co-authored-by: Ayaan Zaidi <hi@obviy.us>
2026-04-17 08:12:55 +05:30
Ayaan Zaidi
685f9903ec fix: unify responses api capability detection (#67918)
* fix: unify responses api capability detection

* fix: unify responses api capability detection (#67918)
2026-04-17 08:03:19 +05:30
Peter Steinberger
24431e5114 build: declare qa-lab aimock runtime dependency 2026-04-17 02:57:18 +01:00
Peter Steinberger
ee856ab31f test: speed up safe-bins exec harness 2026-04-17 02:57:18 +01:00
Peter Steinberger
acd86a06cd test: preserve tool helpers in embedded runner mocks 2026-04-17 02:57:18 +01:00
Peter Steinberger
77e6e4cf87 refactor: move memory embeddings into provider plugins 2026-04-17 02:57:18 +01:00
918 changed files with 35229 additions and 30104 deletions

View File

@@ -52,7 +52,11 @@ function ghGraphQL(query, options = {}) {
function failOnGraphQLFailure(result, message) {
if (result?.gh_failed) {
const details = (result.stderr || result.stdout || `gh exited with status ${result.status}`).trim();
const details = (
result.stderr ||
result.stdout ||
`gh exited with status ${result.status}`
).trim();
fail(`${message}: ${details}`);
}
if (Array.isArray(result?.errors) && result.errors.length > 0) {
@@ -73,9 +77,7 @@ function formatGraphQLAfterClause(cursor) {
}
function findDiscussionCommentNode(nodes, discussionCommentDbId) {
return (
nodes.find((node) => String(node.databaseId) === String(discussionCommentDbId)) || null
);
return nodes.find((node) => String(node.databaseId) === String(discussionCommentDbId)) || null;
}
function fetchDiscussionReplyPage(commentNodeId, cursor) {
@@ -169,9 +171,13 @@ function fetchDiscussionComment(discussionNumber, discussionCommentDbId) {
while (!reply && hasMoreReplies) {
const replyPage = fetchDiscussionReplyPage(topLevelComment.id, replyCursor);
failOnGraphQLFailure(replyPage, `Failed to fetch replies for discussion comment ${topLevelComment.id}`);
failOnGraphQLFailure(
replyPage,
`Failed to fetch replies for discussion comment ${topLevelComment.id}`,
);
const replies = replyPage?.data?.node?.replies;
if (!replies) fail(`Failed to paginate replies for discussion comment ${topLevelComment.id}`);
if (!replies)
fail(`Failed to paginate replies for discussion comment ${topLevelComment.id}`);
reply = findDiscussionCommentNode(replies.nodes, discussionCommentDbId);
hasMoreReplies = replies.pageInfo.hasNextPage;
@@ -189,9 +195,7 @@ function fetchDiscussionComment(discussionNumber, discussionCommentDbId) {
}
function createDiscussionComment(discussionNodeId, body, replyToNodeId) {
const replyToClause = replyToNodeId
? `, replyToId: "${escapeGraphQLString(replyToNodeId)}"`
: "";
const replyToClause = replyToNodeId ? `, replyToId: "${escapeGraphQLString(replyToNodeId)}"` : "";
const result = ghGraphQL(
`mutation { addDiscussionComment(input: { discussionId: "${escapeGraphQLString(discussionNodeId)}"${replyToClause}, body: "${escapeGraphQLString(body)}" }) { comment { id url } } }`,
);
@@ -261,7 +265,10 @@ function cmdFetchContent(locationJson) {
const discussionNumber = urlMatch[1];
const discussionCommentDbId = urlMatch[2];
const { discussionId, comment } = fetchDiscussionComment(discussionNumber, discussionCommentDbId);
const { discussionId, comment } = fetchDiscussionComment(
discussionNumber,
discussionCommentDbId,
);
if (!comment)
fail(
`Discussion comment #${discussionCommentDbId} not found in discussion #${discussionNumber}`,

View File

@@ -677,9 +677,10 @@ jobs:
-c protocol.version=2 \
-c "http.https://github.com/.extraheader=AUTHORIZATION: basic ${auth_header}" \
fetch --no-tags --prune --no-recurse-submodules --depth=1 origin \
"+${CHECKOUT_SHA}:refs/remotes/origin/ci-target"
"+${CHECKOUT_SHA}:refs/remotes/origin/ci-target" || return 1
git -C "$workdir" checkout --force --detach "$CHECKOUT_SHA"
git -C "$workdir" checkout --force --detach "$CHECKOUT_SHA" || return 1
test -f "$workdir/.github/actions/setup-node-env/action.yml" || return 1
echo "checkout attempt ${attempt}/2 succeeded"
}

View File

@@ -1,12 +1,6 @@
name: CodeQL
on:
pull_request:
branches: [main]
paths-ignore:
- "**/*.md"
- "**/*.mdx"
- "LICENSE"
workflow_dispatch:
schedule:
- cron: "0 6 * * *"

View File

@@ -1,12 +1,12 @@
# Repository Guidelines
- Repo: https://github.com/openclaw/openclaw
- In chat replies, file references must be repo-root relative only (example: `src/telegram/index.ts:80`); never absolute paths or `~/...`.
- In chat replies, file references must be repo-root relative only (example: `extensions/telegram/src/index.ts:80`); never absolute paths or `~/...`.
- Do not edit files covered by security-focused `CODEOWNERS` rules unless a listed owner explicitly asked for the change or is already reviewing it with you. Treat those paths as restricted surfaces, not drive-by cleanup.
## Project Structure & Module Organization
- Source code: `src/` (CLI wiring in `src/cli`, commands in `src/commands`, web provider in `src/provider-web.ts`, infra in `src/infra`, media pipeline in `src/media`).
- Source code: `src/` (CLI wiring in `src/cli`, commands in `src/commands`, infra in `src/infra`, media pipeline in `src/media`, web provider helpers in `src/web` and `src/plugins/web-*provider*.ts`).
- Tests: colocated `*.test.ts`.
- Docs: `docs/` (images, queue, Pi config). Built output lives in `dist/`.
- Nomenclature: use "plugin" / "plugins" in docs, UI, changelogs, and contributor guidance. The bundled workspace plugin tree remains the internal package layout to avoid repo-wide churn from a rename.
@@ -17,8 +17,8 @@
- Installers served from `https://openclaw.ai/*`: live in the sibling repo `../openclaw.ai` (`public/install.sh`, `public/install-cli.sh`, `public/install.ps1`).
- Messaging channels: always consider **all** built-in + extension channels when refactoring shared logic (routing, allowlists, pairing, command gating, onboarding, docs).
- Core channel docs: `docs/channels/`
- Core channel code: `src/telegram`, `src/discord`, `src/slack`, `src/signal`, `src/imessage`, `src/web` (WhatsApp web), `src/channels`, `src/routing`
- Bundled plugin channels: the workspace plugin tree (for example Matrix, Zalo, ZaloUser, Voice Call)
- Core channel code: `src/channels`, `src/routing`, `src/web`
- Bundled plugin channels: `extensions/<channel>/` (for example Discord, Telegram, Slack, Matrix, Zalo, ZaloUser, Voice Call)
- When adding channels/plugins/apps/docs, update `.github/labeler.yml` and create matching GitHub labels (use existing channel/plugin label colors).
## Architecture Boundaries
@@ -73,7 +73,7 @@
- `hooks.internal.entries` is the canonical public hook config model. `hooks.internal.handlers` is compatibility-only input and must not be re-exposed in public schema/help/baseline surfaces.
- Bundled plugin contract boundary:
- Public docs: `docs/plugins/architecture.md`, `docs/plugins/manifest.md`, `docs/plugins/sdk-overview.md`
- Definition files: `src/plugins/contracts/registry.ts`, `src/plugins/types.ts`, `src/plugins/public-artifacts.ts`
- Definition files: `src/plugins/contracts/registry.ts`, `src/plugins/types.ts`, `src/plugins/public-surface-loader.ts`, `src/plugins/public-surface-runtime.ts`, `src/plugins/provider-public-artifacts.ts`, `src/plugins/web-provider-public-artifacts.ts`
- Rule: keep manifest metadata, runtime registration, public SDK exports, and contract tests aligned. Do not create a hidden path around the declared plugin interfaces.
- Extension test boundary:
- Keep extension-owned onboarding/config/provider coverage under the owning bundled plugin package when feasible.
@@ -87,6 +87,8 @@
- `src/plugin-sdk/AGENTS.md` expands public SDK contract rules.
- `src/plugins/AGENTS.md` expands plugin loading, registry, and manifest rules.
- `src/gateway/protocol/AGENTS.md` expands typed Gateway protocol rules.
- `src/gateway/AGENTS.md` expands Gateway server hot-path and plugin artifact rules.
- `src/agents/AGENTS.md` expands agent test/import performance rules.
- `test/helpers/AGENTS.md` and `test/helpers/channels/AGENTS.md` expand shared test helper boundary rules.
- Plugin architecture direction:
- Keep a manifest-first control plane: discovery, validation, enablement, setup hints, and activation planning should stay metadata-driven by default.
@@ -117,8 +119,8 @@
- Runtime baseline: Node **22+** (keep Node + Bun paths working).
- Install deps: `pnpm install`
- If deps are missing (for example `node_modules` missing, `vitest not found`, or `command not found`), run the repos package-manager install command (prefer lockfile/README-defined PM), then rerun the exact requested command once. Apply this to test/build/lint/typecheck/dev commands; if retry still fails, report the command and first actionable error.
- Pre-commit hooks: `prek install`. The hook runs the repo verification flow, including `pnpm check`.
- `FAST_COMMIT=1` skips the repo-wide `pnpm format` and `pnpm check` inside the pre-commit hook only. Use it when you intentionally want a faster commit path and are running equivalent targeted verification manually. It does not change CI and does not change what `pnpm check` itself does.
- Pre-commit hooks are installed by the package `prepare` script (`git config core.hooksPath git-hooks`). The hook formats/lints staged source files and runs `pnpm check` unless the staged change is docs-only or `FAST_COMMIT=1` is set.
- `FAST_COMMIT=1` skips the repo-wide `pnpm check` inside the pre-commit hook only. The hook still runs targeted formatting/linting for staged files and restages formatter changes. Use it when you intentionally want a faster commit path and are running equivalent targeted verification manually. It does not change CI and does not change what `pnpm check` itself does.
- Also supported: `bun install` (keep `pnpm-lock.yaml` + Bun patching in sync when touching deps/patches).
- Prefer Bun for TypeScript execution (scripts, dev, tests): `bun <file.ts>` / `bunx <tool>`.
- Run CLI in dev: `pnpm openclaw ...` (bun) or `pnpm dev`.
@@ -128,8 +130,8 @@
- TypeScript checks: `pnpm tsgo`
- Lint/format: `pnpm check`
- Local agent/dev shells default to host-aware `OPENCLAW_LOCAL_CHECK=1` behavior for `pnpm tsgo` and `pnpm lint`; set `OPENCLAW_LOCAL_CHECK_MODE=throttled` to force the lower-memory profile, `OPENCLAW_LOCAL_CHECK_MODE=full` to keep lock-only behavior, or `OPENCLAW_LOCAL_CHECK=0` in CI/shared runs.
- Format check: `pnpm format` (oxfmt --check)
- Format fix: `pnpm format:fix` (oxfmt --write)
- Format check: `pnpm format:check` (oxfmt --check)
- Format fix: `pnpm format` or `pnpm format:fix` (oxfmt --write)
- Terminology:
- "gate" means a verification command or command set that must be green for the decision you are making.
- A local dev gate is the fast default loop, usually `pnpm check` plus any scoped test you actually need.
@@ -137,8 +139,8 @@
- A CI gate is whatever the relevant workflow enforces for that lane (for example `check`, `check-additional`, `build-smoke`, or release validation).
- Local dev gate: prefer `pnpm check` for the normal edit loop. It keeps the repo-architecture policy guards out of the default local loop.
- CI architecture gate: `check-additional` enforces architecture and boundary policy guards that are intentionally kept out of the default local loop.
- Formatting gate: the pre-commit hook runs `pnpm format` before `pnpm check`. If you want a formatting-only preflight locally, run `pnpm format` explicitly.
- If you need a fast commit loop, `FAST_COMMIT=1 git commit ...` skips the hooks repo-wide `pnpm format` and `pnpm check`; use that only when you are deliberately covering the touched surface some other way.
- Formatting gate: the pre-commit hook runs targeted formatting on staged source files before `pnpm check`. If you want a repo-wide formatting-only preflight locally, run `pnpm format:check` explicitly.
- If you need a fast commit loop, `FAST_COMMIT=1 git commit ...` skips the hooks repo-wide `pnpm check`; targeted formatting/linting still runs, so use that only when you are deliberately covering the touched surface some other way.
- Tests: `pnpm test` (vitest); coverage: `pnpm test:coverage`
- Generated baseline drift detection uses SHA-256 hash files under `docs/.generated/` (`.sha256` files tracked in git; full JSON baselines are gitignored, generated locally for inspection).
- Config schema drift uses `pnpm config:docs:gen` / `pnpm config:docs:check`.
@@ -215,6 +217,8 @@
- Test performance guardrail: when production code already accepts `deps`, callbacks, or runtime injection, use that seam in tests before adding module-level mocks.
- Test performance guardrail: prefer narrow public SDK subpaths such as `models-provider-runtime`, `skill-commands-runtime`, and `reply-dispatch-runtime` over older broad helper barrels when both expose the needed helper.
- Test performance guardrail: treat import-dominated test time as a boundary bug. Refactor the import surface before adding more cases to the slow file.
- Test performance guardrail: when replacing a slow integration test with helper-level coverage, extract the exact production composition into a named helper and test that helper. Do not trade coverage shape for speed without preserving the behavior proof somewhere cheaper.
- Test performance guardrail: for plugin-owned static descriptors used by core tests or cold paths, prefer lightweight public artifacts with full-runtime fallback over loading broad bundled plugin barrels.
- Agents MUST NOT modify baseline, inventory, ignore, snapshot, or expected-failure files to silence failing checks without explicit approval in this chat.
- For targeted/local debugging, use the native root-project entrypoint: `pnpm test <path-or-filter> [vitest args...]` (for example `pnpm test src/commands/onboard-search.test.ts -t "shows registered plugin providers"`); do not default to raw `pnpm vitest run ...` because it bypasses the repo's default config/profile/pool routing.
- Do not set test workers above 16; tried already.
@@ -250,8 +254,8 @@
## Security & Configuration Tips
- Web provider stores creds at `~/.openclaw/credentials/`; rerun `openclaw login` if logged out.
- Pi sessions live under `~/.openclaw/sessions/` by default; the base directory is not configurable.
- Channel/provider state lives under `~/.openclaw/credentials/`; rerun `openclaw channels login` if logged out. Model auth profiles live under `~/.openclaw/agents/<agentId>/agent/auth-profiles.json`; legacy OAuth import still reads `~/.openclaw/credentials/oauth.json`.
- Pi sessions live under `~/.openclaw/agents/<agentId>/sessions/` by default; `session.store` can override the session store path.
- Environment variables: see `~/.profile`.
- Never commit or publish real phone numbers, videos, or live configuration values. Use obviously fake placeholders in docs, tests, and examples.
- Release flow: use the private [maintainer release docs](https://github.com/openclaw/maintainers/blob/main/release/README.md) for the actual runbook, `docs/reference/RELEASING.md` for the public release policy, and `$openclaw-release-maintainer` for the maintainership workflow.
@@ -268,13 +272,13 @@
- Signal: "update fly" => `fly ssh console -a flawd-bot -C "bash -lc 'cd /data/clawd/openclaw && git pull --rebase origin main'"` then `fly machines restart e825232f34d058 -a flawd-bot`.
- CLI progress: use `src/cli/progress.ts` (`osc-progress` + `@clack/prompts` spinner); dont hand-roll spinners/bars.
- Status output: keep tables + ANSI-safe wrapping (`src/terminal/table.ts`); `status --all` = read-only/pasteable, `status --deep` = probes.
- Gateway currently runs only as the menubar app; there is no separate LaunchAgent/helper label installed. Restart via the OpenClaw Mac app or `scripts/restart-mac.sh`; to verify/kill use `launchctl print gui/$UID | grep openclaw` rather than assuming a fixed label. **When debugging on macOS, start/stop the gateway via the app, not ad-hoc tmux sessions; kill any temporary tunnels before handoff.**
- Gateway may run as an app-managed launchd job. Restart the gateway via the app or `openclaw gateway restart`; inspect with `openclaw gateway status --deep` or, for the default profile, `launchctl print gui/$UID/ai.openclaw.gateway`. Use `scripts/restart-mac.sh` when you need to rebuild/relaunch the local macOS app itself. The app LaunchAgent uses `ai.openclaw.mac`. **When debugging on macOS, start/stop the gateway via the app or gateway CLI, not ad-hoc tmux sessions; kill any temporary tunnels before handoff.**
- macOS logs: use `./scripts/clawlog.sh` to query unified logs for the OpenClaw subsystem; it supports follow/tail/category filters and expects passwordless sudo for `/usr/bin/log`.
- If shared guardrails are available locally, review them; otherwise follow this repo's guidance.
- SwiftUI state management (iOS/macOS): prefer the `Observation` framework (`@Observable`, `@Bindable`) over `ObservableObject`/`@StateObject`; dont introduce new `ObservableObject` unless required for compatibility, and migrate existing usages when touching related code.
- Connection providers: when adding a new connection, update every UI surface and docs (macOS app, web UI, mobile if applicable, onboarding/overview docs) and add matching status + configuration forms so provider lists and settings stay in sync.
- Version locations: `package.json` (CLI), `apps/android/app/build.gradle.kts` (versionName/versionCode), `apps/ios/Sources/Info.plist` + `apps/ios/Tests/Info.plist` (CFBundleShortVersionString/CFBundleVersion), `apps/macos/Sources/OpenClaw/Resources/Info.plist` (CFBundleShortVersionString/CFBundleVersion), `docs/install/updating.md` (pinned npm version), and Peekaboo Xcode projects/Info.plists (MARKETING_VERSION/CURRENT_PROJECT_VERSION).
- "Bump version everywhere" means all version locations above **except** `appcast.xml` (only touch appcast when cutting a new macOS Sparkle release).
- Version locations: `package.json` (CLI), `apps/android/app/build.gradle.kts` (versionName/versionCode), `apps/ios/version.json` (source for generated iOS config and Fastlane metadata), `apps/macos/Sources/OpenClaw/Resources/Info.plist` (CFBundleShortVersionString/CFBundleVersion), and `docs/install/updating.md` (pinned npm version).
- "Bump version everywhere" means all version locations above, then run `pnpm ios:version:sync` for iOS generated outputs. Only touch appcast metadata when cutting a new macOS Sparkle release.
- **Restart apps:** “restart iOS/Android apps” means rebuild (recompile/install) and relaunch, not just kill/launch.
- **Device checks:** before testing, verify connected real devices (iOS/Android) before reaching for simulators/emulators.
- Mobile pairing: `ws://` (cleartext) is allowed for private LAN addresses (RFC 1918, link-local, mDNS `.local`) and loopback. Private LAN hosts typically lack PKI-backed identity, so requiring TLS there adds complexity without meaningful security gain. `wss://` is required for Tailscale and public endpoints.

View File

@@ -4,9 +4,43 @@ Docs: https://docs.openclaw.ai
## Unreleased
### Changes
- macOS/gateway: add `screen.snapshot` support for macOS app nodes, including runtime plumbing, default macOS allowlisting, and docs for monitor preview flows. (#67954) Thanks @BunsDev.
### Fixes
- Agents/bootstrap: resolve bootstrap from workspace truth instead of stale session transcript markers, keep embedded bootstrap instructions on a hidden user-context prelude, suppress normal `/new` and `/reset` greetings while `BOOTSTRAP.md` is still pending, and make the embedded runner read the bootstrap ritual before replying normally.
- Onboarding/non-interactive: preserve existing gateway auth tokens during re-onboard so active local gateway clients are not disconnected by an implicit token rotation. (#67821) Thanks @BKF-Gitty.
- Gateway/hello-ok: always report negotiated auth metadata for successful shared-auth handshakes, including control-ui bypass coverage when no device token is issued. (#67810) Thanks @BunsDev.
- OpenAI Codex/Responses: unify native Responses API capability detection so Codex OAuth requests emit the required `store: false` field on the native Responses path. (#67918) Thanks @obviyus.
- WhatsApp/setup: guard personal-phone and allowlist prompt values so setup fails with clear validation errors instead of crashing on undefined prompt text. (#67895) Thanks @lawrence3699.
- Models/config: preserve an existing `models.json` provider `baseUrl` during merge-mode regeneration so custom endpoints do not get reset on restart. (#67893) Thanks @lawrence3699.
- Plugins/discovery: reuse bundled and global plugin discovery results across workspace cache misses so Windows multi-workspace startup stops redoing the shared synchronous scan. (#67940) Thanks @obviyus.
- Plugins/webhooks: enforce synchronous plugin registration with full rollback of failed plugin side effects, and cache SecretRef-backed webhook auth per route so plugin startup and inbound webhook auth stay deterministic. (#67941) Thanks @obviyus.
- Telegram/ACP bindings: drop persisted DM bindings that still point at missing or failed ACP sessions on restart, while preserving plugin-owned bindings and uncertain store reads. (#67822) Thanks @chinar-amrutkar.
- Telegram/streaming: keep a transient preview on the same Telegram message when auto-compaction retries an in-flight answer, so streamed replies no longer appear duplicated after compaction. (#66939) Thanks @rubencu.
- Memory/sqlite-vec: emit the degraded sqlite-vec warning once per degraded episode instead of repeating it for every file write, while preserving the latch across safe-reindex rollback and resetting it when vector state is genuinely rebuilt. (#67898) Thanks @rubencu.
- Reply/block streaming: preserve post-stream incomplete-turn error payloads after block streaming already emitted content, so users get the warning instead of silence. (#67991) Thanks @obviyus.
- Telegram/streaming: clear the compaction replay guard after visible non-final boundaries so a post-tool assistant reply rotates to a fresh preview instead of editing the pre-compaction message. (#67993) Thanks @obviyus.
- Matrix: fix `sessions_spawn --thread` subagent session spawning — thread binding creation, cleanup on session end, and completion-message delivery target resolution now work end-to-end. (#67643) Thanks @eejohnso-ops and @gumadeiras.
- macOS/webchat: enable Undo and Redo in the composer text input by turning on the native `NSTextView` undo manager. (#34962) Thanks @tylerbittner.
- macOS/remote SSH: require an already-trusted host key on the macOS remote command, gateway probe, port tunnel, and pairing probe paths by switching `StrictHostKeyChecking=accept-new` to `StrictHostKeyChecking=yes` and centralizing the shared SSH option fragments in `CommandResolver`, so first-time macOS remote connections no longer silently accept an unknown host key and must be trusted ahead of time via `~/.ssh/known_hosts`. (#68199)
- CLI/configure: show the channel picker before probing statuses and let remove mode delete configured channel blocks directly from config. (#68007) Thanks @gumadeiras.
- OpenAI Codex/OAuth: keep OpenClaw as the canonical owner for imported Codex CLI OAuth sessions, stop writing refreshed credentials back into `.codex`, and prefer fresher OpenClaw credentials over stale imported CLI state so refresh recovery stays stable. Thanks @vincentkoc.
- OpenAI Codex/OAuth: treat the OpenAI TLS prerequisites probe as advisory instead of a hard blocker, so Codex sign-in can still proceed when the speculative Node/OpenSSL precheck fails but the real OAuth flow still works. Thanks @vincentkoc.
- Models status/OAuth health: align OAuth health reporting with the same effective credential view runtime uses, so expired refreshable sessions stop showing healthy by default and fresher imported Codex CLI credentials surface correctly in `models status`, doctor, and gateway auth status. Thanks @vincentkoc.
- OpenAI Codex/OAuth: keep external CLI OAuth imports runtime-only by overlaying fresher Codex CLI credentials without mutating `auth-profiles.json`, so `.codex` stays a bootstrap/runtime input instead of becoming durable OpenClaw state. Thanks @vincentkoc.
- OpenAI Codex/OAuth: drop legacy CLI-manager routing from the remaining bootstrap path so Codex and MiniMax CLI imports are matched by their canonical OpenClaw profile ids instead of stale `managedBy` metadata. Thanks @vincentkoc.
- OpenAI Codex/OAuth: only bootstrap from external CLI OAuth when the local OpenClaw profile is missing or unusable, so healthy local sessions are no longer overridden by fresher `.codex` tokens. Thanks @vincentkoc.
- OpenAI Codex/OAuth: rename the external CLI bootstrap helper, reuse the same usable-oauth check across runtime fallback paths, and add debug logs plus health coverage so bootstrap decisions stay legible. Thanks @vincentkoc.
- Twitch/setup: load Twitch through the bundled setup-entry discovery path and keep setup/status account detection aligned with runtime config. (#68008) Thanks @gumadeiras.
- Feishu/card actions: resolve card-action chat type from the Feishu chat API when stored context is missing, preferring `chat_mode` over `chat_type`, so DM-originated card actions no longer bypass `dmPolicy` by falling through to the group handling path. (#68201)
- Cron/isolated-agent: preserve `trusted: false` on isolated cron awareness events mirrored into the main session, and forward the optional `trusted` flag through the gateway cron wrapper so explicit trust downgrades survive session-key scoping. (#68210)
- Agents/fallback: recognize bare leading ZenMux `402 ...` quota-refresh errors without misclassifying plain numeric `402 ...` text, and keep the embedded fallback regression coverage stable. (#47579) Thanks @bwjoke.
- Failover/google: only treat `INTERNAL` status payloads as retryable timeouts when they also carry a `500` code, so malformed non-500 payloads do not enter the retry path. (#68238) Thanks @altaywtf and @Openbling.
- Agents/tools: filter bundled MCP/LSP tools through the final owner-only and tool-policy pipeline after merging them into the effective tool list, so existing allowlists, deny rules, sandbox policy, subagent policy, and owner-only restrictions apply to bundled tools the same way they apply to core tools. (#68195)
- Gateway/assistant media: require `operator.read` scope for assistant-media file and metadata requests on identity-bearing HTTP auth paths so callers without a read scope can no longer access assistant media. (#68175) Thanks @eleqtrizit.
## 2026.4.15
@@ -113,6 +147,7 @@ Docs: https://docs.openclaw.ai
- Webchat/security: reject remote-host `file://` URLs in the media embedding path. (#67293) Thanks @pgondhi987.
- Dreaming/memory-core: use the ingestion day, not the source file day, for daily recall dedupe so repeat sweeps of the same daily note can increment `dailyCount` across days instead of stalling at `1`. (#67091) Thanks @Bartok9.
- Node-host/tools.exec: let approval binding distinguish known native binaries from mutable shell payload files, while still fail-closing unknown or racy file probes so absolute-path node-host commands like `/usr/bin/whoami` no longer get rejected as unsafe interpreter/runtime commands. (#66731) Thanks @tmimmanuel.
- Codex/gateway: fix gateway crash when the codex-acp subprocess terminates abruptly; an unhandled EPIPE on the child stdin stream now routes through graceful client shutdown, rejecting pending requests instead of propagating as an uncaught exception that crashes the entire gateway daemon and all connected channels. Fixes #67886. (#67947) thanks @openperf
## 2026.4.14

View File

@@ -3,6 +3,12 @@ import Foundation
enum CommandResolver {
private static let projectRootDefaultsKey = "openclaw.gatewayProjectRootPath"
private static let helperName = "openclaw"
static let strictHostKeyCheckingSSHOptions = [
"-o", "StrictHostKeyChecking=yes",
]
static let updateHostKeysSSHOptions = [
"-o", "UpdateHostKeys=yes",
]
static func gatewayEntrypoint(in root: URL) -> String? {
let distEntry = root.appendingPathComponent("dist/index.js").path
@@ -397,9 +403,7 @@ enum CommandResolver {
"""
let options: [String] = [
"-o", "BatchMode=yes",
"-o", "StrictHostKeyChecking=accept-new",
"-o", "UpdateHostKeys=yes",
]
] + self.strictHostKeyCheckingSSHOptions + self.updateHostKeysSSHOptions
let args = self.sshArguments(
target: parsed,
identity: settings.identity,

View File

@@ -146,6 +146,7 @@ final class MacNodeModeCoordinator {
OpenClawCanvasA2UICommand.push.rawValue,
OpenClawCanvasA2UICommand.pushJSONL.rawValue,
OpenClawCanvasA2UICommand.reset.rawValue,
MacNodeScreenCommand.snapshot.rawValue,
MacNodeScreenCommand.record.rawValue,
OpenClawSystemCommand.notify.rawValue,
OpenClawSystemCommand.which.rawValue,

View File

@@ -63,6 +63,8 @@ actor MacNodeRuntime {
return try await self.handleCameraInvoke(req)
case OpenClawLocationCommand.get.rawValue:
return try await self.handleLocationInvoke(req)
case MacNodeScreenCommand.snapshot.rawValue:
return try await self.handleScreenSnapshotInvoke(req)
case MacNodeScreenCommand.record.rawValue:
return try await self.handleScreenRecordInvoke(req)
case OpenClawSystemCommand.run.rawValue:
@@ -352,6 +354,34 @@ actor MacNodeRuntime {
return BridgeInvokeResponse(id: req.id, ok: true, payloadJSON: payload)
}
private func handleScreenSnapshotInvoke(_ req: BridgeInvokeRequest) async throws -> BridgeInvokeResponse {
let params = (try? Self.decodeParams(MacNodeScreenSnapshotParams.self, from: req.paramsJSON)) ??
MacNodeScreenSnapshotParams()
let services = await self.mainActorServices()
let capturedAtMs = Int64(Date().timeIntervalSince1970 * 1000)
let res = try await services.snapshotScreen(
screenIndex: params.screenIndex,
maxWidth: params.maxWidth,
quality: params.quality,
format: params.format)
struct ScreenSnapshotPayload: Encodable {
var format: String
var base64: String
var width: Int
var height: Int
var screenIndex: Int?
var capturedAtMs: Int64
}
let payload = try Self.encodePayload(ScreenSnapshotPayload(
format: res.format.rawValue,
base64: res.data.base64EncodedString(),
width: res.width,
height: res.height,
screenIndex: params.screenIndex,
capturedAtMs: capturedAtMs))
return BridgeInvokeResponse(id: req.id, ok: true, payloadJSON: payload)
}
private func mainActorServices() async -> any MacNodeRuntimeMainActorServices {
if let cachedMainActorServices { return cachedMainActorServices }
let services = await self.makeMainActorServices()

View File

@@ -4,6 +4,13 @@ import OpenClawKit
@MainActor
protocol MacNodeRuntimeMainActorServices: Sendable {
func snapshotScreen(
screenIndex: Int?,
maxWidth: Int?,
quality: Double?,
format: OpenClawScreenSnapshotFormat?) async throws
-> (data: Data, format: OpenClawScreenSnapshotFormat, width: Int, height: Int)
func recordScreen(
screenIndex: Int?,
durationMs: Int?,
@@ -21,9 +28,24 @@ protocol MacNodeRuntimeMainActorServices: Sendable {
@MainActor
final class LiveMacNodeRuntimeMainActorServices: MacNodeRuntimeMainActorServices, @unchecked Sendable {
private let screenSnapshotter = ScreenSnapshotService()
private let screenRecorder = ScreenRecordService()
private let locationService = MacNodeLocationService()
func snapshotScreen(
screenIndex: Int?,
maxWidth: Int?,
quality: Double?,
format: OpenClawScreenSnapshotFormat?) async throws
-> (data: Data, format: OpenClawScreenSnapshotFormat, width: Int, height: Int)
{
try await self.screenSnapshotter.snapshot(
screenIndex: screenIndex,
maxWidth: maxWidth,
quality: quality,
format: format)
}
func recordScreen(
screenIndex: Int?,
durationMs: Int?,

View File

@@ -1,9 +1,18 @@
import Foundation
import OpenClawKit
enum MacNodeScreenCommand: String, Codable {
case snapshot = "screen.snapshot"
case record = "screen.record"
}
struct MacNodeScreenSnapshotParams: Codable, Equatable {
var screenIndex: Int?
var maxWidth: Int?
var quality: Double?
var format: OpenClawScreenSnapshotFormat?
}
struct MacNodeScreenRecordParams: Codable, Equatable {
var screenIndex: Int?
var durationMs: Int?

View File

@@ -483,8 +483,7 @@ final class NodePairingApprovalPrompter {
"-o", "ConnectTimeout=5",
"-o", "NumberOfPasswordPrompts=0",
"-o", "PreferredAuthentications=publickey",
"-o", "StrictHostKeyChecking=accept-new",
]
] + CommandResolver.strictHostKeyCheckingSSHOptions
guard let target = CommandResolver.makeSSHTarget(user: user, host: host, port: port) else {
return false
}

View File

@@ -200,9 +200,7 @@ enum RemoteGatewayProbe {
let options = [
"-o", "BatchMode=yes",
"-o", "ConnectTimeout=5",
"-o", "StrictHostKeyChecking=accept-new",
"-o", "UpdateHostKeys=yes",
]
] + CommandResolver.strictHostKeyCheckingSSHOptions + CommandResolver.updateHostKeysSSHOptions
let args = CommandResolver.sshArguments(
target: parsed,
identity: identity,

View File

@@ -73,14 +73,12 @@ final class RemotePortTunnel {
let options: [String] = [
"-o", "BatchMode=yes",
"-o", "ExitOnForwardFailure=yes",
"-o", "StrictHostKeyChecking=accept-new",
"-o", "UpdateHostKeys=yes",
"-o", "ServerAliveInterval=15",
"-o", "ServerAliveCountMax=3",
"-o", "TCPKeepAlive=yes",
"-N",
"-L", "\(localPort):127.0.0.1:\(resolvedRemotePort)",
]
] + CommandResolver.strictHostKeyCheckingSSHOptions + CommandResolver.updateHostKeysSSHOptions
let identity = settings.identity.trimmingCharacters(in: .whitespacesAndNewlines)
let args = CommandResolver.sshArguments(
target: parsed,

View File

@@ -0,0 +1,109 @@
import AppKit
import Foundation
import OpenClawKit
@preconcurrency import ScreenCaptureKit
@MainActor
final class ScreenSnapshotService {
enum ScreenSnapshotError: LocalizedError {
case noDisplays
case invalidScreenIndex(Int)
case captureFailed(String)
case encodeFailed(String)
var errorDescription: String? {
switch self {
case .noDisplays:
"No displays available for screen snapshot"
case let .invalidScreenIndex(idx):
"Invalid screen index \(idx)"
case let .captureFailed(message):
message
case let .encodeFailed(message):
message
}
}
}
func snapshot(
screenIndex: Int?,
maxWidth: Int?,
quality: Double?,
format: OpenClawScreenSnapshotFormat?) async throws
-> (data: Data, format: OpenClawScreenSnapshotFormat, width: Int, height: Int)
{
let format = format ?? .jpeg
let normalized = Self.normalize(maxWidth: maxWidth, quality: quality, format: format)
let content = try await SCShareableContent.current
let displays = content.displays.sorted { $0.displayID < $1.displayID }
guard !displays.isEmpty else {
throw ScreenSnapshotError.noDisplays
}
let idx = screenIndex ?? 0
guard idx >= 0, idx < displays.count else {
throw ScreenSnapshotError.invalidScreenIndex(idx)
}
let display = displays[idx]
let filter = SCContentFilter(display: display, excludingWindows: [])
let config = SCStreamConfiguration()
let targetSize = Self.targetSize(
width: display.width,
height: display.height,
maxWidth: normalized.maxWidth)
config.width = targetSize.width
config.height = targetSize.height
config.showsCursor = true
let cgImage: CGImage
do {
cgImage = try await SCScreenshotManager.captureImage(
contentFilter: filter,
configuration: config)
} catch {
throw ScreenSnapshotError.captureFailed(error.localizedDescription)
}
let bitmap = NSBitmapImageRep(cgImage: cgImage)
let data: Data
switch format {
case .png:
guard let encoded = bitmap.representation(using: .png, properties: [:]) else {
throw ScreenSnapshotError.encodeFailed("png encode failed")
}
data = encoded
case .jpeg:
guard let encoded = bitmap.representation(
using: .jpeg,
properties: [.compressionFactor: normalized.quality])
else {
throw ScreenSnapshotError.encodeFailed("jpeg encode failed")
}
data = encoded
}
return (data: data, format: format, width: cgImage.width, height: cgImage.height)
}
private static func normalize(
maxWidth: Int?,
quality: Double?,
format: OpenClawScreenSnapshotFormat)
-> (maxWidth: Int, quality: Double)
{
let resolvedMaxWidth = maxWidth.flatMap { $0 > 0 ? $0 : nil } ?? (format == .png ? 900 : 1600)
let resolvedQuality = min(1.0, max(0.05, quality ?? 0.72))
return (maxWidth: resolvedMaxWidth, quality: resolvedQuality)
}
private static func targetSize(width: Int, height: Int, maxWidth: Int) -> (width: Int, height: Int) {
guard width > 0, height > 0, width > maxWidth else {
return (width: width, height: height)
}
let scale = Double(maxWidth) / Double(width)
let targetHeight = max(1, Int((Double(height) * scale).rounded()))
return (width: maxWidth, height: targetHeight)
}
}

View File

@@ -164,6 +164,9 @@ import Testing
} else {
#expect(Bool(false))
}
#expect(cmd.contains("StrictHostKeyChecking=yes"))
#expect(!cmd.contains("StrictHostKeyChecking=accept-new"))
#expect(cmd.contains("UpdateHostKeys=yes"))
#expect(cmd.contains("-i"))
#expect(cmd.contains("/tmp/id_ed25519"))
if let script = cmd.last {

View File

@@ -78,6 +78,19 @@ struct MacNodeRuntimeTests {
@Test func `handle invoke screen record uses injected services`() async throws {
@MainActor
final class FakeMainActorServices: MacNodeRuntimeMainActorServices, @unchecked Sendable {
func snapshotScreen(
screenIndex: Int?,
maxWidth: Int?,
quality: Double?,
format: OpenClawScreenSnapshotFormat?) async throws
-> (data: Data, format: OpenClawScreenSnapshotFormat, width: Int, height: Int)
{
_ = screenIndex
_ = maxWidth
_ = quality
return (Data("snapshot".utf8), format ?? .jpeg, 640, 360)
}
func recordScreen(
screenIndex: Int?,
durationMs: Int?,
@@ -127,6 +140,94 @@ struct MacNodeRuntimeTests {
#expect(!payload.base64.isEmpty)
}
@Test func `handle invoke screen snapshot uses injected services`() async throws {
@MainActor
final class FakeMainActorServices: MacNodeRuntimeMainActorServices, @unchecked Sendable {
var snapshotCalledAtMs: Int64?
func snapshotScreen(
screenIndex: Int?,
maxWidth: Int?,
quality: Double?,
format: OpenClawScreenSnapshotFormat?) async throws
-> (data: Data, format: OpenClawScreenSnapshotFormat, width: Int, height: Int)
{
self.snapshotCalledAtMs = Int64(Date().timeIntervalSince1970 * 1000)
#expect(screenIndex == 0)
#expect(maxWidth == 800)
#expect(quality == 0.5)
return (Data("ok".utf8), format ?? .jpeg, 800, 450)
}
func recordScreen(
screenIndex: Int?,
durationMs: Int?,
fps: Double?,
includeAudio: Bool?,
outPath: String?) async throws -> (path: String, hasAudio: Bool)
{
let url = FileManager().temporaryDirectory
.appendingPathComponent("openclaw-test-screen-record-\(UUID().uuidString).mp4")
try Data("ok".utf8).write(to: url)
return (path: url.path, hasAudio: false)
}
func locationAuthorizationStatus() -> CLAuthorizationStatus {
.authorizedAlways
}
func locationAccuracyAuthorization() -> CLAccuracyAuthorization {
.fullAccuracy
}
func currentLocation(
desiredAccuracy: OpenClawLocationAccuracy,
maxAgeMs: Int?,
timeoutMs: Int?) async throws -> CLLocation
{
_ = desiredAccuracy
_ = maxAgeMs
_ = timeoutMs
return CLLocation(latitude: 0, longitude: 0)
}
}
let services = await MainActor.run { FakeMainActorServices() }
let runtime = MacNodeRuntime(makeMainActorServices: { services })
let params = MacNodeScreenSnapshotParams(
screenIndex: 0,
maxWidth: 800,
quality: 0.5,
format: .jpeg)
let json = try String(data: JSONEncoder().encode(params), encoding: .utf8)
let response = await runtime.handleInvoke(
BridgeInvokeRequest(
id: "req-screen-snapshot",
command: MacNodeScreenCommand.snapshot.rawValue,
paramsJSON: json))
#expect(response.ok == true)
let payloadJSON = try #require(response.payloadJSON)
struct Payload: Decodable {
var format: String
var base64: String
var width: Int
var height: Int
var capturedAtMs: Int64
}
let payload = try JSONDecoder().decode(Payload.self, from: Data(payloadJSON.utf8))
#expect(payload.format == "jpeg")
#expect(payload.base64 == Data("ok".utf8).base64EncodedString())
#expect(payload.width == 800)
#expect(payload.height == 450)
#expect(payload.capturedAtMs > 0)
let snapshotCalledAtMs = await MainActor.run { services.snapshotCalledAtMs }
#expect(snapshotCalledAtMs != nil)
#expect(payload.capturedAtMs <= snapshotCalledAtMs!)
}
@Test func `handle invoke browser proxy uses injected request`() async {
let runtime = MacNodeRuntime(browserProxyRequest: { paramsJSON in
#expect(paramsJSON?.contains("/tabs") == true)

View File

@@ -444,34 +444,18 @@ private struct ChatComposerTextView: NSViewRepresentable {
func makeCoordinator() -> Coordinator { Coordinator(self) }
func makeNSView(context: Context) -> NSScrollView {
let textView = ChatComposerNSTextView()
textView.delegate = context.coordinator
textView.drawsBackground = false
textView.isRichText = false
textView.isAutomaticQuoteSubstitutionEnabled = false
textView.isAutomaticTextReplacementEnabled = false
textView.isAutomaticDashSubstitutionEnabled = false
textView.isAutomaticSpellingCorrectionEnabled = false
textView.font = .systemFont(ofSize: 14, weight: .regular)
textView.textContainer?.lineBreakMode = .byWordWrapping
textView.textContainer?.lineFragmentPadding = 0
textView.textContainerInset = NSSize(width: 2, height: 4)
textView.focusRingType = .none
let textView = ChatComposerTextViewFactory.makeConfiguredTextView()
guard let composerTextView = textView as? ChatComposerNSTextView else {
preconditionFailure("ChatComposerTextViewFactory must return ChatComposerNSTextView")
}
composerTextView.delegate = context.coordinator
textView.minSize = .zero
textView.maxSize = NSSize(width: CGFloat.greatestFiniteMagnitude, height: CGFloat.greatestFiniteMagnitude)
textView.isHorizontallyResizable = false
textView.isVerticallyResizable = true
textView.autoresizingMask = [.width]
textView.textContainer?.containerSize = NSSize(width: 0, height: CGFloat.greatestFiniteMagnitude)
textView.textContainer?.widthTracksTextView = true
textView.string = self.text
textView.onSend = { [weak textView] in
textView?.window?.makeFirstResponder(nil)
composerTextView.string = self.text
composerTextView.onSend = { [weak composerTextView] in
composerTextView?.window?.makeFirstResponder(nil)
self.onSend()
}
textView.onPasteImageAttachment = self.onPasteImageAttachment
composerTextView.onPasteImageAttachment = self.onPasteImageAttachment
let scroll = NSScrollView()
scroll.drawsBackground = false
@@ -522,6 +506,34 @@ private struct ChatComposerTextView: NSViewRepresentable {
}
}
enum ChatComposerTextViewFactory {
// Internal for @testable import coverage of composer text view defaults.
@MainActor
static func makeConfiguredTextView() -> NSTextView {
let textView = ChatComposerNSTextView()
textView.drawsBackground = false
textView.isRichText = false
textView.isAutomaticQuoteSubstitutionEnabled = false
textView.isAutomaticTextReplacementEnabled = false
textView.isAutomaticDashSubstitutionEnabled = false
textView.isAutomaticSpellingCorrectionEnabled = false
textView.font = .systemFont(ofSize: 14, weight: .regular)
textView.textContainer?.lineBreakMode = .byWordWrapping
textView.textContainer?.lineFragmentPadding = 0
textView.textContainerInset = NSSize(width: 2, height: 4)
textView.focusRingType = .none
textView.allowsUndo = true
textView.minSize = .zero
textView.maxSize = NSSize(width: CGFloat.greatestFiniteMagnitude, height: CGFloat.greatestFiniteMagnitude)
textView.isHorizontallyResizable = false
textView.isVerticallyResizable = true
textView.autoresizingMask = [.width]
textView.textContainer?.containerSize = NSSize(width: 0, height: CGFloat.greatestFiniteMagnitude)
textView.textContainer?.widthTracksTextView = true
return textView
}
}
private final class ChatComposerNSTextView: NSTextView {
var onSend: (() -> Void)?
var onPasteImageAttachment: ((_ data: Data, _ fileName: String, _ mimeType: String) -> Void)?

View File

@@ -1,9 +1,34 @@
import Foundation
public enum OpenClawScreenCommand: String, Codable, Sendable {
case snapshot = "screen.snapshot"
case record = "screen.record"
}
public enum OpenClawScreenSnapshotFormat: String, Codable, Sendable {
case jpeg
case png
}
public struct OpenClawScreenSnapshotParams: Codable, Sendable, Equatable {
public var screenIndex: Int?
public var maxWidth: Int?
public var quality: Double?
public var format: OpenClawScreenSnapshotFormat?
public init(
screenIndex: Int? = nil,
maxWidth: Int? = nil,
quality: Double? = nil,
format: OpenClawScreenSnapshotFormat? = nil)
{
self.screenIndex = screenIndex
self.maxWidth = maxWidth
self.quality = quality
self.format = format
}
}
public struct OpenClawScreenRecordParams: Codable, Sendable, Equatable {
public var screenIndex: Int?
public var durationMs: Int?

View File

@@ -0,0 +1,15 @@
#if os(macOS)
import AppKit
import Testing
@testable import OpenClawChatUI
@Suite
@MainActor
struct ChatComposerTextViewTests {
@Test func configuredComposerTextViewEnablesUndo() {
let textView = ChatComposerTextViewFactory.makeConfiguredTextView()
#expect(textView.allowsUndo)
}
}
#endif

View File

@@ -1,2 +1,2 @@
9683f324fae8f455f2b64d7e152a77009941e4c7558521bca2510d8bcf573af9 plugin-sdk-api-baseline.json
097bf226e4e857e9296d0851852a2963c6263d176c4c470452d9a8efd36988e5 plugin-sdk-api-baseline.jsonl
052943a9f1eb82a49452b6715f4c08faeb650d16a36c150a3c726ff392ecad0d plugin-sdk-api-baseline.json
a5077395f009f5064331dc1c38bb2d6d2864299d3c1fbd9e40956c1700fa253c plugin-sdk-api-baseline.jsonl

View File

@@ -120,7 +120,7 @@ can write back through the mounted workspace.
Seed assets live in `qa/`:
- `qa/scenarios/index.md`
- `qa/scenarios/*.md`
- `qa/scenarios/<theme>/*.md`
These are intentionally in git so the QA plan is visible to both humans and the
agent.
@@ -129,6 +129,7 @@ agent.
the source of truth for one test run and should define:
- scenario metadata
- optional category, capability, lane, and risk metadata
- docs and code refs
- optional plugin requirements
- optional gateway config patch
@@ -139,6 +140,10 @@ and cross-cutting. For example, markdown scenarios can combine transport-side
helpers with browser-side helpers that drive the embedded Control UI through the
Gateway `browser.request` seam without adding a special-case runner.
Scenario files should be grouped by product capability rather than source tree
folder. Keep scenario IDs stable when files move; use `docsRefs` and `codeRefs`
for implementation traceability.
The baseline list should stay broad enough to cover:
- DM and channel chat

View File

@@ -89,7 +89,21 @@ Gateway → Client:
```
`server`, `features`, `snapshot`, and `policy` are all required by the schema
(`src/gateway/protocol/schema/frames.ts`). `auth` and `canvasHostUrl` are optional.
(`src/gateway/protocol/schema/frames.ts`). `canvasHostUrl` is optional. `auth`
reports the negotiated role/scopes when available, and includes `deviceToken`
when the gateway issues one.
When no device token is issued, `hello-ok.auth` can still report the negotiated
permissions:
```json
{
"auth": {
"role": "operator",
"scopes": ["operator.read", "operator.write"]
}
}
```
When a device token is issued, `hello-ok` also includes:

View File

@@ -213,7 +213,7 @@ The minimum adoption bar for a new channel is:
4. Mount the runner as `openclaw qa <runner>` instead of registering a competing root command.
Runner plugins should declare `qaRunners` in `openclaw.plugin.json` and export a matching `qaRunnerCliRegistrations` array from `runtime-api.ts`.
Keep `runtime-api.ts` light; lazy CLI and runner execution should stay behind separate entrypoints.
5. Author or adapt markdown scenarios under `qa/scenarios/`.
5. Author or adapt markdown scenarios under the themed `qa/scenarios/` directories.
6. Use the generic scenario helpers for new scenarios.
7. Keep existing compatibility aliases working unless the repo is doing an intentional migration.

View File

@@ -55,7 +55,7 @@ The macOS app presents itself as a node. Common commands:
- Canvas: `canvas.present`, `canvas.navigate`, `canvas.eval`, `canvas.snapshot`, `canvas.a2ui.*`
- Camera: `camera.snap`, `camera.clip`
- Screen: `screen.record`
- Screen: `screen.snapshot`, `screen.record`
- System: `system.run`, `system.notify`
The node reports a `permissions` map so agents can decide whats allowed.

View File

@@ -185,7 +185,9 @@ Keep inbound mention handling split in two layers:
- plugin-owned evidence gathering
- shared policy evaluation
Use `openclaw/plugin-sdk/channel-inbound` for the shared layer.
Use `openclaw/plugin-sdk/channel-mention-gating` for mention-policy decisions.
Use `openclaw/plugin-sdk/channel-inbound` only when you need the broader inbound
helper barrel.
Good fit for plugin-local logic:
@@ -255,6 +257,11 @@ bundled channel plugins that already depend on runtime injection:
- `implicitMentionKindWhen`
- `resolveInboundMentionDecision`
If you only need `implicitMentionKindWhen` and
`resolveInboundMentionDecision`, import from
`openclaw/plugin-sdk/channel-mention-gating` to avoid loading unrelated inbound
runtime helpers.
The older `resolveMentionGating*` helpers remain on
`openclaw/plugin-sdk/channel-inbound` as compatibility exports only. New code
should use `resolveInboundMentionDecision({ facts, policy })`.

View File

@@ -318,7 +318,7 @@ Current bundled provider examples:
| `plugin-sdk/memory-core` | Bundled memory-core helpers | Memory manager/config/file/CLI helper surface |
| `plugin-sdk/memory-core-engine-runtime` | Memory engine runtime facade | Memory index/search runtime facade |
| `plugin-sdk/memory-core-host-engine-foundation` | Memory host foundation engine | Memory host foundation engine exports |
| `plugin-sdk/memory-core-host-engine-embeddings` | Memory host embedding engine | Memory host embedding engine exports |
| `plugin-sdk/memory-core-host-engine-embeddings` | Memory host embedding engine | Memory embedding contracts, registry access, local provider, and generic batch/remote helpers; concrete remote providers live in their owning plugins |
| `plugin-sdk/memory-core-host-engine-qmd` | Memory host QMD engine | Memory host QMD engine exports |
| `plugin-sdk/memory-core-host-engine-storage` | Memory host storage engine | Memory host storage engine exports |
| `plugin-sdk/memory-core-host-multimodal` | Memory host multimodal helpers | Memory host multimodal helpers |

View File

@@ -88,6 +88,7 @@ explicitly promotes one as public.
| `plugin-sdk/channel-config-helpers` | `createHybridChannelConfigAdapter` |
| `plugin-sdk/channel-config-schema` | Channel config schema types |
| `plugin-sdk/telegram-command-config` | Telegram custom-command normalization/validation helpers with bundled-contract fallback |
| `plugin-sdk/command-gating` | Narrow command authorization gate helpers |
| `plugin-sdk/channel-policy` | `resolveChannelGroupRequireMention` |
| `plugin-sdk/channel-lifecycle` | `createAccountStatusSink` |
| `plugin-sdk/inbound-envelope` | Shared inbound route + envelope builder helpers |
@@ -95,6 +96,7 @@ explicitly promotes one as public.
| `plugin-sdk/messaging-targets` | Target parsing/matching helpers |
| `plugin-sdk/outbound-media` | Shared outbound media loading helpers |
| `plugin-sdk/outbound-runtime` | Outbound identity/send delegate helpers |
| `plugin-sdk/poll-runtime` | Narrow poll normalization helpers |
| `plugin-sdk/thread-bindings-runtime` | Thread-binding lifecycle and adapter helpers |
| `plugin-sdk/agent-media-payload` | Legacy agent media payload builder |
| `plugin-sdk/conversation-runtime` | Conversation/thread binding, pairing, and configured-binding helpers |
@@ -108,7 +110,10 @@ explicitly promotes one as public.
| `plugin-sdk/group-access` | Shared group-access decision helpers |
| `plugin-sdk/direct-dm` | Shared direct-DM auth/guard helpers |
| `plugin-sdk/interactive-runtime` | Interactive reply payload normalization/reduction helpers |
| `plugin-sdk/channel-inbound` | Inbound debounce, mention matching, mention-policy helpers, and envelope helpers |
| `plugin-sdk/channel-inbound` | Compatibility barrel for inbound debounce, mention matching, mention-policy helpers, and envelope helpers |
| `plugin-sdk/channel-mention-gating` | Narrow mention-policy helpers without the broader inbound runtime surface |
| `plugin-sdk/channel-location` | Channel location context and formatting helpers |
| `plugin-sdk/channel-logging` | Channel logging helpers for inbound drops and typing/ack failures |
| `plugin-sdk/channel-send-result` | Reply result types |
| `plugin-sdk/channel-actions` | `createMessageToolButtonsSchema`, `createMessageToolCardSchema` |
| `plugin-sdk/channel-targets` | Target parsing/matching helpers |
@@ -166,6 +171,7 @@ explicitly promotes one as public.
| `plugin-sdk/secret-ref-runtime` | Narrow `coerceSecretRef` and SecretRef typing helpers for secret-contract/config parsing |
| `plugin-sdk/security-runtime` | Shared trust, DM gating, external-content, and secret-collection helpers |
| `plugin-sdk/ssrf-policy` | Host allowlist and private-network SSRF policy helpers |
| `plugin-sdk/ssrf-dispatcher` | Narrow pinned-dispatcher helpers without the broad infra runtime surface |
| `plugin-sdk/ssrf-runtime` | Pinned-dispatcher, SSRF-guarded fetch, and SSRF policy helpers |
| `plugin-sdk/secret-input` | Secret input parsing helpers |
| `plugin-sdk/webhook-ingress` | Webhook request/target helpers |
@@ -187,6 +193,7 @@ explicitly promotes one as public.
| `plugin-sdk/gateway-runtime` | Gateway client and channel-status patch helpers |
| `plugin-sdk/config-runtime` | Config load/write helpers |
| `plugin-sdk/telegram-command-config` | Telegram command-name/description normalization and duplicate/conflict checks, even when the bundled Telegram contract surface is unavailable |
| `plugin-sdk/text-autolink-runtime` | File-reference autolink detection without the broad text-runtime barrel |
| `plugin-sdk/approval-runtime` | Exec/plugin approval helpers, approval-capability builders, auth/profile helpers, native routing/runtime helpers |
| `plugin-sdk/reply-runtime` | Shared inbound/reply runtime helpers, chunking, dispatch, heartbeat, reply planner |
| `plugin-sdk/reply-dispatch-runtime` | Narrow reply dispatch/finalize helpers |
@@ -211,6 +218,7 @@ explicitly promotes one as public.
| `plugin-sdk/file-lock` | Re-entrant file-lock helpers |
| `plugin-sdk/persistent-dedupe` | Disk-backed dedupe cache helpers |
| `plugin-sdk/acp-runtime` | ACP runtime/session and reply-dispatch helpers |
| `plugin-sdk/acp-binding-resolve-runtime` | Read-only ACP binding resolution without lifecycle startup imports |
| `plugin-sdk/agent-config-primitives` | Narrow agent runtime config-schema primitives |
| `plugin-sdk/boolean-param` | Loose boolean param reader |
| `plugin-sdk/dangerous-name-runtime` | Dangerous-name matching resolution helpers |
@@ -226,6 +234,12 @@ explicitly promotes one as public.
| `plugin-sdk/diagnostic-runtime` | Diagnostic flag and event helpers |
| `plugin-sdk/error-runtime` | Error graph, formatting, shared error classification helpers, `isApprovalNotFoundError` |
| `plugin-sdk/fetch-runtime` | Wrapped fetch, proxy, and pinned lookup helpers |
| `plugin-sdk/runtime-fetch` | Dispatcher-aware runtime fetch without proxy/guarded-fetch imports |
| `plugin-sdk/response-limit-runtime` | Bounded response-body reader without the broad media runtime surface |
| `plugin-sdk/session-binding-runtime` | Current conversation binding state without configured binding routing or pairing stores |
| `plugin-sdk/session-store-runtime` | Session-store read helpers without broad config writes/maintenance imports |
| `plugin-sdk/context-visibility-runtime` | Context visibility resolution and supplemental context filtering without broad config/security imports |
| `plugin-sdk/string-coerce-runtime` | Narrow primitive record/string coercion and normalization helpers without markdown/logging imports |
| `plugin-sdk/host-runtime` | Hostname and SCP host normalization helpers |
| `plugin-sdk/retry-runtime` | Retry config and retry runner helpers |
| `plugin-sdk/agent-runtime` | Agent dir/identity/workspace helpers |
@@ -264,7 +278,7 @@ explicitly promotes one as public.
| `plugin-sdk/memory-core` | Bundled memory-core helper surface for manager/config/file/CLI helpers |
| `plugin-sdk/memory-core-engine-runtime` | Memory index/search runtime facade |
| `plugin-sdk/memory-core-host-engine-foundation` | Memory host foundation engine exports |
| `plugin-sdk/memory-core-host-engine-embeddings` | Memory host embedding engine exports |
| `plugin-sdk/memory-core-host-engine-embeddings` | Memory host embedding contracts, registry access, local provider, and generic batch/remote helpers |
| `plugin-sdk/memory-core-host-engine-qmd` | Memory host QMD engine exports |
| `plugin-sdk/memory-core-host-engine-storage` | Memory host storage engine exports |
| `plugin-sdk/memory-core-host-multimodal` | Memory host multimodal helpers |

View File

@@ -18,7 +18,7 @@ The desired end state is a generic QA harness that loads powerful scenario defin
## Current State
Primary source of truth now lives in `qa/scenarios/index.md` plus one file per
scenario under `qa/scenarios/*.md`.
scenario under `qa/scenarios/<theme>/*.md`.
Implemented:
@@ -26,7 +26,7 @@ Implemented:
- canonical QA pack metadata
- operator identity
- kickoff mission
- `qa/scenarios/*.md`
- `qa/scenarios/<theme>/*.md`
- one markdown file per scenario
- scenario metadata
- handler bindings
@@ -107,8 +107,8 @@ These categories matter because they drive DSL requirements. A flat list of prom
### Single source of truth
Use `qa/scenarios/index.md` plus `qa/scenarios/*.md` as the authored source of
truth.
Use `qa/scenarios/index.md` plus `qa/scenarios/<theme>/*.md` as the authored
source of truth.
The pack should stay:
@@ -363,7 +363,7 @@ Generated compatibility:
Done.
- added `qa/scenarios/index.md`
- split scenarios into `qa/scenarios/*.md`
- split scenarios into `qa/scenarios/<theme>/*.md`
- added parser for named markdown YAML pack content
- validated with zod
- switched consumers to the parsed pack

View File

@@ -82,7 +82,10 @@ html.dark .nav-tabs-underline {
border-radius: 8px;
background: color-mix(in oklab, rgb(var(--primary)) 4%, transparent);
text-decoration: none;
transition: transform 0.16s ease, border-color 0.16s ease, background 0.16s ease;
transition:
transform 0.16s ease,
border-color 0.16s ease,
background 0.16s ease;
}
.showcase-actions a:first-child {

View File

@@ -60,6 +60,10 @@ third-party plugins see.
- Do not rely on eager global registry seeding or import-time side effects to
make a plugin “available”. Plugin availability should come from manifest
ownership plus targeted activation.
- When core needs plugin-owned static data on a hot path, expose a lightweight
top-level artifact such as `gateway-auth-api.ts`, `message-tool-api.ts`, or a
similarly narrow `*-api.ts`. Reuse the same local helper from the artifact and
the full plugin so fast paths do not drift from runtime behavior.
## Expanding The Boundary

View File

@@ -119,7 +119,7 @@ describe("active-memory plugin", () => {
runEmbeddedPiAgent.mockResolvedValue({
payloads: [{ text: "- lemon pepper wings\n- blue cheese" }],
});
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
});
afterEach(async () => {
@@ -425,7 +425,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
allowedChatTypes: ["direct", "group"],
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
const result = await hooks.before_prompt_build(
{ prompt: "what wings should we order?", messages: [] },
@@ -513,7 +513,7 @@ describe("active-memory plugin", () => {
searchMode: "inherit",
},
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -602,7 +602,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
queryMode: "message",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -630,7 +630,7 @@ describe("active-memory plugin", () => {
queryMode: "message",
promptStyle: "preference-only",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -675,7 +675,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
thinking: "medium",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -701,7 +701,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
promptAppend: "Prefer stable long-term preferences over one-off events.",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -730,7 +730,7 @@ describe("active-memory plugin", () => {
promptOverride: "Custom memory prompt. Return NONE or one user fact.",
promptAppend: "Extra custom instruction.",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -802,7 +802,7 @@ describe("active-memory plugin", () => {
api.pluginConfig = {
agents: ["main"],
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{ prompt: "what wings should i order? temp transcript", messages: [] },
@@ -828,7 +828,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
modelFallbackPolicy: "resolved-only",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
const result = await hooks.before_prompt_build(
{ prompt: "what wings should i order? no fallback", messages: [] },
@@ -851,7 +851,7 @@ describe("active-memory plugin", () => {
modelFallback: "google/gemini-3-flash",
modelFallbackPolicy: "default-remote",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{ prompt: "what wings should i order? custom fallback", messages: [] },
@@ -878,7 +878,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
modelFallbackPolicy: "default-remote",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
const result = await hooks.before_prompt_build(
{ prompt: "what wings should i order? built-in fallback", messages: [] },
@@ -1027,7 +1027,7 @@ describe("active-memory plugin", () => {
timeoutMs: 250,
logging: true,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
let lastAbortSignal: AbortSignal | undefined;
runEmbeddedPiAgent.mockImplementation(async (params: { abortSignal?: AbortSignal }) => {
lastAbortSignal = params.abortSignal;
@@ -1073,7 +1073,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
logging: true,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{ prompt: "what wings should i order? session id cache", messages: [] },
@@ -1107,7 +1107,7 @@ describe("active-memory plugin", () => {
timeoutMs: 250,
logging: true,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
runEmbeddedPiAgent.mockImplementationOnce(async (params: { timeoutMs?: number }) => {
await new Promise((resolve) => setTimeout(resolve, (params.timeoutMs ?? 0) + 25));
return {
@@ -1145,7 +1145,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
logging: true,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{ prompt: "what wings should i order? log sanitization", messages: [] },
@@ -1179,7 +1179,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
logging: true,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
const hugeSession = `agent:main:${"x".repeat(500)}`;
await hooks.before_prompt_build(
@@ -1423,7 +1423,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
queryMode: "message",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -1451,7 +1451,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
queryMode: "full",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -1482,7 +1482,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
queryMode: "recent",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -1536,7 +1536,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
queryMode: "recent",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -1578,7 +1578,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
queryMode: "recent",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -1611,7 +1611,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
queryMode: "recent",
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{
@@ -1619,8 +1619,7 @@ describe("active-memory plugin", () => {
messages: [
{
role: "user",
content:
"Active Memory: I really do want you to remember that I prefer aisle seats.",
content: "Active Memory: I really do want you to remember that I prefer aisle seats.",
},
{
role: "user",
@@ -1674,7 +1673,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
maxSummaryChars: 40,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
runEmbeddedPiAgent.mockResolvedValueOnce({
payloads: [
{
@@ -1708,7 +1707,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
maxSummaryChars: 90,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
await hooks.before_prompt_build(
{ prompt: "what wings should i order? prompt-count-check", messages: [] },
@@ -1758,7 +1757,7 @@ describe("active-memory plugin", () => {
transcriptDir: "active-memory-subagents",
logging: true,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
const mkdirSpy = vi.spyOn(fs, "mkdir").mockResolvedValue(undefined);
const mkdtempSpy = vi.spyOn(fs, "mkdtemp");
const rmSpy = vi.spyOn(fs, "rm").mockResolvedValue(undefined);
@@ -1802,7 +1801,7 @@ describe("active-memory plugin", () => {
transcriptDir: "C:/temp/escape",
logging: true,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
const mkdirSpy = vi.spyOn(fs, "mkdir").mockResolvedValue(undefined);
await hooks.before_prompt_build(
@@ -1839,7 +1838,7 @@ describe("active-memory plugin", () => {
transcriptDir: "active-memory-subagents",
logging: true,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
const mkdirSpy = vi.spyOn(fs, "mkdir").mockResolvedValue(undefined);
await hooks.before_prompt_build(
@@ -1906,7 +1905,7 @@ describe("active-memory plugin", () => {
agents: ["main"],
logging: true,
};
await plugin.register(api as unknown as OpenClawPluginApi);
plugin.register(api as unknown as OpenClawPluginApi);
for (let index = 0; index <= 1000; index += 1) {
await hooks.before_prompt_build(

View File

@@ -1527,7 +1527,9 @@ function extractRecentTurns(messages: unknown[]): ActiveRecallRecentTurn[] {
}
const rawText = extractTextContent(typed.content);
const text =
role === "assistant" ? stripRecalledContextNoise(rawText) : stripInjectedActiveMemoryPrefixOnly(rawText);
role === "assistant"
? stripRecalledContextNoise(rawText)
: stripInjectedActiveMemoryPrefixOnly(rawText);
if (!text) {
continue;
}

View File

@@ -1,7 +1,10 @@
import { normalizeLowercaseStringOrEmpty } from "../../shared/string-coerce.js";
import { sanitizeAndNormalizeEmbedding } from "./embedding-vectors.js";
import { debugEmbeddingsLog } from "./embeddings-debug.js";
import type { EmbeddingProvider, EmbeddingProviderOptions } from "./embeddings.types.js";
import {
debugEmbeddingsLog,
sanitizeAndNormalizeEmbedding,
type MemoryEmbeddingProvider,
type MemoryEmbeddingProviderCreateOptions,
} from "openclaw/plugin-sdk/memory-core-host-engine-embeddings";
import { normalizeLowercaseStringOrEmpty } from "openclaw/plugin-sdk/text-runtime";
// ---------------------------------------------------------------------------
// Types & constants
@@ -254,8 +257,8 @@ function parseCohereBatch(family: Family, raw: string): number[][] {
// ---------------------------------------------------------------------------
export async function createBedrockEmbeddingProvider(
options: EmbeddingProviderOptions,
): Promise<{ provider: EmbeddingProvider; client: BedrockEmbeddingClient }> {
options: MemoryEmbeddingProviderCreateOptions,
): Promise<{ provider: MemoryEmbeddingProvider; client: BedrockEmbeddingClient }> {
const client = resolveBedrockEmbeddingClient(options);
const { BedrockRuntimeClient, InvokeModelCommand } = await loadSdk();
const sdk = new BedrockRuntimeClient({ region: client.region });
@@ -333,7 +336,7 @@ export async function createBedrockEmbeddingProvider(
// ---------------------------------------------------------------------------
export function resolveBedrockEmbeddingClient(
options: EmbeddingProviderOptions,
options: MemoryEmbeddingProviderCreateOptions,
): BedrockEmbeddingClient {
const model = normalizeBedrockEmbeddingModel(options.model);
const spec = resolveSpec(model);

View File

@@ -0,0 +1,37 @@
import {
isMissingEmbeddingApiKeyError,
type MemoryEmbeddingProviderAdapter,
} from "openclaw/plugin-sdk/memory-core-host-engine-embeddings";
import {
createBedrockEmbeddingProvider,
DEFAULT_BEDROCK_EMBEDDING_MODEL,
} from "./embedding-provider.js";
export const bedrockMemoryEmbeddingProviderAdapter: MemoryEmbeddingProviderAdapter = {
id: "bedrock",
defaultModel: DEFAULT_BEDROCK_EMBEDDING_MODEL,
transport: "remote",
authProviderId: "amazon-bedrock",
autoSelectPriority: 60,
allowExplicitWhenConfiguredAuto: true,
shouldContinueAutoSelection: isMissingEmbeddingApiKeyError,
create: async (options) => {
const { provider, client } = await createBedrockEmbeddingProvider({
...options,
provider: "bedrock",
fallback: "none",
});
return {
provider,
runtime: {
id: "bedrock",
cacheKeyData: {
provider: "bedrock",
region: client.region,
model: client.model,
dimensions: client.dimensions,
},
},
};
},
};

View File

@@ -2,6 +2,9 @@
"id": "amazon-bedrock",
"enabledByDefault": true,
"providers": ["amazon-bedrock"],
"contracts": {
"memoryEmbeddingProviders": ["bedrock"]
},
"configSchema": {
"type": "object",
"additionalProperties": false,

View File

@@ -5,7 +5,9 @@
"description": "OpenClaw Amazon Bedrock provider plugin",
"type": "module",
"dependencies": {
"@aws-sdk/client-bedrock": "3.1028.0"
"@aws-sdk/client-bedrock": "3.1028.0",
"@aws-sdk/client-bedrock-runtime": "3.1028.0",
"@aws-sdk/credential-provider-node": "3.972.30"
},
"devDependencies": {
"@openclaw/plugin-sdk": "workspace:*"

View File

@@ -14,6 +14,7 @@ import {
resolveBedrockConfigApiKey,
resolveImplicitBedrockProvider,
} from "./api.js";
import { bedrockMemoryEmbeddingProviderAdapter } from "./memory-embedding-adapter.js";
type GuardrailConfig = {
guardrailIdentifier: string;
@@ -78,6 +79,8 @@ export function registerAmazonBedrockPlugin(api: OpenClawPluginApi): void {
const pluginConfig = (api.pluginConfig ?? {}) as AmazonBedrockPluginConfig;
const guardrail = pluginConfig.guardrail;
api.registerMemoryEmbeddingProvider(bedrockMemoryEmbeddingProviderAdapter);
const baseWrapStreamFn = ({ modelId, streamFn }: { modelId: string; streamFn?: StreamFn }) =>
isAnthropicBedrockModel(modelId) ? streamFn : createBedrockNoCacheWrapper(streamFn);

View File

@@ -1,4 +1,5 @@
export { CLAUDE_CLI_BACKEND_ID, isClaudeCliProvider } from "./cli-shared.js";
export { buildAnthropicProvider } from "./register.runtime.js";
export {
createAnthropicBetaHeadersWrapper,
createAnthropicFastModeWrapper,

View File

@@ -18,7 +18,10 @@ import {
upsertAuthProfile,
validateAnthropicSetupToken,
} from "openclaw/plugin-sdk/provider-auth";
import { cloneFirstTemplateModel } from "openclaw/plugin-sdk/provider-model-shared";
import {
cloneFirstTemplateModel,
type ProviderPlugin,
} from "openclaw/plugin-sdk/provider-model-shared";
import { fetchClaudeUsage } from "openclaw/plugin-sdk/provider-usage";
import { normalizeLowercaseStringOrEmpty } from "openclaw/plugin-sdk/text-runtime";
import * as claudeCliAuth from "./cli-auth-seam.js";
@@ -395,11 +398,10 @@ async function runAnthropicCliMigrationNonInteractive(ctx: {
};
}
export function registerAnthropicPlugin(api: OpenClawPluginApi): void {
export function buildAnthropicProvider(): ProviderPlugin {
const providerId = "anthropic";
const defaultAnthropicModel = DEFAULT_ANTHROPIC_MODEL;
api.registerCliBackend(buildAnthropicCliBackend());
api.registerProvider({
return {
id: providerId,
label: "Anthropic",
docsPath: "/providers/models",
@@ -505,6 +507,11 @@ export function registerAnthropicPlugin(api: OpenClawPluginApi): void {
store: ctx.store,
profileId: ctx.profileId,
}),
});
};
}
export function registerAnthropicPlugin(api: OpenClawPluginApi): void {
api.registerCliBackend(buildAnthropicCliBackend());
api.registerProvider(buildAnthropicProvider());
api.registerMediaUnderstandingProvider(anthropicMediaUnderstandingProvider);
}

View File

@@ -2,3 +2,7 @@ export {
collectRuntimeConfigAssignments,
secretTargetRegistryEntries,
} from "./src/secret-contract.js";
export {
__testing as blueBubblesConversationBindingTesting,
createBlueBubblesConversationBindingManager,
} from "./src/conversation-bindings.js";

View File

@@ -0,0 +1,46 @@
/** Resolve the correct targetId after a navigation that may trigger a renderer swap. */
export async function resolveTargetIdAfterNavigate(opts: {
oldTargetId: string;
navigatedUrl: string;
listTabs: () => Promise<Array<{ targetId: string; url: string }>>;
retryDelayMs?: number;
}): Promise<string> {
let currentTargetId = opts.oldTargetId;
try {
const pickReplacement = (
tabs: Array<{ targetId: string; url: string }>,
options?: { allowSingleTabFallback?: boolean },
): { targetId: string; shouldRetry: boolean } => {
if (tabs.some((tab) => tab.targetId === opts.oldTargetId)) {
return { targetId: opts.oldTargetId, shouldRetry: false };
}
const byUrl = tabs.filter((tab) => tab.url === opts.navigatedUrl);
if (byUrl.length === 1) {
return { targetId: byUrl[0]?.targetId ?? opts.oldTargetId, shouldRetry: false };
}
const uniqueReplacement = byUrl.filter((tab) => tab.targetId !== opts.oldTargetId);
if (uniqueReplacement.length === 1) {
return {
targetId: uniqueReplacement[0]?.targetId ?? opts.oldTargetId,
shouldRetry: false,
};
}
if (options?.allowSingleTabFallback && tabs.length === 1) {
return { targetId: tabs[0]?.targetId ?? opts.oldTargetId, shouldRetry: false };
}
return { targetId: opts.oldTargetId, shouldRetry: true };
};
const first = pickReplacement(await opts.listTabs());
currentTargetId = first.targetId;
if (first.shouldRetry) {
await new Promise((r) => setTimeout(r, opts.retryDelayMs ?? 800));
currentTargetId = pickReplacement(await opts.listTabs(), {
allowSingleTabFallback: true,
}).targetId;
}
} catch {
// Best-effort: fall back to pre-navigation targetId.
}
return currentTargetId;
}

View File

@@ -1,5 +1,5 @@
import { beforeEach, describe, expect, it, vi } from "vitest";
import { resolveTargetIdAfterNavigate } from "./agent.snapshot.js";
import { describe, expect, it } from "vitest";
import { resolveTargetIdAfterNavigate } from "./agent.snapshot-target.js";
type Tab = { targetId: string; url: string };
@@ -8,10 +8,6 @@ function staticListTabs(tabs: Tab[]): () => Promise<Tab[]> {
}
describe("resolveTargetIdAfterNavigate", () => {
beforeEach(() => {
vi.useRealTimers();
});
it("returns original targetId when old target still exists (no swap)", async () => {
const result = await resolveTargetIdAfterNavigate({
oldTargetId: "old-123",
@@ -37,6 +33,7 @@ describe("resolveTargetIdAfterNavigate", () => {
const result = await resolveTargetIdAfterNavigate({
oldTargetId: "old-123",
navigatedUrl: "https://example.com",
retryDelayMs: 0,
listTabs: staticListTabs([
{ targetId: "preexisting-000", url: "https://example.com" },
{ targetId: "fresh-777", url: "https://example.com" },
@@ -47,12 +44,12 @@ describe("resolveTargetIdAfterNavigate", () => {
});
it("retries and resolves targetId when first listTabs has no URL match", async () => {
vi.useFakeTimers();
let calls = 0;
const result$ = resolveTargetIdAfterNavigate({
const result = await resolveTargetIdAfterNavigate({
oldTargetId: "old-123",
navigatedUrl: "https://delayed.com",
retryDelayMs: 0,
listTabs: async () => {
calls++;
if (calls === 1) {
@@ -62,50 +59,33 @@ describe("resolveTargetIdAfterNavigate", () => {
},
});
await vi.advanceTimersByTimeAsync(800);
const result = await result$;
expect(result).toBe("delayed-999");
expect(calls).toBe(2);
vi.useRealTimers();
});
it("falls back to original targetId when no match found after retry", async () => {
vi.useFakeTimers();
const result$ = resolveTargetIdAfterNavigate({
const result = await resolveTargetIdAfterNavigate({
oldTargetId: "old-123",
navigatedUrl: "https://no-match.com",
retryDelayMs: 0,
listTabs: staticListTabs([
{ targetId: "unrelated-1", url: "https://unrelated.com" },
{ targetId: "unrelated-2", url: "https://unrelated2.com" },
]),
});
await vi.advanceTimersByTimeAsync(800);
const result = await result$;
expect(result).toBe("old-123");
vi.useRealTimers();
});
it("falls back to single remaining tab when no URL match after retry", async () => {
vi.useFakeTimers();
const result$ = resolveTargetIdAfterNavigate({
const result = await resolveTargetIdAfterNavigate({
oldTargetId: "old-123",
navigatedUrl: "https://single-tab.com",
retryDelayMs: 0,
listTabs: staticListTabs([{ targetId: "only-tab", url: "https://some-other.com" }]),
});
await vi.advanceTimersByTimeAsync(800);
const result = await result$;
expect(result).toBe("only-tab");
vi.useRealTimers();
});
it("falls back to original targetId when listTabs throws", async () => {
@@ -120,22 +100,16 @@ describe("resolveTargetIdAfterNavigate", () => {
});
it("keeps the old target when multiple replacement candidates still match after retry", async () => {
vi.useFakeTimers();
const result$ = resolveTargetIdAfterNavigate({
const result = await resolveTargetIdAfterNavigate({
oldTargetId: "old-123",
navigatedUrl: "https://example.com",
retryDelayMs: 0,
listTabs: staticListTabs([
{ targetId: "preexisting-000", url: "https://example.com" },
{ targetId: "fresh-777", url: "https://example.com" },
]),
});
await vi.advanceTimersByTimeAsync(800);
const result = await result$;
expect(result).toBe("old-123");
vi.useRealTimers();
});
});

View File

@@ -32,6 +32,7 @@ import {
withPlaywrightRouteContext,
withRouteTabContext,
} from "./agent.shared.js";
import { resolveTargetIdAfterNavigate } from "./agent.snapshot-target.js";
import {
resolveSnapshotPlan,
shouldUsePlaywrightForAriaSnapshot,
@@ -172,48 +173,6 @@ async function saveBrowserMediaResponse(params: {
});
}
/** Resolve the correct targetId after a navigation that may trigger a renderer swap. */
export async function resolveTargetIdAfterNavigate(opts: {
oldTargetId: string;
navigatedUrl: string;
listTabs: () => Promise<Array<{ targetId: string; url: string }>>;
}): Promise<string> {
let currentTargetId = opts.oldTargetId;
try {
const pickReplacement = (
tabs: Array<{ targetId: string; url: string }>,
options?: { allowSingleTabFallback?: boolean },
) => {
if (tabs.some((tab) => tab.targetId === opts.oldTargetId)) {
return opts.oldTargetId;
}
const byUrl = tabs.filter((tab) => tab.url === opts.navigatedUrl);
if (byUrl.length === 1) {
return byUrl[0]?.targetId ?? opts.oldTargetId;
}
const uniqueReplacement = byUrl.filter((tab) => tab.targetId !== opts.oldTargetId);
if (uniqueReplacement.length === 1) {
return uniqueReplacement[0]?.targetId ?? opts.oldTargetId;
}
if (options?.allowSingleTabFallback && tabs.length === 1) {
return tabs[0]?.targetId ?? opts.oldTargetId;
}
return opts.oldTargetId;
};
currentTargetId = pickReplacement(await opts.listTabs());
if (currentTargetId === opts.oldTargetId) {
await new Promise((r) => setTimeout(r, 800));
currentTargetId = pickReplacement(await opts.listTabs(), {
allowSingleTabFallback: true,
});
}
} catch {
// Best-effort: fall back to pre-navigation targetId
}
return currentTargetId;
}
export function registerBrowserAgentSnapshotRoutes(
app: BrowserRouteRegistrar,
ctx: BrowserRouteContext,

View File

@@ -1,176 +0,0 @@
import { describe, expect, it } from "vitest";
import {
installAgentContractHooks,
startServerAndBase,
} from "./server.agent-contract.test-harness.js";
import {
setBrowserControlServerEvaluateEnabled,
setBrowserControlServerProfiles,
} from "./server.control-server.test-harness.js";
import { getBrowserTestFetch } from "./test-fetch.js";
type ActErrorResponse = {
error?: string;
code?: string;
};
type ActErrorHttpResponse = {
status: number;
body: ActErrorResponse;
};
async function postActAndReadError(base: string, body?: unknown): Promise<ActErrorHttpResponse> {
const realFetch = getBrowserTestFetch();
const response = await realFetch(`${base}/act`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: body === undefined ? undefined : JSON.stringify(body),
});
return {
status: response.status,
body: (await response.json()) as ActErrorResponse,
};
}
describe("browser control server", () => {
installAgentContractHooks();
const slowTimeoutMs = process.platform === "win32" ? 40_000 : 20_000;
it(
"returns ACT_KIND_REQUIRED when kind is missing",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {});
expect(response.status).toBe(400);
expect(response.body.code).toBe("ACT_KIND_REQUIRED");
expect(response.body.error).toContain("kind is required");
},
slowTimeoutMs,
);
it(
"returns ACT_INVALID_REQUEST for malformed action payloads",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "click",
ref: {},
});
expect(response.status).toBe(400);
expect(response.body.code).toBe("ACT_INVALID_REQUEST");
expect(response.body.error).toContain("click requires ref or selector");
},
slowTimeoutMs,
);
it(
"returns ACT_EXISTING_SESSION_UNSUPPORTED for unsupported existing-session actions",
async () => {
setBrowserControlServerProfiles({
openclaw: {
color: "#FF4500",
driver: "existing-session",
},
});
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "batch",
actions: [{ kind: "press", key: "Enter" }],
});
expect(response.status).toBe(501);
expect(response.body.code).toBe("ACT_EXISTING_SESSION_UNSUPPORTED");
expect(response.body.error).toContain("batch");
},
slowTimeoutMs,
);
it(
"returns ACT_TARGET_ID_MISMATCH for batched action targetId overrides",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "batch",
actions: [{ kind: "click", ref: "5", targetId: "other-tab" }],
});
expect(response.status).toBe(403);
expect(response.body.code).toBe("ACT_TARGET_ID_MISMATCH");
expect(response.body.error).toContain("batched action targetId must match request targetId");
},
slowTimeoutMs,
);
it(
"returns ACT_TARGET_ID_MISMATCH for top-level action targetId overrides",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "click",
ref: "5",
// Intentionally non-string: route-level target selection ignores this,
// while action normalization stringifies it.
targetId: 12345,
});
expect(response.status).toBe(403);
expect(response.body.code).toBe("ACT_TARGET_ID_MISMATCH");
expect(response.body.error).toContain("action targetId must match request targetId");
},
slowTimeoutMs,
);
it(
"returns ACT_SELECTOR_UNSUPPORTED for selector on unsupported action kinds",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "evaluate",
fn: "() => 1",
selector: "#submit",
});
expect(response.status).toBe(400);
expect(response.body.code).toBe("ACT_SELECTOR_UNSUPPORTED");
expect(response.body.error).toContain("'selector' is not supported");
},
slowTimeoutMs,
);
it(
"returns ACT_INVALID_REQUEST for malformed unsupported selector actions before selector gating",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "press",
selector: "#submit",
});
expect(response.status).toBe(400);
expect(response.body.code).toBe("ACT_INVALID_REQUEST");
expect(response.body.error).toContain("press requires key");
},
slowTimeoutMs,
);
it(
"returns ACT_EVALUATE_DISABLED when evaluate is blocked by config",
async () => {
setBrowserControlServerEvaluateEnabled(false);
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "evaluate",
fn: "() => 1",
});
expect(response.status).toBe(403);
expect(response.body.code).toBe("ACT_EVALUATE_DISABLED");
expect(response.body.error).toContain("browser.evaluateEnabled=false");
},
slowTimeoutMs,
);
});

View File

@@ -0,0 +1,577 @@
import fs from "node:fs";
import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";
import { DEFAULT_AI_SNAPSHOT_MAX_CHARS } from "./constants.js";
import {
installAgentContractHooks,
postJson,
startServerAndBase,
} from "./server.agent-contract.test-harness.js";
import {
cleanupBrowserControlServerTestContext,
getBrowserControlServerBaseUrl,
getBrowserControlServerTestState,
getCdpMocks,
getPwMocks,
makeResponse,
resetBrowserControlServerTestContext,
setBrowserControlServerEvaluateEnabled,
setBrowserControlServerProfiles,
setBrowserControlServerReachable,
startBrowserControlServerFromConfig,
} from "./server.control-server.test-harness.js";
import { getBrowserTestFetch } from "./test-fetch.js";
type ActErrorResponse = {
error?: string;
code?: string;
};
type ActErrorHttpResponse = {
status: number;
body: ActErrorResponse;
};
async function postActAndReadError(base: string, body?: unknown): Promise<ActErrorHttpResponse> {
const realFetch = getBrowserTestFetch();
const response = await realFetch(`${base}/act`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: body === undefined ? undefined : JSON.stringify(body),
});
return {
status: response.status,
body: (await response.json()) as ActErrorResponse,
};
}
const state = getBrowserControlServerTestState();
const cdpMocks = getCdpMocks();
const pwMocks = getPwMocks();
describe("browser control server", () => {
installAgentContractHooks();
const slowTimeoutMs = process.platform === "win32" ? 40_000 : 20_000;
it(
"returns ACT_KIND_REQUIRED when kind is missing",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {});
expect(response.status).toBe(400);
expect(response.body.code).toBe("ACT_KIND_REQUIRED");
expect(response.body.error).toContain("kind is required");
},
slowTimeoutMs,
);
it(
"returns ACT_INVALID_REQUEST for malformed action payloads",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "click",
ref: {},
});
expect(response.status).toBe(400);
expect(response.body.code).toBe("ACT_INVALID_REQUEST");
expect(response.body.error).toContain("click requires ref or selector");
},
slowTimeoutMs,
);
it(
"returns ACT_EXISTING_SESSION_UNSUPPORTED for unsupported existing-session actions",
async () => {
setBrowserControlServerProfiles({
openclaw: {
color: "#FF4500",
driver: "existing-session",
},
});
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "batch",
actions: [{ kind: "press", key: "Enter" }],
});
expect(response.status).toBe(501);
expect(response.body.code).toBe("ACT_EXISTING_SESSION_UNSUPPORTED");
expect(response.body.error).toContain("batch");
},
slowTimeoutMs,
);
it(
"returns ACT_TARGET_ID_MISMATCH for batched action targetId overrides",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "batch",
actions: [{ kind: "click", ref: "5", targetId: "other-tab" }],
});
expect(response.status).toBe(403);
expect(response.body.code).toBe("ACT_TARGET_ID_MISMATCH");
expect(response.body.error).toContain("batched action targetId must match request targetId");
},
slowTimeoutMs,
);
it(
"returns ACT_TARGET_ID_MISMATCH for top-level action targetId overrides",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "click",
ref: "5",
// Intentionally non-string: route-level target selection ignores this,
// while action normalization stringifies it.
targetId: 12345,
});
expect(response.status).toBe(403);
expect(response.body.code).toBe("ACT_TARGET_ID_MISMATCH");
expect(response.body.error).toContain("action targetId must match request targetId");
},
slowTimeoutMs,
);
it(
"returns ACT_SELECTOR_UNSUPPORTED for selector on unsupported action kinds",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "evaluate",
fn: "() => 1",
selector: "#submit",
});
expect(response.status).toBe(400);
expect(response.body.code).toBe("ACT_SELECTOR_UNSUPPORTED");
expect(response.body.error).toContain("'selector' is not supported");
},
slowTimeoutMs,
);
it(
"returns ACT_INVALID_REQUEST for malformed unsupported selector actions before selector gating",
async () => {
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "press",
selector: "#submit",
});
expect(response.status).toBe(400);
expect(response.body.code).toBe("ACT_INVALID_REQUEST");
expect(response.body.error).toContain("press requires key");
},
slowTimeoutMs,
);
it(
"returns ACT_EVALUATE_DISABLED when evaluate is blocked by config",
async () => {
setBrowserControlServerEvaluateEnabled(false);
const base = await startServerAndBase();
const response = await postActAndReadError(base, {
kind: "evaluate",
fn: "() => 1",
});
expect(response.status).toBe(403);
expect(response.body.code).toBe("ACT_EVALUATE_DISABLED");
expect(response.body.error).toContain("browser.evaluateEnabled=false");
},
slowTimeoutMs,
);
it("agent contract: snapshot endpoints", async () => {
const base = await startServerAndBase();
const realFetch = getBrowserTestFetch();
const snapAria = (await realFetch(`${base}/snapshot?format=aria&limit=1`).then((r) =>
r.json(),
)) as { ok: boolean; format?: string };
expect(snapAria.ok).toBe(true);
expect(snapAria.format).toBe("aria");
expect(cdpMocks.snapshotAria).toHaveBeenCalledWith({
wsUrl: "ws://127.0.0.1/devtools/page/abcd1234",
limit: 1,
});
const snapAi = (await realFetch(`${base}/snapshot?format=ai`).then((r) => r.json())) as {
ok: boolean;
format?: string;
};
expect(snapAi.ok).toBe(true);
expect(snapAi.format).toBe("ai");
expect(pwMocks.snapshotAiViaPlaywright).toHaveBeenCalledWith({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
maxChars: DEFAULT_AI_SNAPSHOT_MAX_CHARS,
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
});
const snapAiZero = (await realFetch(`${base}/snapshot?format=ai&maxChars=0`).then((r) =>
r.json(),
)) as { ok: boolean; format?: string };
expect(snapAiZero.ok).toBe(true);
expect(snapAiZero.format).toBe("ai");
const [lastCall] = pwMocks.snapshotAiViaPlaywright.mock.calls.at(-1) ?? [];
expect(lastCall).toEqual({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
});
});
it("agent contract: navigation + common act commands", async () => {
const base = await startServerAndBase();
const realFetch = getBrowserTestFetch();
const nav = await postJson<{ ok: boolean; targetId?: string }>(`${base}/navigate`, {
url: "https://example.com",
});
expect(nav.ok).toBe(true);
expect(typeof nav.targetId).toBe("string");
expect(pwMocks.navigateViaPlaywright).toHaveBeenCalledWith(
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
url: "https://example.com",
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
}),
);
const click = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "click",
ref: "1",
button: "left",
modifiers: ["Shift"],
});
expect(click.ok).toBe(true);
expect(pwMocks.clickViaPlaywright).toHaveBeenNthCalledWith(
1,
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
ref: "1",
button: "left",
modifiers: ["Shift"],
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
}),
);
const [clickArgs] = pwMocks.clickViaPlaywright.mock.calls[0] ?? [];
expect((clickArgs as { doubleClick?: boolean }).doubleClick).toBeUndefined();
const clickSelector = await realFetch(`${base}/act`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ kind: "click", selector: "button.save" }),
});
expect(clickSelector.status).toBe(200);
expect(((await clickSelector.json()) as { ok?: boolean }).ok).toBe(true);
expect(pwMocks.clickViaPlaywright).toHaveBeenNthCalledWith(
2,
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
selector: "button.save",
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
}),
);
const [clickSelectorArgs] = pwMocks.clickViaPlaywright.mock.calls[1] ?? [];
expect((clickSelectorArgs as { doubleClick?: boolean }).doubleClick).toBeUndefined();
const type = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "type",
ref: "1",
text: "",
});
expect(type.ok).toBe(true);
expect(pwMocks.typeViaPlaywright).toHaveBeenNthCalledWith(
1,
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
ref: "1",
text: "",
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
}),
);
const [typeArgs] = pwMocks.typeViaPlaywright.mock.calls[0] ?? [];
expect((typeArgs as { submit?: boolean }).submit).toBeUndefined();
expect((typeArgs as { slowly?: boolean }).slowly).toBeUndefined();
const press = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "press",
key: "Enter",
});
expect(press.ok).toBe(true);
expect(pwMocks.pressKeyViaPlaywright).toHaveBeenCalledWith(
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
key: "Enter",
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
}),
);
const [pressArgs] = pwMocks.pressKeyViaPlaywright.mock.calls[0] ?? [];
expect((pressArgs as { delayMs?: number }).delayMs).toBeUndefined();
const hover = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "hover",
ref: "2",
});
expect(hover.ok).toBe(true);
expect(pwMocks.hoverViaPlaywright).toHaveBeenCalledWith(
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
ref: "2",
}),
);
const [hoverArgs] = pwMocks.hoverViaPlaywright.mock.calls[0] ?? [];
expect((hoverArgs as { timeoutMs?: number }).timeoutMs).toBeUndefined();
const scroll = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "scrollIntoView",
ref: "2",
});
expect(scroll.ok).toBe(true);
expect(pwMocks.scrollIntoViewViaPlaywright).toHaveBeenCalledWith(
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
ref: "2",
}),
);
const [scrollArgs] = pwMocks.scrollIntoViewViaPlaywright.mock.calls[0] ?? [];
expect((scrollArgs as { timeoutMs?: number }).timeoutMs).toBeUndefined();
const drag = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "drag",
startRef: "3",
endRef: "4",
});
expect(drag.ok).toBe(true);
expect(pwMocks.dragViaPlaywright).toHaveBeenCalledWith(
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
startRef: "3",
endRef: "4",
}),
);
const [dragArgs] = pwMocks.dragViaPlaywright.mock.calls[0] ?? [];
expect((dragArgs as { timeoutMs?: number }).timeoutMs).toBeUndefined();
});
it("POST /tabs/open?profile=unknown returns 404", async () => {
await startBrowserControlServerFromConfig();
const base = getBrowserControlServerBaseUrl();
const realFetch = getBrowserTestFetch();
const result = await realFetch(`${base}/tabs/open?profile=unknown`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ url: "https://example.com" }),
});
expect(result.status).toBe(404);
const body = (await result.json()) as { error: string };
expect(body.error).toContain("not found");
});
it("POST /tabs/open returns 400 for invalid URLs", async () => {
setBrowserControlServerReachable(true);
await startBrowserControlServerFromConfig();
const base = getBrowserControlServerBaseUrl();
const realFetch = getBrowserTestFetch();
const result = await realFetch(`${base}/tabs/open`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ url: "not a url" }),
});
expect(result.status).toBe(400);
const body = (await result.json()) as { error: string };
expect(body.error).toContain("Invalid URL:");
});
});
describe("profile CRUD endpoints", () => {
beforeEach(async () => {
await resetBrowserControlServerTestContext();
vi.stubGlobal(
"fetch",
vi.fn(async (url: string) => {
const u = url;
if (u.includes("/json/list")) {
return makeResponse([]);
}
return makeResponse({}, { ok: false, status: 500, text: "unexpected" });
}),
);
});
afterEach(async () => {
await cleanupBrowserControlServerTestContext();
});
it("validates profile create/delete endpoints", async () => {
await startBrowserControlServerFromConfig();
const base = getBrowserControlServerBaseUrl();
const realFetch = getBrowserTestFetch();
const createMissingName = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({}),
});
expect(createMissingName.status).toBe(400);
const createMissingNameBody = (await createMissingName.json()) as { error: string };
expect(createMissingNameBody.error).toContain("name is required");
const createInvalidName = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "Invalid Name!" }),
});
expect(createInvalidName.status).toBe(400);
const createInvalidNameBody = (await createInvalidName.json()) as { error: string };
expect(createInvalidNameBody.error).toContain("invalid profile name");
const createDuplicate = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "openclaw" }),
});
expect(createDuplicate.status).toBe(409);
const createDuplicateBody = (await createDuplicate.json()) as { error: string };
expect(createDuplicateBody.error).toContain("already exists");
const createRemote = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "remote", cdpUrl: "http://10.0.0.42:9222" }),
});
expect(createRemote.status).toBe(200);
const createRemoteBody = (await createRemote.json()) as {
profile?: string;
cdpUrl?: string;
isRemote?: boolean;
};
expect(createRemoteBody.profile).toBe("remote");
expect(createRemoteBody.cdpUrl).toBe("http://10.0.0.42:9222");
expect(createRemoteBody.isRemote).toBe(true);
const createBadRemote = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "badremote", cdpUrl: "ftp://bad" }),
});
expect(createBadRemote.status).toBe(400);
const createBadRemoteBody = (await createBadRemote.json()) as { error: string };
expect(createBadRemoteBody.error).toContain("cdpUrl");
const createClawd = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "legacyclawd", driver: "clawd" }),
});
expect(createClawd.status).toBe(200);
const createClawdBody = (await createClawd.json()) as {
profile?: string;
transport?: string;
cdpPort?: number | null;
userDataDir?: string | null;
};
expect(createClawdBody.profile).toBe("legacyclawd");
expect(createClawdBody.transport).toBe("cdp");
expect(createClawdBody.cdpPort).toBeTypeOf("number");
expect(createClawdBody.userDataDir).toBeNull();
const explicitUserDataDir = "/tmp/openclaw-brave-profile";
await fs.promises.mkdir(explicitUserDataDir, { recursive: true });
const createExistingSession = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({
name: "brave-live",
driver: "existing-session",
userDataDir: explicitUserDataDir,
}),
});
expect(createExistingSession.status).toBe(200);
const createExistingSessionBody = (await createExistingSession.json()) as {
profile?: string;
transport?: string;
userDataDir?: string | null;
};
expect(createExistingSessionBody.profile).toBe("brave-live");
expect(createExistingSessionBody.transport).toBe("chrome-mcp");
expect(createExistingSessionBody.userDataDir).toBe(explicitUserDataDir);
const createBadExistingSession = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({
name: "bad-live",
userDataDir: explicitUserDataDir,
}),
});
expect(createBadExistingSession.status).toBe(400);
const createBadExistingSessionBody = (await createBadExistingSession.json()) as {
error: string;
};
expect(createBadExistingSessionBody.error).toContain("driver=existing-session is required");
const createLegacyDriver = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "legacy", driver: "extension" }),
});
expect(createLegacyDriver.status).toBe(400);
const createLegacyDriverBody = (await createLegacyDriver.json()) as { error: string };
expect(createLegacyDriverBody.error).toContain('unsupported profile driver "extension"');
const deleteMissing = await realFetch(`${base}/profiles/nonexistent`, {
method: "DELETE",
});
expect(deleteMissing.status).toBe(404);
const deleteMissingBody = (await deleteMissing.json()) as { error: string };
expect(deleteMissingBody.error).toContain("not found");
const deleteDefault = await realFetch(`${base}/profiles/openclaw`, {
method: "DELETE",
});
expect(deleteDefault.status).toBe(400);
const deleteDefaultBody = (await deleteDefault.json()) as { error: string };
expect(deleteDefaultBody.error).toContain("cannot delete the default profile");
const deleteInvalid = await realFetch(`${base}/profiles/Invalid-Name!`, {
method: "DELETE",
});
expect(deleteInvalid.status).toBe(400);
const deleteInvalidBody = (await deleteInvalid.json()) as { error: string };
expect(deleteInvalidBody.error).toContain("invalid profile name");
});
});

View File

@@ -1,217 +0,0 @@
import { describe, expect, it } from "vitest";
import { DEFAULT_AI_SNAPSHOT_MAX_CHARS } from "./constants.js";
import {
installAgentContractHooks,
postJson,
startServerAndBase,
} from "./server.agent-contract.test-harness.js";
import {
getBrowserControlServerTestState,
getCdpMocks,
getPwMocks,
} from "./server.control-server.test-harness.js";
import { getBrowserTestFetch } from "./test-fetch.js";
const state = getBrowserControlServerTestState();
const cdpMocks = getCdpMocks();
const pwMocks = getPwMocks();
describe("browser control server", () => {
installAgentContractHooks();
it("agent contract: snapshot endpoints", async () => {
const base = await startServerAndBase();
const realFetch = getBrowserTestFetch();
const snapAria = (await realFetch(`${base}/snapshot?format=aria&limit=1`).then((r) =>
r.json(),
)) as { ok: boolean; format?: string };
expect(snapAria.ok).toBe(true);
expect(snapAria.format).toBe("aria");
expect(cdpMocks.snapshotAria).toHaveBeenCalledWith({
wsUrl: "ws://127.0.0.1/devtools/page/abcd1234",
limit: 1,
});
const snapAi = (await realFetch(`${base}/snapshot?format=ai`).then((r) => r.json())) as {
ok: boolean;
format?: string;
};
expect(snapAi.ok).toBe(true);
expect(snapAi.format).toBe("ai");
expect(pwMocks.snapshotAiViaPlaywright).toHaveBeenCalledWith({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
maxChars: DEFAULT_AI_SNAPSHOT_MAX_CHARS,
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
});
const snapAiZero = (await realFetch(`${base}/snapshot?format=ai&maxChars=0`).then((r) =>
r.json(),
)) as { ok: boolean; format?: string };
expect(snapAiZero.ok).toBe(true);
expect(snapAiZero.format).toBe("ai");
const [lastCall] = pwMocks.snapshotAiViaPlaywright.mock.calls.at(-1) ?? [];
expect(lastCall).toEqual({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
});
});
it("agent contract: navigation + common act commands", async () => {
const base = await startServerAndBase();
const realFetch = getBrowserTestFetch();
const nav = await postJson<{ ok: boolean; targetId?: string }>(`${base}/navigate`, {
url: "https://example.com",
});
expect(nav.ok).toBe(true);
expect(typeof nav.targetId).toBe("string");
expect(pwMocks.navigateViaPlaywright).toHaveBeenCalledWith(
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
url: "https://example.com",
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
}),
);
const click = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "click",
ref: "1",
button: "left",
modifiers: ["Shift"],
});
expect(click.ok).toBe(true);
expect(pwMocks.clickViaPlaywright).toHaveBeenNthCalledWith(
1,
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
ref: "1",
button: "left",
modifiers: ["Shift"],
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
}),
);
const [clickArgs] = pwMocks.clickViaPlaywright.mock.calls[0] ?? [];
expect((clickArgs as { doubleClick?: boolean }).doubleClick).toBeUndefined();
const clickSelector = await realFetch(`${base}/act`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ kind: "click", selector: "button.save" }),
});
expect(clickSelector.status).toBe(200);
expect(((await clickSelector.json()) as { ok?: boolean }).ok).toBe(true);
expect(pwMocks.clickViaPlaywright).toHaveBeenNthCalledWith(
2,
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
selector: "button.save",
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
}),
);
const [clickSelectorArgs] = pwMocks.clickViaPlaywright.mock.calls[1] ?? [];
expect((clickSelectorArgs as { doubleClick?: boolean }).doubleClick).toBeUndefined();
const type = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "type",
ref: "1",
text: "",
});
expect(type.ok).toBe(true);
expect(pwMocks.typeViaPlaywright).toHaveBeenNthCalledWith(
1,
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
ref: "1",
text: "",
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
}),
);
const [typeArgs] = pwMocks.typeViaPlaywright.mock.calls[0] ?? [];
expect((typeArgs as { submit?: boolean }).submit).toBeUndefined();
expect((typeArgs as { slowly?: boolean }).slowly).toBeUndefined();
const press = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "press",
key: "Enter",
});
expect(press.ok).toBe(true);
expect(pwMocks.pressKeyViaPlaywright).toHaveBeenCalledWith(
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
key: "Enter",
ssrfPolicy: {
dangerouslyAllowPrivateNetwork: true,
},
}),
);
const [pressArgs] = pwMocks.pressKeyViaPlaywright.mock.calls[0] ?? [];
expect((pressArgs as { delayMs?: number }).delayMs).toBeUndefined();
const hover = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "hover",
ref: "2",
});
expect(hover.ok).toBe(true);
expect(pwMocks.hoverViaPlaywright).toHaveBeenCalledWith(
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
ref: "2",
}),
);
const [hoverArgs] = pwMocks.hoverViaPlaywright.mock.calls[0] ?? [];
expect((hoverArgs as { timeoutMs?: number }).timeoutMs).toBeUndefined();
const scroll = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "scrollIntoView",
ref: "2",
});
expect(scroll.ok).toBe(true);
expect(pwMocks.scrollIntoViewViaPlaywright).toHaveBeenCalledWith(
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
ref: "2",
}),
);
const [scrollArgs] = pwMocks.scrollIntoViewViaPlaywright.mock.calls[0] ?? [];
expect((scrollArgs as { timeoutMs?: number }).timeoutMs).toBeUndefined();
const drag = await postJson<{ ok: boolean }>(`${base}/act`, {
kind: "drag",
startRef: "3",
endRef: "4",
});
expect(drag.ok).toBe(true);
expect(pwMocks.dragViaPlaywright).toHaveBeenCalledWith(
expect.objectContaining({
cdpUrl: state.cdpBaseUrl,
targetId: "abcd1234",
startRef: "3",
endRef: "4",
}),
);
const [dragArgs] = pwMocks.dragViaPlaywright.mock.calls[0] ?? [];
expect((dragArgs as { timeoutMs?: number }).timeoutMs).toBeUndefined();
});
});

View File

@@ -348,7 +348,6 @@ async function loadBrowserServerModule(): Promise<BrowserServerModule> {
if (browserServerModule) {
return browserServerModule;
}
vi.resetModules();
browserServerModule = await import("../server.js");
return browserServerModule;
}
@@ -484,7 +483,6 @@ export async function startBrowserControlServerFromConfig() {
export async function stopBrowserControlServer(): Promise<void> {
const server = browserServerModule;
browserServerModule = null;
if (!server) {
return;
}

View File

@@ -1,207 +0,0 @@
import fs from "node:fs";
import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";
import {
cleanupBrowserControlServerTestContext,
getBrowserControlServerBaseUrl,
installBrowserControlServerHooks,
makeResponse,
resetBrowserControlServerTestContext,
setBrowserControlServerReachable,
startBrowserControlServerFromConfig,
} from "./server.control-server.test-harness.js";
import { getBrowserTestFetch } from "./test-fetch.js";
describe("browser control server", () => {
installBrowserControlServerHooks();
it("POST /tabs/open?profile=unknown returns 404", async () => {
await startBrowserControlServerFromConfig();
const base = getBrowserControlServerBaseUrl();
const realFetch = getBrowserTestFetch();
const result = await realFetch(`${base}/tabs/open?profile=unknown`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ url: "https://example.com" }),
});
expect(result.status).toBe(404);
const body = (await result.json()) as { error: string };
expect(body.error).toContain("not found");
});
it("POST /tabs/open returns 400 for invalid URLs", async () => {
setBrowserControlServerReachable(true);
await startBrowserControlServerFromConfig();
const base = getBrowserControlServerBaseUrl();
const realFetch = getBrowserTestFetch();
const result = await realFetch(`${base}/tabs/open`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ url: "not a url" }),
});
expect(result.status).toBe(400);
const body = (await result.json()) as { error: string };
expect(body.error).toContain("Invalid URL:");
});
});
describe("profile CRUD endpoints", () => {
beforeEach(async () => {
await resetBrowserControlServerTestContext();
vi.stubGlobal(
"fetch",
vi.fn(async (url: string) => {
const u = url;
if (u.includes("/json/list")) {
return makeResponse([]);
}
return makeResponse({}, { ok: false, status: 500, text: "unexpected" });
}),
);
});
afterEach(async () => {
await cleanupBrowserControlServerTestContext();
});
it("validates profile create/delete endpoints", async () => {
await startBrowserControlServerFromConfig();
const base = getBrowserControlServerBaseUrl();
const realFetch = getBrowserTestFetch();
const createMissingName = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({}),
});
expect(createMissingName.status).toBe(400);
const createMissingNameBody = (await createMissingName.json()) as { error: string };
expect(createMissingNameBody.error).toContain("name is required");
const createInvalidName = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "Invalid Name!" }),
});
expect(createInvalidName.status).toBe(400);
const createInvalidNameBody = (await createInvalidName.json()) as { error: string };
expect(createInvalidNameBody.error).toContain("invalid profile name");
const createDuplicate = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "openclaw" }),
});
expect(createDuplicate.status).toBe(409);
const createDuplicateBody = (await createDuplicate.json()) as { error: string };
expect(createDuplicateBody.error).toContain("already exists");
const createRemote = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "remote", cdpUrl: "http://10.0.0.42:9222" }),
});
expect(createRemote.status).toBe(200);
const createRemoteBody = (await createRemote.json()) as {
profile?: string;
cdpUrl?: string;
isRemote?: boolean;
};
expect(createRemoteBody.profile).toBe("remote");
expect(createRemoteBody.cdpUrl).toBe("http://10.0.0.42:9222");
expect(createRemoteBody.isRemote).toBe(true);
const createBadRemote = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "badremote", cdpUrl: "ftp://bad" }),
});
expect(createBadRemote.status).toBe(400);
const createBadRemoteBody = (await createBadRemote.json()) as { error: string };
expect(createBadRemoteBody.error).toContain("cdpUrl");
const createClawd = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "legacyclawd", driver: "clawd" }),
});
expect(createClawd.status).toBe(200);
const createClawdBody = (await createClawd.json()) as {
profile?: string;
transport?: string;
cdpPort?: number | null;
userDataDir?: string | null;
};
expect(createClawdBody.profile).toBe("legacyclawd");
expect(createClawdBody.transport).toBe("cdp");
expect(createClawdBody.cdpPort).toBeTypeOf("number");
expect(createClawdBody.userDataDir).toBeNull();
const explicitUserDataDir = "/tmp/openclaw-brave-profile";
await fs.promises.mkdir(explicitUserDataDir, { recursive: true });
const createExistingSession = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({
name: "brave-live",
driver: "existing-session",
userDataDir: explicitUserDataDir,
}),
});
expect(createExistingSession.status).toBe(200);
const createExistingSessionBody = (await createExistingSession.json()) as {
profile?: string;
transport?: string;
userDataDir?: string | null;
};
expect(createExistingSessionBody.profile).toBe("brave-live");
expect(createExistingSessionBody.transport).toBe("chrome-mcp");
expect(createExistingSessionBody.userDataDir).toBe(explicitUserDataDir);
const createBadExistingSession = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({
name: "bad-live",
userDataDir: explicitUserDataDir,
}),
});
expect(createBadExistingSession.status).toBe(400);
const createBadExistingSessionBody = (await createBadExistingSession.json()) as {
error: string;
};
expect(createBadExistingSessionBody.error).toContain("driver=existing-session is required");
const createLegacyDriver = await realFetch(`${base}/profiles/create`, {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ name: "legacy", driver: "extension" }),
});
expect(createLegacyDriver.status).toBe(400);
const createLegacyDriverBody = (await createLegacyDriver.json()) as { error: string };
expect(createLegacyDriverBody.error).toContain('unsupported profile driver "extension"');
const deleteMissing = await realFetch(`${base}/profiles/nonexistent`, {
method: "DELETE",
});
expect(deleteMissing.status).toBe(404);
const deleteMissingBody = (await deleteMissing.json()) as { error: string };
expect(deleteMissingBody.error).toContain("not found");
const deleteDefault = await realFetch(`${base}/profiles/openclaw`, {
method: "DELETE",
});
expect(deleteDefault.status).toBe(400);
const deleteDefaultBody = (await deleteDefault.json()) as { error: string };
expect(deleteDefaultBody.error).toContain("cannot delete the default profile");
const deleteInvalid = await realFetch(`${base}/profiles/Invalid-Name!`, {
method: "DELETE",
});
expect(deleteInvalid.status).toBe(400);
const deleteInvalidBody = (await deleteInvalid.json()) as { error: string };
expect(deleteInvalidBody.error).toContain("invalid profile name");
});
});

View File

@@ -199,6 +199,39 @@ describe("CodexAppServerClient", () => {
expect(process.kill).toHaveBeenCalledWith("SIGKILL");
expect(process.unref).toHaveBeenCalledTimes(1);
});
it("handles stdin write errors without crashing the process", async () => {
const harness = createClientHarness();
clients.push(harness.client);
// Start a pending request so we can verify it gets properly rejected.
const pending = harness.client.request("test/method");
// Simulate the child process closing its pipe — a write to the now-dead
// stdin emits an asynchronous EPIPE error on the stream.
harness.process.stdin.destroy(Object.assign(new Error("write EPIPE"), { code: "EPIPE" }));
// The pending request must be rejected with the pipe error rather than
// an unhandled exception tearing down the gateway.
await expect(pending).rejects.toThrow("write EPIPE");
// Subsequent requests are rejected immediately (client is closed).
await expect(harness.client.request("another/method")).rejects.toThrow(
"codex app-server client is closed",
);
});
it("does not write to stdin after the child process exits", async () => {
const harness = createClientHarness();
clients.push(harness.client);
// Simulate the child process exiting.
harness.process.emit("exit", 1, null);
// A notification after exit must not attempt a write.
harness.client.notify("late/event", { data: "ignored" });
expect(harness.writes).toHaveLength(0);
});
it("reads the Codex version from the app-server user agent", () => {
expect(readCodexVersionFromUserAgent("Codex Desktop/0.118.0")).toBe("0.118.0");
expect(readCodexVersionFromUserAgent("openclaw/0.118.0 (macOS; test)")).toBe("0.118.0");

View File

@@ -74,6 +74,13 @@ export class CodexAppServerClient {
),
);
});
// Guard against unhandled EPIPE / write-after-close errors on the stdin
// stream. When the child process terminates abruptly the pipe can break
// before the "exit" event fires, so a pending writeMessage() produces an
// asynchronous error on stdin that would otherwise crash the gateway.
child.stdin.on?.("error", (error) =>
this.closeWithError(error instanceof Error ? error : new Error(String(error))),
);
}
static start(options?: Partial<CodexAppServerStartOptions>): CodexAppServerClient {
@@ -212,6 +219,9 @@ export class CodexAppServerClient {
}
private writeMessage(message: RpcRequest | RpcResponse): void {
if (this.closed) {
return;
}
this.child.stdin.write(`${JSON.stringify(message)}\n`);
}
@@ -300,7 +310,9 @@ export class CodexAppServerClient {
return;
}
this.closed = true;
this.lines.close();
this.rejectPendingRequests(error);
closeCodexAppServerTransport(this.child);
}
private rejectPendingRequests(error: Error): void {

View File

@@ -4,6 +4,7 @@ export type CodexAppServerTransport = {
end?: () => unknown;
destroy?: () => unknown;
unref?: () => unknown;
on?: (event: "error", listener: (error: Error) => void) => unknown;
};
stdout: NodeJS.ReadableStream & {
destroy?: () => unknown;

View File

@@ -42,7 +42,7 @@ describeLive("comfy live", () => {
beforeAll(async () => {
cfg = withPluginsEnabled(loadConfig());
agentDir = resolveOpenClawAgentDir();
await plugin.register(
plugin.register(
createTestPluginApi({
config: cfg as never,
registerImageGenerationProvider(provider) {

View File

@@ -92,7 +92,7 @@ function registerPairCommand(params?: {
pluginConfig?: Record<string, unknown>;
}): OpenClawPluginCommandDefinition {
let command: OpenClawPluginCommandDefinition | undefined;
void registerDevicePair.register(
registerDevicePair.register(
createApi({
...params,
registerCommand: (nextCommand) => {

View File

@@ -1,4 +1,5 @@
export { createThreadBindingManager } from "./src/monitor/thread-bindings.manager.js";
export { __testing as discordThreadBindingTesting } from "./src/monitor/thread-bindings.manager.js";
export {
listDiscordDirectoryGroupsFromConfig,
listDiscordDirectoryPeersFromConfig,

View File

@@ -1,4 +1,4 @@
import { beforeAll, beforeEach, describe, expect, it, vi } from "vitest";
import { afterEach, beforeAll, beforeEach, describe, expect, it, vi } from "vitest";
const {
GatewayIntents,
@@ -221,6 +221,9 @@ describe("createDiscordGatewayPlugin", () => {
}
beforeEach(() => {
vi.unstubAllEnvs();
vi.stubEnv("OPENCLAW_DEBUG_PROXY_ENABLED", "");
vi.stubEnv("OPENCLAW_DEBUG_PROXY_URL", "");
vi.stubGlobal("fetch", globalFetchMock);
vi.useRealTimers();
baseRegisterClientSpy.mockClear();
@@ -236,6 +239,11 @@ describe("createDiscordGatewayPlugin", () => {
resetLastAgent();
});
afterEach(() => {
vi.useRealTimers();
vi.unstubAllEnvs();
});
it("uses safe gateway metadata lookup without proxy", async () => {
const runtime = createRuntime();
const plugin = createDiscordGatewayPlugin({

View File

@@ -0,0 +1,21 @@
import { describe, expect, it } from "vitest";
import { createDiscordPluginBase } from "./shared.js";
describe("createDiscordPluginBase", () => {
it("owns Discord native command name overrides", () => {
const plugin = createDiscordPluginBase({ setup: {} as never });
expect(
plugin.commands?.resolveNativeCommandName?.({
commandKey: "tts",
defaultName: "tts",
}),
).toBe("voice");
expect(
plugin.commands?.resolveNativeCommandName?.({
commandKey: "status",
defaultName: "status",
}),
).toBe("status");
});
});

View File

@@ -1,4 +1,5 @@
import { describe, expect, it } from "vitest";
import { createExaWebSearchProvider as createContractExaWebSearchProvider } from "../web-search-contract-api.js";
import { __testing, createExaWebSearchProvider } from "./exa-web-search-provider.js";
describe("exa web search provider", () => {
@@ -15,6 +16,31 @@ describe("exa web search provider", () => {
expect(applied.plugins?.entries?.exa?.enabled).toBe(true);
});
it("keeps the lightweight contract surface aligned with provider metadata", () => {
const provider = createExaWebSearchProvider();
const contractProvider = createContractExaWebSearchProvider();
if (!contractProvider.applySelectionConfig) {
throw new Error("Expected contract applySelectionConfig to be defined");
}
const applied = contractProvider.applySelectionConfig({});
expect(contractProvider).toMatchObject({
id: provider.id,
label: provider.label,
hint: provider.hint,
onboardingScopes: provider.onboardingScopes,
credentialLabel: provider.credentialLabel,
envVars: provider.envVars,
placeholder: provider.placeholder,
signupUrl: provider.signupUrl,
docsUrl: provider.docsUrl,
autoDetectOrder: provider.autoDetectOrder,
credentialPath: provider.credentialPath,
});
expect(contractProvider.createTool({ config: {}, searchConfig: {} })).toBeNull();
expect(applied.plugins?.entries?.exa?.enabled).toBe(true);
});
it("prefers scoped configured api keys over environment fallbacks", () => {
expect(__testing.resolveExaApiKey({ apiKey: "exa-secret" })).toBe("exa-secret");
});

View File

@@ -0,0 +1,29 @@
import {
createWebSearchProviderContractFields,
type WebSearchProviderPlugin,
} from "openclaw/plugin-sdk/provider-web-search-contract";
export function createExaWebSearchProvider(): WebSearchProviderPlugin {
const credentialPath = "plugins.entries.exa.config.webSearch.apiKey";
return {
id: "exa",
label: "Exa Search",
hint: "Neural + keyword search with date filters and content extraction",
onboardingScopes: ["text-inference"],
credentialLabel: "Exa API key",
envVars: ["EXA_API_KEY"],
placeholder: "exa-...",
signupUrl: "https://exa.ai/",
docsUrl: "https://docs.openclaw.ai/tools/web",
autoDetectOrder: 65,
credentialPath,
...createWebSearchProviderContractFields({
credentialPath,
searchCredential: { type: "scoped", scopeId: "exa" },
configuredCredential: { pluginId: "exa" },
selectionPluginId: "exa",
}),
createTool: () => null,
};
}

View File

@@ -25,9 +25,14 @@ vi.mock("./bot.js", () => ({
handleFeishuMessage: vi.fn(),
}));
const createFeishuClientMock = vi.hoisted(() => vi.fn());
const sendCardFeishuMock = vi.hoisted(() => vi.fn());
const sendMessageFeishuMock = vi.hoisted(() => vi.fn());
vi.mock("./client.js", () => ({
createFeishuClient: createFeishuClientMock,
}));
vi.mock("./send.js", () => ({
sendCardFeishu: sendCardFeishuMock,
sendMessageFeishu: sendMessageFeishuMock,
@@ -89,6 +94,13 @@ describe("Feishu Card Action Handler", () => {
beforeEach(() => {
vi.clearAllMocks();
createFeishuClientMock.mockReset().mockReturnValue({
im: {
chat: {
get: vi.fn().mockResolvedValue({ code: 0, data: { chat_type: "group" } }),
},
},
});
vi.mocked(handleFeishuMessage)
.mockReset()
.mockResolvedValue(undefined as never);
@@ -354,6 +366,142 @@ describe("Feishu Card Action Handler", () => {
);
});
it("resolves DM chat type from the Feishu chat API when card context omits it", async () => {
createFeishuClientMock.mockReturnValueOnce({
im: {
chat: {
get: vi.fn().mockResolvedValue({ code: 0, data: { chat_type: "p2p" } }),
},
},
});
const event = createCardActionEvent({
token: "tok9b",
chatId: "oc_dm_chat_123",
actionValue: { text: "/help" },
});
await handleFeishuCardAction({ cfg, event, runtime });
expect(handleFeishuMessage).toHaveBeenCalledWith(
expect.objectContaining({
event: expect.objectContaining({
message: expect.objectContaining({
chat_id: "oc_dm_chat_123",
chat_type: "p2p",
}),
}),
}),
);
expect(createFeishuClientMock).toHaveBeenCalledTimes(1);
});
it("uses resolved DM chat type when building approval cards without stored context", async () => {
createFeishuClientMock.mockReturnValueOnce({
im: {
chat: {
get: vi.fn().mockResolvedValue({ code: 0, data: { chat_mode: "p2p" } }),
},
},
});
const event = createCardActionEvent({
token: "tok9c",
chatId: "oc_dm_chat_234",
actionValue: createFeishuCardInteractionEnvelope({
k: "meta",
a: FEISHU_APPROVAL_REQUEST_ACTION,
m: {
command: "/new",
prompt: "Start a fresh session?",
},
c: {
u: "u123",
h: "oc_dm_chat_234",
e: Date.now() + 60_000,
},
}),
});
await handleFeishuCardAction({ cfg, event, runtime, accountId: "main" });
expect(sendCardFeishuMock).toHaveBeenCalledWith(
expect.objectContaining({
card: expect.objectContaining({
body: expect.objectContaining({
elements: expect.arrayContaining([
expect.objectContaining({
tag: "action",
actions: expect.arrayContaining([
expect.objectContaining({
value: expect.objectContaining({
c: expect.objectContaining({
t: "p2p",
}),
}),
}),
]),
}),
]),
}),
}),
}),
);
expect(createFeishuClientMock).toHaveBeenCalledTimes(1);
});
it("falls back to p2p when Feishu chat API returns an error", async () => {
createFeishuClientMock.mockReturnValueOnce({
im: {
chat: {
get: vi.fn().mockResolvedValue({ code: 99, msg: "not found" }),
},
},
});
const event = createCardActionEvent({
token: "tok9d",
chatId: "oc_unknown_chat_456",
actionValue: { text: "/help" },
});
await handleFeishuCardAction({ cfg, event, runtime });
expect(handleFeishuMessage).toHaveBeenCalledWith(
expect.objectContaining({
event: expect.objectContaining({
message: expect.objectContaining({
chat_type: "p2p",
}),
}),
}),
);
});
it("falls back to p2p when Feishu chat API throws", async () => {
createFeishuClientMock.mockReturnValueOnce({
im: {
chat: {
get: vi.fn().mockRejectedValue(new Error("network failure")),
},
},
});
const event = createCardActionEvent({
token: "tok9e",
chatId: "oc_broken_chat_789",
actionValue: { text: "/help" },
});
await handleFeishuCardAction({ cfg, event, runtime });
expect(handleFeishuMessage).toHaveBeenCalledWith(
expect.objectContaining({
event: expect.objectContaining({
message: expect.objectContaining({
chat_type: "p2p",
}),
}),
}),
);
});
it("drops duplicate structured callback tokens", async () => {
const event = createStructuredQuickActionEvent({
token: "tok10",

View File

@@ -2,6 +2,7 @@ import type { ClawdbotConfig, RuntimeEnv } from "../runtime-api.js";
import { resolveFeishuRuntimeAccount } from "./accounts.js";
import { handleFeishuMessage, type FeishuMessageEvent } from "./bot.js";
import { decodeFeishuCardAction, buildFeishuCardActionTextFallback } from "./card-interaction.js";
import { createFeishuClient } from "./client.js";
import {
createApprovalCard,
FEISHU_APPROVAL_CANCEL_ACTION,
@@ -104,7 +105,7 @@ function releaseFeishuCardActionToken(params: { token: string; accountId: string
function buildSyntheticMessageEvent(
event: FeishuCardActionEvent,
content: string,
chatType?: "p2p" | "group",
chatType: "p2p" | "group",
): FeishuMessageEvent {
return {
sender: {
@@ -117,7 +118,7 @@ function buildSyntheticMessageEvent(
message: {
message_id: `card-action-${event.token}`,
chat_id: event.context.chat_id || event.operator.open_id,
chat_type: chatType ?? (event.context.chat_id ? "group" : "p2p"),
chat_type: chatType,
message_type: "text",
content: JSON.stringify({ text: content }),
},
@@ -136,20 +137,124 @@ async function dispatchSyntheticCommand(params: {
cfg: ClawdbotConfig;
event: FeishuCardActionEvent;
command: string;
account: ReturnType<typeof resolveFeishuRuntimeAccount>;
botOpenId?: string;
runtime?: RuntimeEnv;
accountId?: string;
chatType?: "p2p" | "group";
}): Promise<void> {
const resolvedChatType = await resolveCardActionChatType({
event: params.event,
account: params.account,
chatType: params.chatType,
log: params.runtime?.log ?? console.log,
});
await handleFeishuMessage({
cfg: params.cfg,
event: buildSyntheticMessageEvent(params.event, params.command, params.chatType),
event: buildSyntheticMessageEvent(params.event, params.command, resolvedChatType),
botOpenId: params.botOpenId,
runtime: params.runtime,
accountId: params.accountId,
});
}
// Feishu's im.chat.get returns two fields:
// chat_mode: conversation type — "p2p" | "group" | "topic"
// chat_type: privacy classification — "private" | "public"
// We check chat_mode first because it directly indicates conversation type.
// "private" maps to "p2p" as the safe-failure direction (restrictive DM
// policy) — a private group chat misclassified as p2p is safer than the
// reverse. "topic" and "public" are treated as group semantics.
function normalizeResolvedCardActionChatType(value: unknown): "p2p" | "group" | undefined {
if (value === "group" || value === "topic" || value === "public") {
return "group";
}
if (value === "p2p" || value === "private") {
return "p2p";
}
return undefined;
}
const resolvedChatTypeCache = new Map<string, { value: "p2p" | "group"; expiresAt: number }>();
const CHAT_TYPE_CACHE_TTL_MS = 30 * 60_000;
const CHAT_TYPE_CACHE_MAX_SIZE = 5_000;
function pruneChatTypeCache(now: number): void {
for (const [key, entry] of resolvedChatTypeCache.entries()) {
if (entry.expiresAt <= now) {
resolvedChatTypeCache.delete(key);
}
}
if (resolvedChatTypeCache.size > CHAT_TYPE_CACHE_MAX_SIZE) {
const excess = resolvedChatTypeCache.size - CHAT_TYPE_CACHE_MAX_SIZE;
const iter = resolvedChatTypeCache.keys();
for (let i = 0; i < excess; i++) {
const key = iter.next().value;
if (key !== undefined) {
resolvedChatTypeCache.delete(key);
}
}
}
}
function sanitizeLogValue(v: string): string {
return v.replace(/[\r\n]/g, " ").slice(0, 500);
}
async function resolveCardActionChatType(params: {
event: FeishuCardActionEvent;
account: ReturnType<typeof resolveFeishuRuntimeAccount>;
chatType?: "p2p" | "group";
log: (message: string) => void;
}): Promise<"p2p" | "group"> {
const explicitChatType = normalizeResolvedCardActionChatType(params.chatType);
if (explicitChatType) {
return explicitChatType;
}
const chatId = params.event.context.chat_id?.trim();
if (!chatId) {
return "p2p";
}
const cacheKey = `${params.account.accountId}:${chatId}`;
const now = Date.now();
pruneChatTypeCache(now);
const cached = resolvedChatTypeCache.get(cacheKey);
if (cached) {
return cached.value;
}
try {
const response = (await createFeishuClient(params.account).im.chat.get({
path: { chat_id: chatId },
})) as { code?: number; msg?: string; data?: { chat_type?: unknown; chat_mode?: unknown } };
if (response.code === 0) {
const resolvedChatType =
normalizeResolvedCardActionChatType(response.data?.chat_mode) ??
normalizeResolvedCardActionChatType(response.data?.chat_type);
if (resolvedChatType) {
resolvedChatTypeCache.set(cacheKey, { value: resolvedChatType, expiresAt: now + CHAT_TYPE_CACHE_TTL_MS });
return resolvedChatType;
}
params.log(
`feishu[${params.account.accountId}]: card action missing chat type for chat; defaulting to p2p`,
);
} else {
params.log(
`feishu[${params.account.accountId}]: failed to resolve chat type: ${sanitizeLogValue(response.msg ?? "unknown error")}; defaulting to p2p`,
);
}
} catch (err) {
const message = err instanceof Error ? err.message : "unknown";
params.log(
`feishu[${params.account.accountId}]: failed to resolve chat type: ${sanitizeLogValue(message)}; defaulting to p2p`,
);
}
return "p2p";
}
async function sendInvalidInteractionNotice(params: {
cfg: ClawdbotConfig;
event: FeishuCardActionEvent;
@@ -246,7 +351,12 @@ export async function handleFeishuCardAction(params: {
prompt,
sessionKey: envelope.c?.s,
expiresAt: Date.now() + FEISHU_APPROVAL_CARD_TTL_MS,
chatType: envelope.c?.t ?? (event.context.chat_id ? "group" : "p2p"),
chatType: await resolveCardActionChatType({
event,
account,
chatType: envelope.c?.t,
log,
}),
confirmLabel: command === "/reset" ? "Reset" : "Confirm",
}),
accountId,
@@ -282,10 +392,11 @@ export async function handleFeishuCardAction(params: {
cfg,
event,
command,
account,
botOpenId: params.botOpenId,
runtime,
accountId,
chatType: envelope.c?.t ?? (event.context.chat_id ? "group" : "p2p"),
chatType: envelope.c?.t,
});
completeFeishuCardActionToken({ token: event.token, accountId: account.accountId });
return;
@@ -311,6 +422,7 @@ export async function handleFeishuCardAction(params: {
cfg,
event,
command: content,
account,
botOpenId: params.botOpenId,
runtime,
accountId,

View File

@@ -4,7 +4,6 @@ const resolveFirstGithubTokenMock = vi.hoisted(() => vi.fn());
const resolveCopilotApiTokenMock = vi.hoisted(() => vi.fn());
const resolveConfiguredSecretInputStringMock = vi.hoisted(() => vi.fn());
const fetchWithSsrFGuardMock = vi.hoisted(() => vi.fn());
const createGitHubCopilotEmbeddingProviderMock = vi.hoisted(() => vi.fn());
vi.mock("./auth.js", () => ({
resolveFirstGithubToken: resolveFirstGithubTokenMock,
@@ -19,10 +18,6 @@ vi.mock("openclaw/plugin-sdk/github-copilot-token", () => ({
resolveCopilotApiToken: resolveCopilotApiTokenMock,
}));
vi.mock("openclaw/plugin-sdk/memory-core-host-engine-embeddings", () => ({
createGitHubCopilotEmbeddingProvider: createGitHubCopilotEmbeddingProviderMock,
}));
vi.mock("openclaw/plugin-sdk/ssrf-runtime", () => ({
fetchWithSsrFGuard: fetchWithSsrFGuardMock,
}));
@@ -73,15 +68,6 @@ describe("githubCopilotMemoryEmbeddingProviderAdapter", () => {
source: "test",
baseUrl: TEST_BASE_URL,
});
createGitHubCopilotEmbeddingProviderMock.mockImplementation(async (client) => ({
provider: {
id: "github-copilot",
model: client.model,
embedQuery: async () => [0.1, 0.2, 0.3],
embedBatch: async (texts: string[]) => texts.map(() => [0.1, 0.2, 0.3]),
},
client,
}));
});
afterEach(() => {
@@ -89,7 +75,6 @@ describe("githubCopilotMemoryEmbeddingProviderAdapter", () => {
resolveConfiguredSecretInputStringMock.mockReset();
resolveFirstGithubTokenMock.mockReset();
resolveCopilotApiTokenMock.mockReset();
createGitHubCopilotEmbeddingProviderMock.mockReset();
fetchWithSsrFGuardMock.mockReset();
});
@@ -113,12 +98,8 @@ describe("githubCopilotMemoryEmbeddingProviderAdapter", () => {
const result = await githubCopilotMemoryEmbeddingProviderAdapter.create(defaultCreateOptions());
expect(result.provider?.model).toBe("text-embedding-3-small");
expect(createGitHubCopilotEmbeddingProviderMock).toHaveBeenCalledWith(
expect.objectContaining({
baseUrl: TEST_BASE_URL,
githubToken: "gh_test_token_123",
model: "text-embedding-3-small",
}),
expect(resolveCopilotApiTokenMock).toHaveBeenCalledWith(
expect.objectContaining({ githubToken: "gh_test_token_123" }),
);
});
@@ -217,14 +198,12 @@ describe("githubCopilotMemoryEmbeddingProviderAdapter", () => {
} as never);
expect(resolveFirstGithubTokenMock).toHaveBeenCalled();
expect(createGitHubCopilotEmbeddingProviderMock).toHaveBeenCalledWith({
baseUrl: "https://proxy.example/v1",
env: process.env,
fetchImpl: fetch,
githubToken: "gh_remote_token",
headers: { "X-Proxy-Token": "proxy" },
model: "text-embedding-3-small",
});
expect(resolveCopilotApiTokenMock).toHaveBeenCalledWith(
expect.objectContaining({
env: process.env,
githubToken: "gh_remote_token",
}),
);
const discoveryCall = fetchWithSsrFGuardMock.mock.calls[0]?.[0] as {
init: { headers: Record<string, string> };

View File

@@ -4,7 +4,10 @@ import {
resolveCopilotApiToken,
} from "openclaw/plugin-sdk/github-copilot-token";
import {
createGitHubCopilotEmbeddingProvider,
buildRemoteBaseUrlPolicy,
sanitizeAndNormalizeEmbedding,
withRemoteHttpResponse,
type MemoryEmbeddingProvider,
type MemoryEmbeddingProviderAdapter,
} from "openclaw/plugin-sdk/memory-core-host-engine-embeddings";
import { fetchWithSsrFGuard, type SsrFPolicy } from "openclaw/plugin-sdk/ssrf-runtime";
@@ -44,6 +47,15 @@ type CopilotModelEntry = {
supported_endpoints?: unknown;
};
type GitHubCopilotEmbeddingClient = {
githubToken: string;
model: string;
baseUrl?: string;
headers?: Record<string, string>;
env?: NodeJS.ProcessEnv;
fetchImpl?: typeof fetch;
};
function isCopilotSetupError(err: unknown): boolean {
if (!(err instanceof Error)) {
return false;
@@ -147,9 +159,126 @@ function pickBestModel(available: string[], userModel?: string): string {
throw new Error("No embedding models available from GitHub Copilot");
}
function parseGitHubCopilotEmbeddingPayload(payload: unknown, expectedCount: number): number[][] {
if (!payload || typeof payload !== "object") {
throw new Error("GitHub Copilot embeddings response missing data[]");
}
const data = (payload as { data?: unknown }).data;
if (!Array.isArray(data)) {
throw new Error("GitHub Copilot embeddings response missing data[]");
}
const vectors = Array.from<number[] | undefined>({ length: expectedCount });
for (const entry of data) {
if (!entry || typeof entry !== "object") {
throw new Error("GitHub Copilot embeddings response contains an invalid entry");
}
const indexValue = (entry as { index?: unknown }).index;
const embedding = (entry as { embedding?: unknown }).embedding;
const index = typeof indexValue === "number" ? indexValue : Number.NaN;
if (!Number.isInteger(index)) {
throw new Error("GitHub Copilot embeddings response contains an invalid index");
}
if (index < 0 || index >= expectedCount) {
throw new Error("GitHub Copilot embeddings response contains an out-of-range index");
}
if (vectors[index] !== undefined) {
throw new Error("GitHub Copilot embeddings response contains duplicate indexes");
}
if (!Array.isArray(embedding) || !embedding.every((value) => typeof value === "number")) {
throw new Error("GitHub Copilot embeddings response contains an invalid embedding");
}
vectors[index] = sanitizeAndNormalizeEmbedding(embedding);
}
for (let index = 0; index < expectedCount; index += 1) {
if (vectors[index] === undefined) {
throw new Error("GitHub Copilot embeddings response missing vectors for some inputs");
}
}
return vectors as number[][];
}
async function resolveGitHubCopilotEmbeddingSession(client: GitHubCopilotEmbeddingClient): Promise<{
baseUrl: string;
headers: Record<string, string>;
}> {
const token = await resolveCopilotApiToken({
githubToken: client.githubToken,
env: client.env,
fetchImpl: client.fetchImpl,
});
const baseUrl = client.baseUrl?.trim() || token.baseUrl || DEFAULT_COPILOT_API_BASE_URL;
return {
baseUrl,
headers: {
...COPILOT_HEADERS_STATIC,
...client.headers,
Authorization: `Bearer ${token.token}`,
},
};
}
async function createGitHubCopilotEmbeddingProvider(
client: GitHubCopilotEmbeddingClient,
): Promise<{ provider: MemoryEmbeddingProvider; client: GitHubCopilotEmbeddingClient }> {
const initialSession = await resolveGitHubCopilotEmbeddingSession(client);
const embed = async (input: string[]): Promise<number[][]> => {
if (input.length === 0) {
return [];
}
const session = await resolveGitHubCopilotEmbeddingSession(client);
const url = `${session.baseUrl.replace(/\/$/, "")}/embeddings`;
return await withRemoteHttpResponse({
url,
fetchImpl: client.fetchImpl,
ssrfPolicy: buildRemoteBaseUrlPolicy(session.baseUrl),
init: {
method: "POST",
headers: session.headers,
body: JSON.stringify({ model: client.model, input }),
},
onResponse: async (response) => {
if (!response.ok) {
throw new Error(
`GitHub Copilot embeddings HTTP ${response.status}: ${await response.text()}`,
);
}
let payload: unknown;
try {
payload = await response.json();
} catch {
throw new Error("GitHub Copilot embeddings returned invalid JSON");
}
return parseGitHubCopilotEmbeddingPayload(payload, input.length);
},
});
};
return {
provider: {
id: COPILOT_EMBEDDING_PROVIDER_ID,
model: client.model,
embedQuery: async (text) => {
const [vector] = await embed([text]);
return vector ?? [];
},
embedBatch: embed,
},
client: {
...client,
baseUrl: initialSession.baseUrl,
},
};
}
export const githubCopilotMemoryEmbeddingProviderAdapter: MemoryEmbeddingProviderAdapter = {
id: COPILOT_EMBEDDING_PROVIDER_ID,
transport: "remote",
authProviderId: COPILOT_EMBEDDING_PROVIDER_ID,
autoSelectPriority: 15,
allowExplicitWhenConfiguredAuto: true,
shouldContinueAutoSelection: (err: unknown) => isCopilotSetupError(err),

View File

@@ -2,11 +2,8 @@ import {
resolveProviderHttpRequestConfig,
type ProviderRequestTransportOverrides,
} from "openclaw/plugin-sdk/provider-http";
import {
applyAgentDefaultModelPrimary,
type OpenClawConfig,
} from "openclaw/plugin-sdk/provider-onboard";
import { parseGoogleOauthApiKey } from "./oauth-token-shared.js";
export { applyGoogleGeminiModelDefault, GOOGLE_GEMINI_DEFAULT_MODEL } from "./onboard.js";
import {
DEFAULT_GOOGLE_API_BASE_URL,
normalizeGoogleApiBaseUrl,
@@ -24,6 +21,8 @@ export {
shouldNormalizeGoogleGenerativeAiProviderConfig,
shouldNormalizeGoogleProviderConfig,
} from "./provider-policy.js";
export { buildGoogleGeminiCliProvider } from "./gemini-cli-provider.js";
export { buildGoogleProvider } from "./provider-registration.js";
export function parseGeminiAuth(apiKey: string): { headers: Record<string, string> } {
const parsed = apiKey.startsWith("{") ? parseGoogleOauthApiKey(apiKey) : null;
@@ -88,27 +87,3 @@ export function resolveGoogleGenerativeAiHttpRequestConfig(params: {
transport: params.transport,
});
}
export const GOOGLE_GEMINI_DEFAULT_MODEL = "google/gemini-3.1-pro-preview";
export function applyGoogleGeminiModelDefault(cfg: OpenClawConfig): {
next: OpenClawConfig;
changed: boolean;
} {
const current = cfg.agents?.defaults?.model as unknown;
const currentPrimary =
typeof current === "string"
? current.trim() || undefined
: current &&
typeof current === "object" &&
typeof (current as { primary?: unknown }).primary === "string"
? ((current as { primary: string }).primary || "").trim() || undefined
: undefined;
if (currentPrimary === GOOGLE_GEMINI_DEFAULT_MODEL) {
return { next: cfg, changed: false };
}
return {
next: applyAgentDefaultModelPrimary(cfg, GOOGLE_GEMINI_DEFAULT_MODEL),
changed: true,
};
}

View File

@@ -1,14 +1,15 @@
import crypto from "node:crypto";
import {
buildEmbeddingBatchGroupOptions,
runEmbeddingBatchGroups,
type EmbeddingBatchExecutionParams,
} from "./batch-runner.js";
import { buildBatchHeaders, normalizeBatchBaseUrl } from "./batch-utils.js";
import { sanitizeAndNormalizeEmbedding } from "./embedding-vectors.js";
import { debugEmbeddingsLog } from "./embeddings-debug.js";
import type { GeminiEmbeddingClient, GeminiTextEmbeddingRequest } from "./embeddings-gemini.js";
import { hashText } from "./internal.js";
import { withRemoteHttpResponse } from "./remote-http.js";
buildBatchHeaders,
debugEmbeddingsLog,
normalizeBatchBaseUrl,
sanitizeAndNormalizeEmbedding,
withRemoteHttpResponse,
} from "openclaw/plugin-sdk/memory-core-host-engine-embeddings";
import type { GeminiEmbeddingClient, GeminiTextEmbeddingRequest } from "./embedding-provider.js";
export type GeminiBatchRequest = {
custom_id: string;
@@ -40,6 +41,10 @@ export type GeminiBatchOutputLine = {
};
const GEMINI_BATCH_MAX_REQUESTS = 50000;
function hashText(text: string): string {
return crypto.createHash("sha256").update(text).digest("hex");
}
function getGeminiUploadUrl(baseUrl: string): string {
if (baseUrl.includes("/v1beta")) {
return baseUrl.replace(/\/v1beta\/?$/, "/upload/v1beta");

View File

@@ -1,84 +1,40 @@
import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";
import * as authModule from "../../agents/model-auth.js";
import { afterEach, describe, expect, it, vi } from "vitest";
import {
buildGeminiEmbeddingRequest,
buildGeminiTextEmbeddingRequest,
createGeminiEmbeddingProvider,
DEFAULT_GEMINI_EMBEDDING_MODEL,
GEMINI_EMBEDDING_2_MODELS,
isGeminiEmbedding2Model,
normalizeGeminiModel,
resolveGeminiOutputDimensionality,
} from "./embeddings-gemini-request.js";
import {
createGeminiBatchFetchMock,
createJsonResponseFetchMock,
installFetchMock,
mockResolvedProviderKey,
parseFetchBody,
readFirstFetchRequest,
type JsonFetchMock,
} from "./embeddings-provider.test-support.js";
const { resolveApiKeyForProviderMock } = vi.hoisted(() => ({
resolveApiKeyForProviderMock: vi.fn(),
}));
vi.mock("../../agents/model-auth.js", () => {
return {
resolveApiKeyForProvider: resolveApiKeyForProviderMock,
requireApiKey: (auth: { apiKey?: string; mode?: string }, provider: string) => {
if (auth.apiKey) {
return auth.apiKey;
}
throw new Error(`No API key resolved for provider "${provider}" (auth mode: ${auth.mode}).`);
},
};
});
vi.mock("../../agents/api-key-rotation.js", () => ({
collectProviderApiKeysForExecution: (params: { primaryApiKey?: string }) =>
params.primaryApiKey ? [params.primaryApiKey] : [],
executeWithApiKeyRotation: async <T>(params: {
apiKeys: string[];
execute: (apiKey: string) => Promise<T>;
}) => {
const apiKey = params.apiKeys[0];
if (!apiKey) {
throw new Error('No API keys configured for provider "google".');
}
return await params.execute(apiKey);
},
}));
beforeEach(() => {
vi.useRealTimers();
vi.doUnmock("undici");
});
} from "./embedding-provider.js";
afterEach(() => {
vi.doUnmock("undici");
vi.resetAllMocks();
vi.restoreAllMocks();
vi.unstubAllGlobals();
});
type GeminiProviderOptions = Parameters<
typeof import("./embeddings-gemini.js").createGeminiEmbeddingProvider
>[0];
async function createProviderWithFetch(
fetchMock: JsonFetchMock,
options: Partial<GeminiProviderOptions> & { model: string },
) {
installFetchMock(fetchMock as unknown as typeof globalThis.fetch);
mockResolvedProviderKey(authModule.resolveApiKeyForProvider);
const { createGeminiEmbeddingProvider } = await import("./embeddings-gemini.js");
const { provider } = await createGeminiEmbeddingProvider({
config: {} as never,
provider: "gemini",
fallback: "none",
...options,
function installFetchMock(
handler: (input: RequestInfo | URL, init?: RequestInit) => unknown,
): ReturnType<typeof vi.fn> {
const fetchMock = vi.fn(async (input: RequestInfo | URL, init?: RequestInit) => {
return new Response(JSON.stringify(handler(input, init)), {
status: 200,
headers: { "Content-Type": "application/json" },
});
});
return provider;
vi.stubGlobal("fetch", fetchMock);
return fetchMock;
}
function fetchJsonBody(fetchMock: ReturnType<typeof vi.fn>, index: number): unknown {
const init = fetchMock.mock.calls[index]?.[1] as RequestInit | undefined;
const body = init?.body;
if (typeof body !== "string") {
throw new Error("Expected JSON string request body.");
}
return JSON.parse(body) as unknown;
}
describe("Gemini embedding request helpers", () => {
@@ -149,24 +105,9 @@ describe("Gemini embedding request helpers", () => {
});
});
describe("gemini embedding provider", () => {
describe("Gemini embedding provider", () => {
it("handles legacy and v2 request/response behavior", async () => {
const legacyFetch = createGeminiBatchFetchMock(2);
const legacyProvider = await createProviderWithFetch(legacyFetch, {
model: "gemini-embedding-001",
});
await legacyProvider.embedQuery("test query");
await legacyProvider.embedBatch(["text1", "text2"]);
expect(parseFetchBody(legacyFetch, 0)).toMatchObject({
taskType: "RETRIEVAL_QUERY",
content: { parts: [{ text: "test query" }] },
});
expect(parseFetchBody(legacyFetch, 0)).not.toHaveProperty("outputDimensionality");
expect(parseFetchBody(legacyFetch, 1)).not.toHaveProperty("outputDimensionality");
const v2Fetch = createJsonResponseFetchMock((input) => {
const fetchMock = installFetchMock((input) => {
const url = input instanceof URL ? input.href : typeof input === "string" ? input : input.url;
return url.endsWith(":batchEmbedContents")
? {
@@ -176,16 +117,22 @@ describe("gemini embedding provider", () => {
}
: { embedding: { values: [3, 4, Number.NaN] } };
});
const v2Provider = await createProviderWithFetch(v2Fetch, {
const { provider } = await createGeminiEmbeddingProvider({
config: {} as never,
provider: "gemini",
remote: { apiKey: "test-key" },
model: "gemini-embedding-2-preview",
outputDimensionality: 768,
taskType: "SEMANTIC_SIMILARITY",
fallback: "none",
});
await expect(v2Provider.embedQuery(" ")).resolves.toEqual([]);
await expect(v2Provider.embedBatch([])).resolves.toEqual([]);
await expect(v2Provider.embedQuery("test query")).resolves.toEqual([0.6, 0.8, 0]);
const structuredBatch = await v2Provider.embedBatchInputs?.([
await expect(provider.embedQuery(" ")).resolves.toEqual([]);
await expect(provider.embedBatch([])).resolves.toEqual([]);
await expect(provider.embedQuery("test query")).resolves.toEqual([0.6, 0.8, 0]);
const structuredBatch = await provider.embedBatchInputs?.([
{
text: "Image file: diagram.png",
parts: [
@@ -206,38 +153,39 @@ describe("gemini embedding provider", () => {
[0, 0, 1],
]);
const { url } = readFirstFetchRequest(v2Fetch);
expect(url).toBe(
expect(fetchMock.mock.calls[0]?.[0]).toBe(
"https://generativelanguage.googleapis.com/v1beta/models/gemini-embedding-2-preview:embedContent",
);
expect(parseFetchBody(v2Fetch, 0)).toMatchObject({
expect(fetchJsonBody(fetchMock, 0)).toMatchObject({
outputDimensionality: 768,
taskType: "SEMANTIC_SIMILARITY",
content: { parts: [{ text: "test query" }] },
});
expect(parseFetchBody(v2Fetch, 1).requests).toEqual([
{
model: "models/gemini-embedding-2-preview",
content: {
parts: [
{ text: "Image file: diagram.png" },
{ inlineData: { mimeType: "image/png", data: "img" } },
],
expect(fetchJsonBody(fetchMock, 1)).toMatchObject({
requests: [
{
model: "models/gemini-embedding-2-preview",
content: {
parts: [
{ text: "Image file: diagram.png" },
{ inlineData: { mimeType: "image/png", data: "img" } },
],
},
taskType: "SEMANTIC_SIMILARITY",
outputDimensionality: 768,
},
taskType: "SEMANTIC_SIMILARITY",
outputDimensionality: 768,
},
{
model: "models/gemini-embedding-2-preview",
content: {
parts: [
{ text: "Audio file: note.wav" },
{ inlineData: { mimeType: "audio/wav", data: "aud" } },
],
{
model: "models/gemini-embedding-2-preview",
content: {
parts: [
{ text: "Audio file: note.wav" },
{ inlineData: { mimeType: "audio/wav", data: "aud" } },
],
},
taskType: "SEMANTIC_SIMILARITY",
outputDimensionality: 768,
},
taskType: "SEMANTIC_SIMILARITY",
outputDimensionality: 768,
},
]);
],
});
});
});

View File

@@ -1,44 +1,22 @@
import { parseGeminiAuth } from "openclaw/plugin-sdk/image-generation-core";
import {
buildRemoteBaseUrlPolicy,
debugEmbeddingsLog,
sanitizeAndNormalizeEmbedding,
withRemoteHttpResponse,
type EmbeddingInput,
type MemoryEmbeddingProvider,
type MemoryEmbeddingProviderCreateOptions,
} from "openclaw/plugin-sdk/memory-core-host-engine-embeddings";
import { resolveMemorySecretInputString } from "openclaw/plugin-sdk/memory-core-host-secret";
import {
collectProviderApiKeysForExecution,
executeWithApiKeyRotation,
} from "../../agents/api-key-rotation.js";
import { requireApiKey, resolveApiKeyForProvider } from "../../agents/model-auth.js";
import { parseGeminiAuth } from "../../infra/gemini-auth.js";
import {
DEFAULT_GOOGLE_API_BASE_URL,
normalizeGoogleApiBaseUrl,
} from "../../infra/google-api-base-url.js";
import type { SsrFPolicy } from "../../infra/net/ssrf.js";
import { normalizeOptionalString } from "../../shared/string-coerce.js";
import type { EmbeddingInput } from "./embedding-inputs.js";
import { sanitizeAndNormalizeEmbedding } from "./embedding-vectors.js";
import { debugEmbeddingsLog } from "./embeddings-debug.js";
import {
buildGeminiEmbeddingRequest,
buildGeminiTextEmbeddingRequest,
isGeminiEmbedding2Model,
normalizeGeminiModel,
resolveGeminiOutputDimensionality,
} from "./embeddings-gemini-request.js";
import type { EmbeddingProvider, EmbeddingProviderOptions } from "./embeddings.types.js";
import { buildRemoteBaseUrlPolicy, withRemoteHttpResponse } from "./remote-http.js";
import { resolveMemorySecretInputString } from "./secret-input.js";
export {
buildGeminiEmbeddingRequest,
buildGeminiTextEmbeddingRequest,
DEFAULT_GEMINI_EMBEDDING_MODEL,
GEMINI_EMBEDDING_2_MODELS,
isGeminiEmbedding2Model,
normalizeGeminiModel,
resolveGeminiOutputDimensionality,
type GeminiEmbeddingRequest,
type GeminiInlinePart,
type GeminiPart,
type GeminiTaskType,
type GeminiTextEmbeddingRequest,
type GeminiTextPart,
} from "./embeddings-gemini-request.js";
requireApiKey,
resolveApiKeyForProvider,
} from "openclaw/plugin-sdk/provider-auth-runtime";
import type { SsrFPolicy } from "openclaw/plugin-sdk/ssrf-runtime";
import { normalizeOptionalString } from "openclaw/plugin-sdk/text-runtime";
export type GeminiEmbeddingClient = {
baseUrl: string;
@@ -50,9 +28,111 @@ export type GeminiEmbeddingClient = {
outputDimensionality?: number;
};
export const DEFAULT_GEMINI_EMBEDDING_MODEL = "gemini-embedding-001";
const DEFAULT_GOOGLE_API_BASE_URL = "https://generativelanguage.googleapis.com/v1beta";
const GEMINI_MAX_INPUT_TOKENS: Record<string, number> = {
"text-embedding-004": 2048,
"gemini-embedding-001": 2048,
"gemini-embedding-2-preview": 8192,
};
export type GeminiTaskType = NonNullable<MemoryEmbeddingProviderCreateOptions["taskType"]>;
// --- gemini-embedding-2-preview support ---
export const GEMINI_EMBEDDING_2_MODELS = new Set([
"gemini-embedding-2-preview",
// Add the GA model name here once released.
]);
const GEMINI_EMBEDDING_2_DEFAULT_DIMENSIONS = 3072;
const GEMINI_EMBEDDING_2_VALID_DIMENSIONS = [768, 1536, 3072] as const;
export type GeminiTextPart = { text: string };
export type GeminiInlinePart = {
inlineData: { mimeType: string; data: string };
};
export type GeminiPart = GeminiTextPart | GeminiInlinePart;
export type GeminiEmbeddingRequest = {
content: { parts: GeminiPart[] };
taskType: GeminiTaskType;
outputDimensionality?: number;
model?: string;
};
export type GeminiTextEmbeddingRequest = GeminiEmbeddingRequest;
/** Builds the text-only Gemini embedding request shape used across direct and batch APIs. */
export function buildGeminiTextEmbeddingRequest(params: {
text: string;
taskType: GeminiTaskType;
outputDimensionality?: number;
modelPath?: string;
}): GeminiTextEmbeddingRequest {
return buildGeminiEmbeddingRequest({
input: { text: params.text },
taskType: params.taskType,
outputDimensionality: params.outputDimensionality,
modelPath: params.modelPath,
});
}
export function buildGeminiEmbeddingRequest(params: {
input: EmbeddingInput;
taskType: GeminiTaskType;
outputDimensionality?: number;
modelPath?: string;
}): GeminiEmbeddingRequest {
const request: GeminiEmbeddingRequest = {
content: {
parts: params.input.parts?.map((part) =>
part.type === "text"
? ({ text: part.text } satisfies GeminiTextPart)
: ({
inlineData: { mimeType: part.mimeType, data: part.data },
} satisfies GeminiInlinePart),
) ?? [{ text: params.input.text }],
},
taskType: params.taskType,
};
if (params.modelPath) {
request.model = params.modelPath;
}
if (params.outputDimensionality != null) {
request.outputDimensionality = params.outputDimensionality;
}
return request;
}
/**
* Returns true if the given model name is a gemini-embedding-2 variant that
* supports `outputDimensionality` and extended task types.
*/
export function isGeminiEmbedding2Model(model: string): boolean {
return GEMINI_EMBEDDING_2_MODELS.has(model);
}
/**
* Validate and return the `outputDimensionality` for gemini-embedding-2 models.
* Returns `undefined` for older models (they don't support the param).
*/
export function resolveGeminiOutputDimensionality(
model: string,
requested?: number,
): number | undefined {
if (!isGeminiEmbedding2Model(model)) {
return undefined;
}
if (requested == null) {
return GEMINI_EMBEDDING_2_DEFAULT_DIMENSIONS;
}
const valid: readonly number[] = GEMINI_EMBEDDING_2_VALID_DIMENSIONS;
if (!valid.includes(requested)) {
throw new Error(
`Invalid outputDimensionality ${requested} for ${model}. Valid values: ${valid.join(", ")}`,
);
}
return requested;
}
function resolveRemoteApiKey(remoteApiKey: unknown): string | undefined {
const trimmed = resolveMemorySecretInputString({
value: remoteApiKey,
@@ -67,6 +147,21 @@ function resolveRemoteApiKey(remoteApiKey: unknown): string | undefined {
return trimmed;
}
export function normalizeGeminiModel(model: string): string {
const trimmed = model.trim();
if (!trimmed) {
return DEFAULT_GEMINI_EMBEDDING_MODEL;
}
const withoutPrefix = trimmed.replace(/^models\//, "");
if (withoutPrefix.startsWith("gemini/")) {
return withoutPrefix.slice("gemini/".length);
}
if (withoutPrefix.startsWith("google/")) {
return withoutPrefix.slice("google/".length);
}
return withoutPrefix;
}
async function fetchGeminiEmbeddingPayload(params: {
client: GeminiEmbeddingClient;
endpoint: string;
@@ -120,9 +215,30 @@ function buildGeminiModelPath(model: string): string {
return model.startsWith("models/") ? model : `models/${model}`;
}
function normalizeGoogleApiBaseUrl(baseUrl: string): string {
const trimmed = baseUrl.trim().replace(/\/+$/, "");
if (!trimmed) {
return DEFAULT_GOOGLE_API_BASE_URL;
}
try {
const url = new URL(trimmed);
url.hash = "";
url.search = "";
if (
url.origin.toLowerCase() === "https://generativelanguage.googleapis.com" &&
url.pathname.replace(/\/+$/, "") === ""
) {
url.pathname = "/v1beta";
}
return url.toString().replace(/\/+$/, "");
} catch {
return trimmed;
}
}
export async function createGeminiEmbeddingProvider(
options: EmbeddingProviderOptions,
): Promise<{ provider: EmbeddingProvider; client: GeminiEmbeddingClient }> {
options: MemoryEmbeddingProviderCreateOptions,
): Promise<{ provider: MemoryEmbeddingProvider; client: GeminiEmbeddingClient }> {
const client = await resolveGeminiEmbeddingClient(options);
const baseUrl = client.baseUrl.replace(/\/$/, "");
const embedUrl = `${baseUrl}/${client.modelPath}:embedContent`;
@@ -190,7 +306,7 @@ export async function createGeminiEmbeddingProvider(
}
export async function resolveGeminiEmbeddingClient(
options: EmbeddingProviderOptions,
options: MemoryEmbeddingProviderCreateOptions,
): Promise<GeminiEmbeddingClient> {
const remote = options.remote;
const remoteApiKey = resolveRemoteApiKey(remote?.apiKey);

View File

@@ -4,6 +4,7 @@ import type {
ProviderFetchUsageSnapshotContext,
} from "openclaw/plugin-sdk/plugin-entry";
import { buildOauthProviderAuthResult } from "openclaw/plugin-sdk/provider-auth-result";
import type { ProviderPlugin } from "openclaw/plugin-sdk/provider-model-shared";
import { buildProviderToolCompatFamilyHooks } from "openclaw/plugin-sdk/provider-tools";
import { fetchGeminiUsage } from "openclaw/plugin-sdk/provider-usage";
import { formatGoogleOauthApiKey, parseGoogleUsageToken } from "./oauth-token-shared.js";
@@ -29,8 +30,8 @@ async function fetchGeminiCliUsage(ctx: ProviderFetchUsageSnapshotContext) {
return await fetchGeminiUsage(ctx.token, ctx.timeoutMs, ctx.fetchFn, PROVIDER_ID);
}
export function registerGoogleGeminiCliProvider(api: OpenClawPluginApi) {
api.registerProvider({
export function buildGoogleGeminiCliProvider(): ProviderPlugin {
return {
id: PROVIDER_ID,
label: PROVIDER_LABEL,
docsPath: "/providers/models",
@@ -128,5 +129,9 @@ export function registerGoogleGeminiCliProvider(api: OpenClawPluginApi) {
};
},
fetchUsageSnapshot: async (ctx) => await fetchGeminiCliUsage(ctx),
});
};
}
export function registerGoogleGeminiCliProvider(api: OpenClawPluginApi) {
api.registerProvider(buildGoogleGeminiCliProvider());
}

View File

@@ -0,0 +1,121 @@
import type { MusicGenerationProvider } from "openclaw/plugin-sdk/music-generation";
import { isProviderApiKeyConfigured } from "openclaw/plugin-sdk/provider-auth";
import type {
VideoGenerationProvider,
VideoGenerationProviderConfiguredContext,
} from "openclaw/plugin-sdk/video-generation";
export const DEFAULT_GOOGLE_MUSIC_MODEL = "lyria-3-clip-preview";
export const GOOGLE_PRO_MUSIC_MODEL = "lyria-3-pro-preview";
export const GOOGLE_MAX_INPUT_IMAGES = 10;
export const DEFAULT_GOOGLE_VIDEO_MODEL = "veo-3.1-fast-generate-preview";
export const GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS = [4, 6, 8] as const;
export const GOOGLE_VIDEO_MIN_DURATION_SECONDS = GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS[0];
export const GOOGLE_VIDEO_MAX_DURATION_SECONDS =
GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS[GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS.length - 1];
function isGoogleProviderConfigured(
ctx: { agentDir?: string } | VideoGenerationProviderConfiguredContext,
): boolean {
return isProviderApiKeyConfigured({
provider: "google",
agentDir: ctx.agentDir,
});
}
export function createGoogleMusicGenerationProviderMetadata(): Omit<
MusicGenerationProvider,
"generateMusic"
> {
return {
id: "google",
label: "Google",
defaultModel: DEFAULT_GOOGLE_MUSIC_MODEL,
models: [DEFAULT_GOOGLE_MUSIC_MODEL, GOOGLE_PRO_MUSIC_MODEL],
isConfigured: isGoogleProviderConfigured,
capabilities: {
generate: {
maxTracks: 1,
supportsLyrics: true,
supportsInstrumental: true,
supportsFormat: true,
supportedFormatsByModel: {
[DEFAULT_GOOGLE_MUSIC_MODEL]: ["mp3"],
[GOOGLE_PRO_MUSIC_MODEL]: ["mp3", "wav"],
},
},
edit: {
enabled: true,
maxTracks: 1,
maxInputImages: GOOGLE_MAX_INPUT_IMAGES,
supportsLyrics: true,
supportsInstrumental: true,
supportsFormat: true,
supportedFormatsByModel: {
[DEFAULT_GOOGLE_MUSIC_MODEL]: ["mp3"],
[GOOGLE_PRO_MUSIC_MODEL]: ["mp3", "wav"],
},
},
},
};
}
export function createGoogleVideoGenerationProviderMetadata(): Omit<
VideoGenerationProvider,
"generateVideo"
> {
return {
id: "google",
label: "Google",
defaultModel: DEFAULT_GOOGLE_VIDEO_MODEL,
models: [
DEFAULT_GOOGLE_VIDEO_MODEL,
"veo-3.1-generate-preview",
"veo-3.1-lite-generate-preview",
"veo-3.0-fast-generate-001",
"veo-3.0-generate-001",
"veo-2.0-generate-001",
],
isConfigured: isGoogleProviderConfigured,
capabilities: {
generate: {
maxVideos: 1,
maxDurationSeconds: GOOGLE_VIDEO_MAX_DURATION_SECONDS,
supportedDurationSeconds: [...GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS],
aspectRatios: ["16:9", "9:16"],
resolutions: ["720P", "1080P"],
supportsAspectRatio: true,
supportsResolution: true,
supportsSize: true,
supportsAudio: true,
},
imageToVideo: {
enabled: true,
maxVideos: 1,
maxInputImages: 1,
maxDurationSeconds: GOOGLE_VIDEO_MAX_DURATION_SECONDS,
supportedDurationSeconds: [...GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS],
aspectRatios: ["16:9", "9:16"],
resolutions: ["720P", "1080P"],
supportsAspectRatio: true,
supportsResolution: true,
supportsSize: true,
supportsAudio: true,
},
videoToVideo: {
enabled: true,
maxVideos: 1,
maxInputVideos: 1,
maxDurationSeconds: GOOGLE_VIDEO_MAX_DURATION_SECONDS,
supportedDurationSeconds: [...GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS],
aspectRatios: ["16:9", "9:16"],
resolutions: ["720P", "1080P"],
supportsAspectRatio: true,
supportsResolution: true,
supportsSize: true,
supportsAudio: true,
},
},
};
}

View File

@@ -1,16 +1,23 @@
import type { ImageGenerationProvider } from "openclaw/plugin-sdk/image-generation";
import type { MediaUnderstandingProvider } from "openclaw/plugin-sdk/media-understanding";
import type { MusicGenerationProvider } from "openclaw/plugin-sdk/music-generation";
import { definePluginEntry } from "openclaw/plugin-sdk/plugin-entry";
import type { VideoGenerationProvider } from "openclaw/plugin-sdk/video-generation";
import { buildGoogleGeminiCliBackend } from "./cli-backend.js";
import { registerGoogleGeminiCliProvider } from "./gemini-cli-provider.js";
import { buildGoogleMusicGenerationProvider } from "./music-generation-provider.js";
import {
createGoogleMusicGenerationProviderMetadata,
createGoogleVideoGenerationProviderMetadata,
} from "./generation-provider-metadata.js";
import { geminiMemoryEmbeddingProviderAdapter } from "./memory-embedding-adapter.js";
import { registerGoogleProvider } from "./provider-registration.js";
import { buildGoogleSpeechProvider } from "./speech-provider.js";
import { createGeminiWebSearchProvider } from "./src/gemini-web-search-provider.js";
import { buildGoogleVideoGenerationProvider } from "./video-generation-provider.js";
let googleImageGenerationProviderPromise: Promise<ImageGenerationProvider> | null = null;
let googleMediaUnderstandingProviderPromise: Promise<MediaUnderstandingProvider> | null = null;
let googleMusicGenerationProviderPromise: Promise<MusicGenerationProvider> | null = null;
let googleVideoGenerationProviderPromise: Promise<VideoGenerationProvider> | null = null;
type GoogleMediaUnderstandingProvider = Required<
Pick<
@@ -37,6 +44,24 @@ async function loadGoogleMediaUnderstandingProvider(): Promise<MediaUnderstandin
return await googleMediaUnderstandingProviderPromise;
}
async function loadGoogleMusicGenerationProvider(): Promise<MusicGenerationProvider> {
if (!googleMusicGenerationProviderPromise) {
googleMusicGenerationProviderPromise = import("./music-generation-provider.js").then((mod) =>
mod.buildGoogleMusicGenerationProvider(),
);
}
return await googleMusicGenerationProviderPromise;
}
async function loadGoogleVideoGenerationProvider(): Promise<VideoGenerationProvider> {
if (!googleVideoGenerationProviderPromise) {
googleVideoGenerationProviderPromise = import("./video-generation-provider.js").then((mod) =>
mod.buildGoogleVideoGenerationProvider(),
);
}
return await googleVideoGenerationProviderPromise;
}
async function loadGoogleRequiredMediaUnderstandingProvider(): Promise<GoogleMediaUnderstandingProvider> {
const provider = await loadGoogleMediaUnderstandingProvider();
if (
@@ -103,6 +128,22 @@ function createLazyGoogleMediaUnderstandingProvider(): MediaUnderstandingProvide
};
}
function createLazyGoogleMusicGenerationProvider(): MusicGenerationProvider {
return {
...createGoogleMusicGenerationProviderMetadata(),
generateMusic: async (...args) =>
await (await loadGoogleMusicGenerationProvider()).generateMusic(...args),
};
}
function createLazyGoogleVideoGenerationProvider(): VideoGenerationProvider {
return {
...createGoogleVideoGenerationProviderMetadata(),
generateVideo: async (...args) =>
await (await loadGoogleVideoGenerationProvider()).generateVideo(...args),
};
}
export default definePluginEntry({
id: "google",
name: "Google Plugin",
@@ -111,11 +152,12 @@ export default definePluginEntry({
api.registerCliBackend(buildGoogleGeminiCliBackend());
registerGoogleGeminiCliProvider(api);
registerGoogleProvider(api);
api.registerMemoryEmbeddingProvider(geminiMemoryEmbeddingProviderAdapter);
api.registerImageGenerationProvider(createLazyGoogleImageGenerationProvider());
api.registerMediaUnderstandingProvider(createLazyGoogleMediaUnderstandingProvider());
api.registerMusicGenerationProvider(buildGoogleMusicGenerationProvider());
api.registerMusicGenerationProvider(createLazyGoogleMusicGenerationProvider());
api.registerSpeechProvider(buildGoogleSpeechProvider());
api.registerVideoGenerationProvider(buildGoogleVideoGenerationProvider());
api.registerVideoGenerationProvider(createLazyGoogleVideoGenerationProvider());
api.registerWebSearchProvider(createGeminiWebSearchProvider());
},
});

View File

@@ -0,0 +1,79 @@
import {
hasNonTextEmbeddingParts,
isMissingEmbeddingApiKeyError,
mapBatchEmbeddingsByIndex,
sanitizeEmbeddingCacheHeaders,
type MemoryEmbeddingProviderAdapter,
} from "openclaw/plugin-sdk/memory-core-host-engine-embeddings";
import { runGeminiEmbeddingBatches } from "./embedding-batch.js";
import {
buildGeminiEmbeddingRequest,
createGeminiEmbeddingProvider,
DEFAULT_GEMINI_EMBEDDING_MODEL,
} from "./embedding-provider.js";
function supportsGeminiMultimodalEmbeddings(model: string): boolean {
const normalized = model
.trim()
.replace(/^models\//, "")
.replace(/^(gemini|google)\//, "");
return normalized === "gemini-embedding-2-preview";
}
export const geminiMemoryEmbeddingProviderAdapter: MemoryEmbeddingProviderAdapter = {
id: "gemini",
defaultModel: DEFAULT_GEMINI_EMBEDDING_MODEL,
transport: "remote",
authProviderId: "google",
autoSelectPriority: 30,
allowExplicitWhenConfiguredAuto: true,
supportsMultimodalEmbeddings: ({ model }) => supportsGeminiMultimodalEmbeddings(model),
shouldContinueAutoSelection: isMissingEmbeddingApiKeyError,
create: async (options) => {
const { provider, client } = await createGeminiEmbeddingProvider({
...options,
provider: "gemini",
fallback: "none",
});
return {
provider,
runtime: {
id: "gemini",
cacheKeyData: {
provider: "gemini",
baseUrl: client.baseUrl,
model: client.model,
outputDimensionality: client.outputDimensionality,
headers: sanitizeEmbeddingCacheHeaders(client.headers, [
"authorization",
"x-goog-api-key",
]),
},
batchEmbed: async (batch) => {
if (batch.chunks.some((chunk) => hasNonTextEmbeddingParts(chunk.embeddingInput))) {
return null;
}
const byCustomId = await runGeminiEmbeddingBatches({
gemini: client,
agentId: batch.agentId,
requests: batch.chunks.map((chunk, index) => ({
custom_id: String(index),
request: buildGeminiEmbeddingRequest({
input: chunk.embeddingInput ?? { text: chunk.text },
taskType: "RETRIEVAL_DOCUMENT",
modelPath: client.modelPath,
outputDimensionality: client.outputDimensionality,
}),
})),
wait: batch.wait,
concurrency: batch.concurrency,
pollIntervalMs: batch.pollIntervalMs,
timeoutMs: batch.timeoutMs,
debug: batch.debug,
});
return mapBatchEmbeddingsByIndex(byCustomId, batch.chunks.length);
},
},
};
},
};

View File

@@ -5,15 +5,17 @@ import type {
MusicGenerationProvider,
MusicGenerationRequest,
} from "openclaw/plugin-sdk/music-generation";
import { isProviderApiKeyConfigured } from "openclaw/plugin-sdk/provider-auth";
import { resolveApiKeyForProvider } from "openclaw/plugin-sdk/provider-auth-runtime";
import { normalizeOptionalString } from "openclaw/plugin-sdk/text-runtime";
import { normalizeGoogleApiBaseUrl } from "./api.js";
import {
createGoogleMusicGenerationProviderMetadata,
DEFAULT_GOOGLE_MUSIC_MODEL,
GOOGLE_MAX_INPUT_IMAGES,
GOOGLE_PRO_MUSIC_MODEL,
} from "./generation-provider-metadata.js";
const DEFAULT_GOOGLE_MUSIC_MODEL = "lyria-3-clip-preview";
const GOOGLE_PRO_MUSIC_MODEL = "lyria-3-pro-preview";
const DEFAULT_TIMEOUT_MS = 180_000;
const GOOGLE_MAX_INPUT_IMAGES = 10;
type GoogleInlineDataPart = {
mimeType?: string;
@@ -99,39 +101,7 @@ function extractTracks(params: { payload: GoogleGenerateMusicResponse; model: st
export function buildGoogleMusicGenerationProvider(): MusicGenerationProvider {
return {
id: "google",
label: "Google",
defaultModel: DEFAULT_GOOGLE_MUSIC_MODEL,
models: [DEFAULT_GOOGLE_MUSIC_MODEL, GOOGLE_PRO_MUSIC_MODEL],
isConfigured: ({ agentDir }) =>
isProviderApiKeyConfigured({
provider: "google",
agentDir,
}),
capabilities: {
generate: {
maxTracks: 1,
supportsLyrics: true,
supportsInstrumental: true,
supportsFormat: true,
supportedFormatsByModel: {
[DEFAULT_GOOGLE_MUSIC_MODEL]: ["mp3"],
[GOOGLE_PRO_MUSIC_MODEL]: ["mp3", "wav"],
},
},
edit: {
enabled: true,
maxTracks: 1,
maxInputImages: GOOGLE_MAX_INPUT_IMAGES,
supportsLyrics: true,
supportsInstrumental: true,
supportsFormat: true,
supportedFormatsByModel: {
[DEFAULT_GOOGLE_MUSIC_MODEL]: ["mp3"],
[GOOGLE_PRO_MUSIC_MODEL]: ["mp3", "wav"],
},
},
},
...createGoogleMusicGenerationProviderMetadata(),
async generateMusic(req) {
if ((req.inputImages?.length ?? 0) > GOOGLE_MAX_INPUT_IMAGES) {
throw new Error(

View File

@@ -0,0 +1,28 @@
import {
applyAgentDefaultModelPrimary,
type OpenClawConfig,
} from "openclaw/plugin-sdk/provider-onboard";
export const GOOGLE_GEMINI_DEFAULT_MODEL = "google/gemini-3.1-pro-preview";
export function applyGoogleGeminiModelDefault(cfg: OpenClawConfig): {
next: OpenClawConfig;
changed: boolean;
} {
const current = cfg.agents?.defaults?.model as unknown;
const currentPrimary =
typeof current === "string"
? current.trim() || undefined
: current &&
typeof current === "object" &&
typeof (current as { primary?: unknown }).primary === "string"
? ((current as { primary: string }).primary || "").trim() || undefined
: undefined;
if (currentPrimary === GOOGLE_GEMINI_DEFAULT_MODEL) {
return { next: cfg, changed: false };
}
return {
next: applyAgentDefaultModelPrimary(cfg, GOOGLE_GEMINI_DEFAULT_MODEL),
changed: true,
};
}

View File

@@ -46,6 +46,7 @@
},
"contracts": {
"mediaUnderstandingProviders": ["google"],
"memoryEmbeddingProviders": ["gemini"],
"imageGenerationProviders": ["google"],
"musicGenerationProviders": ["google"],
"speechProviders": ["google"],

View File

@@ -1,17 +1,17 @@
import type { OpenClawPluginApi } from "openclaw/plugin-sdk/plugin-entry";
import { createProviderApiKeyAuthMethod } from "openclaw/plugin-sdk/provider-auth-api-key";
import {
GOOGLE_GEMINI_DEFAULT_MODEL,
applyGoogleGeminiModelDefault,
normalizeGoogleProviderConfig,
normalizeGoogleModelId,
resolveGoogleGenerativeAiTransport,
} from "./api.js";
import type { ProviderPlugin } from "openclaw/plugin-sdk/provider-model-shared";
import { normalizeGoogleModelId } from "./model-id.js";
import { GOOGLE_GEMINI_DEFAULT_MODEL, applyGoogleGeminiModelDefault } from "./onboard.js";
import { GOOGLE_GEMINI_PROVIDER_HOOKS } from "./provider-hooks.js";
import { isModernGoogleModel, resolveGoogleGeminiForwardCompatModel } from "./provider-models.js";
import {
normalizeGoogleProviderConfig,
resolveGoogleGenerativeAiTransport,
} from "./provider-policy.js";
export function registerGoogleProvider(api: OpenClawPluginApi) {
api.registerProvider({
export function buildGoogleProvider(): ProviderPlugin {
return {
id: "google",
label: "Google AI Studio",
docsPath: "/providers/models",
@@ -50,5 +50,9 @@ export function registerGoogleProvider(api: OpenClawPluginApi) {
}),
...GOOGLE_GEMINI_PROVIDER_HOOKS,
isModernModelRef: ({ modelId }) => isModernGoogleModel(modelId),
});
};
}
export function registerGoogleProvider(api: OpenClawPluginApi) {
api.registerProvider(buildGoogleProvider());
}

View File

@@ -1,7 +1,6 @@
import { mkdtemp, readFile, rm } from "node:fs/promises";
import path from "node:path";
import { GoogleGenAI } from "@google/genai";
import { isProviderApiKeyConfigured } from "openclaw/plugin-sdk/provider-auth";
import { resolveApiKeyForProvider } from "openclaw/plugin-sdk/provider-auth-runtime";
import {
createProviderOperationDeadline,
@@ -16,15 +15,17 @@ import type {
VideoGenerationRequest,
} from "openclaw/plugin-sdk/video-generation";
import { normalizeGoogleApiBaseUrl } from "./api.js";
import {
createGoogleVideoGenerationProviderMetadata,
DEFAULT_GOOGLE_VIDEO_MODEL,
GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS,
GOOGLE_VIDEO_MAX_DURATION_SECONDS,
GOOGLE_VIDEO_MIN_DURATION_SECONDS,
} from "./generation-provider-metadata.js";
const DEFAULT_GOOGLE_VIDEO_MODEL = "veo-3.1-fast-generate-preview";
const DEFAULT_TIMEOUT_MS = 180_000;
const POLL_INTERVAL_MS = 10_000;
const MAX_POLL_ATTEMPTS = 90;
const GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS = [4, 6, 8] as const;
const GOOGLE_VIDEO_MIN_DURATION_SECONDS = GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS[0];
const GOOGLE_VIDEO_MAX_DURATION_SECONDS =
GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS[GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS.length - 1];
function resolveConfiguredGoogleVideoBaseUrl(req: VideoGenerationRequest): string | undefined {
const configured = normalizeOptionalString(req.cfg?.models?.providers?.google?.baseUrl);
@@ -151,61 +152,7 @@ async function downloadGeneratedVideo(params: {
export function buildGoogleVideoGenerationProvider(): VideoGenerationProvider {
return {
id: "google",
label: "Google",
defaultModel: DEFAULT_GOOGLE_VIDEO_MODEL,
models: [
DEFAULT_GOOGLE_VIDEO_MODEL,
"veo-3.1-generate-preview",
"veo-3.1-lite-generate-preview",
"veo-3.0-fast-generate-001",
"veo-3.0-generate-001",
"veo-2.0-generate-001",
],
isConfigured: ({ agentDir }) =>
isProviderApiKeyConfigured({
provider: "google",
agentDir,
}),
capabilities: {
generate: {
maxVideos: 1,
maxDurationSeconds: GOOGLE_VIDEO_MAX_DURATION_SECONDS,
supportedDurationSeconds: GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS,
aspectRatios: ["16:9", "9:16"],
resolutions: ["720P", "1080P"],
supportsAspectRatio: true,
supportsResolution: true,
supportsSize: true,
supportsAudio: true,
},
imageToVideo: {
enabled: true,
maxVideos: 1,
maxInputImages: 1,
maxDurationSeconds: GOOGLE_VIDEO_MAX_DURATION_SECONDS,
supportedDurationSeconds: GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS,
aspectRatios: ["16:9", "9:16"],
resolutions: ["720P", "1080P"],
supportsAspectRatio: true,
supportsResolution: true,
supportsSize: true,
supportsAudio: true,
},
videoToVideo: {
enabled: true,
maxVideos: 1,
maxInputVideos: 1,
maxDurationSeconds: GOOGLE_VIDEO_MAX_DURATION_SECONDS,
supportedDurationSeconds: GOOGLE_VIDEO_ALLOWED_DURATION_SECONDS,
aspectRatios: ["16:9", "9:16"],
resolutions: ["720P", "1080P"],
supportsAspectRatio: true,
supportsResolution: true,
supportsSize: true,
supportsAudio: true,
},
},
...createGoogleVideoGenerationProviderMetadata(),
async generateVideo(req) {
if ((req.inputImages?.length ?? 0) > 1) {
throw new Error("Google video generation supports at most one input image.");

View File

@@ -1,28 +1 @@
import {
createWebSearchProviderContractFields,
type WebSearchProviderPlugin,
} from "openclaw/plugin-sdk/provider-web-search-config-contract";
export function createGeminiWebSearchProvider(): WebSearchProviderPlugin {
const credentialPath = "plugins.entries.google.config.webSearch.apiKey";
return {
id: "gemini",
label: "Gemini (Google Search)",
hint: "Requires Google Gemini API key · Google Search grounding",
onboardingScopes: ["text-inference"],
credentialLabel: "Google Gemini API key",
envVars: ["GEMINI_API_KEY"],
placeholder: "AIza...",
signupUrl: "https://aistudio.google.com/apikey",
docsUrl: "https://docs.openclaw.ai/tools/web",
autoDetectOrder: 20,
credentialPath,
...createWebSearchProviderContractFields({
credentialPath,
searchCredential: { type: "scoped", scopeId: "gemini" },
configuredCredential: { pluginId: "google" },
}),
createTool: () => null,
};
}
export { createGeminiWebSearchProvider } from "./src/gemini-web-search-provider.js";

View File

@@ -8,7 +8,6 @@ import {
} from "openclaw/plugin-sdk/account-resolution";
import { safeParseJsonWithSchema, safeParseWithSchema } from "openclaw/plugin-sdk/extension-shared";
import { isSecretRef } from "openclaw/plugin-sdk/secret-input";
import { normalizeOptionalString } from "openclaw/plugin-sdk/text-runtime";
import { z } from "zod";
import type { GoogleChatAccountConfig } from "./types.config.js";
@@ -28,6 +27,14 @@ const ENV_SERVICE_ACCOUNT = "GOOGLE_CHAT_SERVICE_ACCOUNT";
const ENV_SERVICE_ACCOUNT_FILE = "GOOGLE_CHAT_SERVICE_ACCOUNT_FILE";
const JsonRecordSchema = z.record(z.string(), z.unknown());
function normalizeOptionalString(value: unknown): string | undefined {
if (typeof value !== "string") {
return undefined;
}
const trimmed = value.trim();
return trimmed ? trimmed : undefined;
}
const {
listAccountIds: listGoogleChatAccountIds,
resolveDefaultAccountId: resolveDefaultGoogleChatAccountId,

View File

@@ -1,3 +1,4 @@
import path from "node:path";
import { beforeAll, beforeEach, describe, expect, it, vi } from "vitest";
const listEnabledGoogleChatAccounts = vi.hoisted(() => vi.fn());
@@ -9,7 +10,6 @@ const sendGoogleChatMessage = vi.hoisted(() => vi.fn());
const uploadGoogleChatAttachment = vi.hoisted(() => vi.fn());
const resolveGoogleChatOutboundSpace = vi.hoisted(() => vi.fn());
const getGoogleChatRuntime = vi.hoisted(() => vi.fn());
const loadOutboundMediaFromUrl = vi.hoisted(() => vi.fn());
vi.mock("./accounts.js", () => ({
listEnabledGoogleChatAccounts,
@@ -32,15 +32,6 @@ vi.mock("./targets.js", () => ({
resolveGoogleChatOutboundSpace,
}));
vi.mock("../runtime-api.js", async () => {
const actual = await vi.importActual<typeof import("../runtime-api.js")>("../runtime-api.js");
return {
...actual,
loadOutboundMediaFromUrl: (...args: Parameters<typeof actual.loadOutboundMediaFromUrl>) =>
(loadOutboundMediaFromUrl as unknown as typeof actual.loadOutboundMediaFromUrl)(...args),
};
});
let googlechatMessageActions: typeof import("./actions.js").googlechatMessageActions;
describe("googlechat message actions", () => {
@@ -161,11 +152,9 @@ describe("googlechat message actions", () => {
config: { mediaMaxMb: 5 },
});
resolveGoogleChatOutboundSpace.mockResolvedValue("spaces/BBB");
loadOutboundMediaFromUrl.mockResolvedValue({
buffer: Buffer.from("local-bytes"),
fileName: "local.txt",
contentType: "text/plain",
});
const localRoot = "/tmp/googlechat-action-test";
const localPath = path.join(localRoot, "local.md");
const readFile = vi.fn(async () => Buffer.from("local-bytes"));
getGoogleChatRuntime.mockReturnValue({
channel: {
media: {
@@ -187,23 +176,22 @@ describe("googlechat message actions", () => {
action: "upload-file",
params: {
to: "spaces/BBB",
path: "/tmp/local.txt",
path: localPath,
message: "notes",
filename: "renamed.txt",
},
cfg: {},
accountId: "default",
mediaLocalRoots: ["/tmp"],
mediaLocalRoots: [localRoot],
mediaReadFile: readFile,
} as never);
expect(loadOutboundMediaFromUrl).toHaveBeenCalledWith(
"/tmp/local.txt",
expect.objectContaining({ mediaLocalRoots: ["/tmp"] }),
);
expect(readFile).toHaveBeenCalledWith(localPath);
expect(uploadGoogleChatAttachment).toHaveBeenCalledWith(
expect.objectContaining({
space: "spaces/BBB",
filename: "renamed.txt",
buffer: Buffer.from("local-bytes"),
}),
);
expect(sendGoogleChatMessage).toHaveBeenCalledWith(

View File

@@ -1,17 +1,17 @@
import type {
ChannelMessageActionAdapter,
ChannelMessageActionName,
OpenClawConfig,
} from "../runtime-api.js";
import {
createActionGate,
extractToolSend,
jsonResult,
loadOutboundMediaFromUrl,
readNumberParam,
readReactionParams,
readStringParam,
} from "../runtime-api.js";
} from "openclaw/plugin-sdk/channel-actions";
import type {
ChannelMessageActionAdapter,
ChannelMessageActionName,
} from "openclaw/plugin-sdk/channel-contract";
import type { OpenClawConfig } from "openclaw/plugin-sdk/config-runtime";
import { loadOutboundMediaFromUrl } from "openclaw/plugin-sdk/outbound-media";
import { extractToolSend } from "openclaw/plugin-sdk/tool-send";
import { listEnabledGoogleChatAccounts, resolveGoogleChatAccount } from "./accounts.js";
import {
createGoogleChatReaction,

View File

@@ -1,6 +1,6 @@
import crypto from "node:crypto";
import { formatErrorMessage } from "openclaw/plugin-sdk/error-runtime";
import { fetchWithSsrFGuard } from "../runtime-api.js";
import { fetchWithSsrFGuard } from "openclaw/plugin-sdk/ssrf-runtime";
import type { ResolvedGoogleChatAccount } from "./accounts.js";
import { getGoogleChatAccessToken } from "./auth.js";
import type { GoogleChatReaction } from "./types.js";

View File

@@ -1,5 +1,5 @@
import { GoogleAuth, OAuth2Client } from "google-auth-library";
import { normalizeLowercaseStringOrEmpty } from "openclaw/plugin-sdk/text-runtime";
import { fetchWithSsrFGuard } from "../runtime-api.js";
import type { ResolvedGoogleChatAccount } from "./accounts.js";
const CHAT_SCOPE = "https://www.googleapis.com/auth/chat.bot";
@@ -16,6 +16,10 @@ const verifyClient = new OAuth2Client();
let cachedCerts: { fetchedAt: number; certs: Record<string, string> } | null = null;
function normalizeLowercaseStringOrEmpty(value: unknown): string {
return typeof value === "string" ? value.trim().toLowerCase() : "";
}
function buildAuthKey(account: ResolvedGoogleChatAccount): string {
if (account.credentialsFile) {
return `file:${account.credentialsFile}`;
@@ -80,13 +84,20 @@ async function fetchChatCerts(): Promise<Record<string, string>> {
if (cachedCerts && now - cachedCerts.fetchedAt < 10 * 60 * 1000) {
return cachedCerts.certs;
}
const res = await fetch(CHAT_CERTS_URL);
if (!res.ok) {
throw new Error(`Failed to fetch Chat certs (${res.status})`);
const { response, release } = await fetchWithSsrFGuard({
url: CHAT_CERTS_URL,
auditContext: "googlechat.auth.certs",
});
try {
if (!response.ok) {
throw new Error(`Failed to fetch Chat certs (${response.status})`);
}
const certs = (await response.json()) as Record<string, string>;
cachedCerts = { fetchedAt: now, certs };
return certs;
} finally {
await release();
}
const certs = (await res.json()) as Record<string, string>;
cachedCerts = { fetchedAt: now, certs };
return certs;
}
export type GoogleChatAudienceType = "app-url" | "project-number";

View File

@@ -1,11 +1,11 @@
import { createLazyRuntimeNamedExport } from "openclaw/plugin-sdk/lazy-runtime";
import type { ChannelAccountSnapshot } from "openclaw/plugin-sdk/status-helpers";
import {
createAccountStatusSink,
runPassiveAccountLifecycle,
type OpenClawConfig,
type ResolvedGoogleChatAccount,
} from "./channel.deps.runtime.js";
} from "openclaw/plugin-sdk/channel-lifecycle";
import type { OpenClawConfig } from "openclaw/plugin-sdk/config-runtime";
import { createLazyRuntimeNamedExport } from "openclaw/plugin-sdk/lazy-runtime";
import type { ChannelAccountSnapshot } from "openclaw/plugin-sdk/status-helpers";
import type { ResolvedGoogleChatAccount } from "./accounts.js";
import type { GoogleChatRuntimeEnv } from "./monitor-types.js";
const loadGoogleChatChannelRuntime = createLazyRuntimeNamedExport(

View File

@@ -18,6 +18,7 @@ import {
import type { ResolvedGoogleChatAccount } from "./accounts.js";
import { sendGoogleChatMessage } from "./api.js";
import type { GoogleChatCoreRuntime } from "./monitor-types.js";
import { isSenderAllowed } from "./sender-allow.js";
import type { GoogleChatAnnotation, GoogleChatMessage, GoogleChatSpace } from "./types.js";
function normalizeUserId(raw?: string | null): string {
@@ -28,42 +29,7 @@ function normalizeUserId(raw?: string | null): string {
return normalizeLowercaseStringOrEmpty(trimmed.replace(/^users\//i, ""));
}
function isEmailLike(value: string): boolean {
// Keep this intentionally loose; allowlists are user-provided config.
return value.includes("@");
}
export function isSenderAllowed(
senderId: string,
senderEmail: string | undefined,
allowFrom: string[],
allowNameMatching = false,
) {
if (allowFrom.includes("*")) {
return true;
}
const normalizedSenderId = normalizeUserId(senderId);
const normalizedEmail = normalizeLowercaseStringOrEmpty(senderEmail ?? "");
return allowFrom.some((entry) => {
const normalized = normalizeLowercaseStringOrEmpty(entry);
if (!normalized) {
return false;
}
// Accept `googlechat:<id>` but treat `users/...` as an *ID* only (deprecated `users/<email>`).
const withoutPrefix = normalized.replace(/^(googlechat|google-chat|gchat):/i, "");
if (withoutPrefix.startsWith("users/")) {
return normalizeUserId(withoutPrefix) === normalizedSenderId;
}
// Raw email allowlist entries are a break-glass override.
if (allowNameMatching && normalizedEmail && isEmailLike(withoutPrefix)) {
return withoutPrefix === normalizedEmail;
}
return withoutPrefix.replace(/^users\//i, "") === normalizedSenderId;
});
}
export { isSenderAllowed } from "./sender-allow.js";
type GoogleChatGroupEntry = {
requireMention?: boolean;

View File

@@ -0,0 +1,55 @@
import type { IncomingMessage, ServerResponse } from "node:http";
import { createWebhookInFlightLimiter } from "openclaw/plugin-sdk/webhook-request-guards";
import { registerWebhookTargetWithPluginRoute } from "openclaw/plugin-sdk/webhook-targets";
import type { WebhookTarget } from "./monitor-types.js";
import { createGoogleChatWebhookRequestHandler } from "./monitor-webhook.js";
import type { GoogleChatEvent } from "./types.js";
type ProcessGoogleChatEvent = (event: GoogleChatEvent, target: WebhookTarget) => Promise<void>;
const webhookTargets = new Map<string, WebhookTarget[]>();
const webhookInFlightLimiter = createWebhookInFlightLimiter();
let processGoogleChatEvent: ProcessGoogleChatEvent = async () => {};
export function setGoogleChatWebhookEventProcessor(processEvent: ProcessGoogleChatEvent): void {
processGoogleChatEvent = processEvent;
}
const googleChatWebhookRequestHandler = createGoogleChatWebhookRequestHandler({
webhookTargets,
webhookInFlightLimiter,
processEvent: async (event, target) => {
await processGoogleChatEvent(event, target);
},
});
export function registerGoogleChatWebhookTarget(target: WebhookTarget): () => void {
return registerWebhookTargetWithPluginRoute({
targetsByPath: webhookTargets,
target,
route: {
auth: "plugin",
match: "exact",
pluginId: "googlechat",
source: "googlechat-webhook",
accountId: target.account.accountId,
log: target.runtime.log,
handler: async (req, res) => {
const handled = await handleGoogleChatWebhookRequest(req, res);
if (!handled && !res.headersSent) {
res.statusCode = 404;
res.setHeader("Content-Type", "text/plain; charset=utf-8");
res.end("Not Found");
}
},
},
}).unregister;
}
export async function handleGoogleChatWebhookRequest(
req: IncomingMessage,
res: ServerResponse,
): Promise<boolean> {
return await googleChatWebhookRequestHandler(req, res);
}

View File

@@ -1,7 +1,7 @@
import type { OpenClawConfig } from "openclaw/plugin-sdk/core";
import type { ResolvedGoogleChatAccount } from "./accounts.js";
import type { GoogleChatAudienceType } from "./auth.js";
import { getGoogleChatRuntime } from "./runtime.js";
import type { getGoogleChatRuntime } from "./runtime.js";
export type GoogleChatRuntimeEnv = {
log?: (message: string) => void;

View File

@@ -8,8 +8,11 @@ const resolveWebhookTargetWithAuthOrReject = vi.hoisted(() => vi.fn());
const withResolvedWebhookRequestPipeline = vi.hoisted(() => vi.fn());
const verifyGoogleChatRequest = vi.hoisted(() => vi.fn());
vi.mock("../runtime-api.js", () => ({
vi.mock("openclaw/plugin-sdk/webhook-request-guards", () => ({
readJsonWebhookBodyOrReject,
}));
vi.mock("openclaw/plugin-sdk/webhook-targets", () => ({
resolveWebhookTargetWithAuthOrReject,
withResolvedWebhookRequestPipeline,
}));

View File

@@ -1,11 +1,10 @@
import type { IncomingMessage, ServerResponse } from "node:http";
import { normalizeLowercaseStringOrEmpty } from "openclaw/plugin-sdk/text-runtime";
import type { WebhookInFlightLimiter } from "openclaw/plugin-sdk/webhook-request-guards";
import { readJsonWebhookBodyOrReject } from "openclaw/plugin-sdk/webhook-request-guards";
import {
readJsonWebhookBodyOrReject,
resolveWebhookTargetWithAuthOrReject,
withResolvedWebhookRequestPipeline,
type WebhookInFlightLimiter,
} from "../runtime-api.js";
} from "openclaw/plugin-sdk/webhook-targets";
import { verifyGoogleChatRequest } from "./auth.js";
import type { WebhookTarget } from "./monitor-types.js";
import type {
@@ -15,6 +14,10 @@ import type {
GoogleChatUser,
} from "./types.js";
function normalizeLowercaseStringOrEmpty(value: unknown): string {
return typeof value === "string" ? value.trim().toLowerCase() : "";
}
function extractBearerToken(header: unknown): string {
const authHeader = Array.isArray(header)
? typeof header[0] === "string"

View File

@@ -1,4 +1,3 @@
import type { IncomingMessage, ServerResponse } from "node:http";
import {
deliverTextOrMediaReply,
resolveSendableOutboundReplyParts,
@@ -7,8 +6,6 @@ import { normalizeOptionalLowercaseString } from "openclaw/plugin-sdk/text-runti
import type { OpenClawConfig } from "../runtime-api.js";
import {
createChannelReplyPipeline,
createWebhookInFlightLimiter,
registerWebhookTargetWithPluginRoute,
resolveInboundRouteEnvelopeBuilderWithRuntime,
resolveWebhookPath,
} from "../runtime-api.js";
@@ -21,27 +18,27 @@ import {
} from "./api.js";
import { type GoogleChatAudienceType } from "./auth.js";
import { applyGoogleChatInboundAccessPolicy, isSenderAllowed } from "./monitor-access.js";
import {
handleGoogleChatWebhookRequest,
registerGoogleChatWebhookTarget,
setGoogleChatWebhookEventProcessor,
} from "./monitor-routing.js";
import type {
GoogleChatCoreRuntime,
GoogleChatMonitorOptions,
GoogleChatRuntimeEnv,
WebhookTarget,
} from "./monitor-types.js";
import { createGoogleChatWebhookRequestHandler } from "./monitor-webhook.js";
import { getGoogleChatRuntime } from "./runtime.js";
import type { GoogleChatAttachment, GoogleChatEvent } from "./types.js";
export type { GoogleChatMonitorOptions, GoogleChatRuntimeEnv } from "./monitor-types.js";
export {
handleGoogleChatWebhookRequest,
registerGoogleChatWebhookTarget,
} from "./monitor-routing.js";
export { isSenderAllowed };
const webhookTargets = new Map<string, WebhookTarget[]>();
const webhookInFlightLimiter = createWebhookInFlightLimiter();
const googleChatWebhookRequestHandler = createGoogleChatWebhookRequestHandler({
webhookTargets,
webhookInFlightLimiter,
processEvent: async (event, target) => {
await processGoogleChatEvent(event, target);
},
});
setGoogleChatWebhookEventProcessor(processGoogleChatEvent);
function logVerbose(core: GoogleChatCoreRuntime, runtime: GoogleChatRuntimeEnv, message: string) {
if (core.logging.shouldLogVerbose()) {
@@ -49,29 +46,6 @@ function logVerbose(core: GoogleChatCoreRuntime, runtime: GoogleChatRuntimeEnv,
}
}
export function registerGoogleChatWebhookTarget(target: WebhookTarget): () => void {
return registerWebhookTargetWithPluginRoute({
targetsByPath: webhookTargets,
target,
route: {
auth: "plugin",
match: "exact",
pluginId: "googlechat",
source: "googlechat-webhook",
accountId: target.account.accountId,
log: target.runtime.log,
handler: async (req, res) => {
const handled = await handleGoogleChatWebhookRequest(req, res);
if (!handled && !res.headersSent) {
res.statusCode = 404;
res.setHeader("Content-Type", "text/plain; charset=utf-8");
res.end("Not Found");
}
},
},
}).unregister;
}
function normalizeAudienceType(value?: string | null): GoogleChatAudienceType | undefined {
const normalized = normalizeOptionalLowercaseString(value);
if (normalized === "app-url" || normalized === "app_url" || normalized === "app") {
@@ -87,13 +61,6 @@ function normalizeAudienceType(value?: string | null): GoogleChatAudienceType |
return undefined;
}
export async function handleGoogleChatWebhookRequest(
req: IncomingMessage,
res: ServerResponse,
): Promise<boolean> {
return await googleChatWebhookRequestHandler(req, res);
}
async function processGoogleChatEvent(event: GoogleChatEvent, target: WebhookTarget) {
const eventType = event.type ?? (event as { eventType?: string }).eventType;
if (eventType !== "MESSAGE") {

View File

@@ -7,7 +7,10 @@ import { createMockServerResponse } from "../../../test/helpers/plugins/mock-htt
import type { OpenClawConfig, PluginRuntime } from "../runtime-api.js";
import type { ResolvedGoogleChatAccount } from "./accounts.js";
import { verifyGoogleChatRequest } from "./auth.js";
import { handleGoogleChatWebhookRequest, registerGoogleChatWebhookTarget } from "./monitor.js";
import {
handleGoogleChatWebhookRequest,
registerGoogleChatWebhookTarget,
} from "./monitor-routing.js";
vi.mock("./auth.js", () => ({
verifyGoogleChatRequest: vi.fn(),

View File

@@ -0,0 +1,48 @@
function normalizeLowercaseStringOrEmpty(value: unknown): string {
return typeof value === "string" ? value.trim().toLowerCase() : "";
}
function normalizeUserId(raw?: string | null): string {
const trimmed = typeof raw === "string" ? raw.trim() : "";
if (!trimmed) {
return "";
}
return normalizeLowercaseStringOrEmpty(trimmed.replace(/^users\//i, ""));
}
function isEmailLike(value: string): boolean {
// Keep this intentionally loose; allowlists are user-provided config.
return value.includes("@");
}
export function isSenderAllowed(
senderId: string,
senderEmail: string | undefined,
allowFrom: string[],
allowNameMatching = false,
) {
if (allowFrom.includes("*")) {
return true;
}
const normalizedSenderId = normalizeUserId(senderId);
const normalizedEmail = normalizeLowercaseStringOrEmpty(senderEmail ?? "");
return allowFrom.some((entry) => {
const normalized = normalizeLowercaseStringOrEmpty(entry);
if (!normalized) {
return false;
}
// Accept `googlechat:<id>` but treat `users/...` as an *ID* only (deprecated `users/<email>`).
const withoutPrefix = normalized.replace(/^(googlechat|google-chat|gchat):/i, "");
if (withoutPrefix.startsWith("users/")) {
return normalizeUserId(withoutPrefix) === normalizedSenderId;
}
// Raw email allowlist entries are a break-glass override.
if (allowNameMatching && normalizedEmail && isEmailLike(withoutPrefix)) {
return withoutPrefix === normalizedEmail;
}
return withoutPrefix.replace(/^users\//i, "") === normalizedSenderId;
});
}

View File

@@ -11,10 +11,6 @@ import {
type ChannelSetupDmPolicy,
type ChannelSetupWizard,
} from "openclaw/plugin-sdk/setup";
import {
normalizeOptionalString,
normalizeStringifiedOptionalString,
} from "openclaw/plugin-sdk/text-runtime";
import { resolveDefaultGoogleChatAccountId, resolveGoogleChatAccount } from "./accounts.js";
const channel = "googlechat" as const;
@@ -23,6 +19,24 @@ const ENV_SERVICE_ACCOUNT_FILE = "GOOGLE_CHAT_SERVICE_ACCOUNT_FILE";
const USE_ENV_FLAG = "__googlechatUseEnv";
const AUTH_METHOD_FLAG = "__googlechatAuthMethod";
function normalizeOptionalString(value: unknown): string | undefined {
if (typeof value !== "string") {
return undefined;
}
const trimmed = value.trim();
return trimmed ? trimmed : undefined;
}
function normalizeStringifiedOptionalString(value: unknown): string | undefined {
if (typeof value === "string") {
return normalizeOptionalString(value);
}
if (typeof value === "number" || typeof value === "boolean" || typeof value === "bigint") {
return normalizeOptionalString(String(value));
}
return undefined;
}
const promptAllowFrom = createPromptParsedAllowFromForAccount({
defaultAccountId: resolveDefaultGoogleChatAccountId,
message: "Google Chat allowFrom (users/<id> or raw email; avoid users/<email>)",

View File

@@ -11,14 +11,14 @@ import {
expectLifecyclePatch,
expectPendingUntilAbort,
startAccountAndTrackLifecycle,
waitForStartedMocks,
} from "../../../test/helpers/plugins/start-account-lifecycle.js";
import type { OpenClawConfig } from "../runtime-api.js";
import { resolveGoogleChatAccount, type ResolvedGoogleChatAccount } from "./accounts.js";
import {
listGoogleChatAccountIds,
resolveGoogleChatAccount,
resolveDefaultGoogleChatAccountId,
} from "./channel.deps.runtime.js";
type ResolvedGoogleChatAccount,
} from "./accounts.js";
import { startGoogleChatGatewayAccount } from "./gateway.js";
import { googlechatSetupAdapter } from "./setup-core.js";
import { googlechatSetupWizard } from "./setup-surface.js";
@@ -27,13 +27,16 @@ const hoisted = vi.hoisted(() => ({
startGoogleChatMonitor: vi.fn(),
}));
vi.mock("./monitor.js", async () => {
const actual = await vi.importActual<typeof import("./monitor.js")>("./monitor.js");
return {
...actual,
vi.mock("./channel.runtime.js", () => ({
googleChatChannelRuntime: {
resolveGoogleChatWebhookPath: ({
account,
}: {
account: { config: { webhookPath?: string } };
}) => account.config.webhookPath ?? "/googlechat",
startGoogleChatMonitor: hoisted.startGoogleChatMonitor,
};
});
},
}));
const googlechatSetupPlugin = {
id: "googlechat",
@@ -65,6 +68,16 @@ function buildAccount(): ResolvedGoogleChatAccount {
};
}
async function waitForGoogleChatMonitorStarted() {
for (let attempt = 0; attempt < 10; attempt += 1) {
if (hoisted.startGoogleChatMonitor.mock.calls.length === 1) {
return;
}
await new Promise<void>((resolve) => setImmediate(resolve));
}
expect(hoisted.startGoogleChatMonitor).toHaveBeenCalledOnce();
}
describe("googlechat setup", () => {
afterEach(() => {
vi.clearAllMocks();
@@ -356,7 +369,7 @@ describe("googlechat setup", () => {
account: buildAccount(),
});
await expectPendingUntilAbort({
waitForStarted: waitForStartedMocks(hoisted.startGoogleChatMonitor),
waitForStarted: waitForGoogleChatMonitorStarted,
isSettled,
abort,
task,

View File

@@ -2,7 +2,7 @@ import { afterEach, describe, expect, it, vi } from "vitest";
import type { ResolvedGoogleChatAccount } from "./accounts.js";
import { downloadGoogleChatMedia, sendGoogleChatMessage } from "./api.js";
import { resolveGoogleChatGroupRequireMention } from "./group-policy.js";
import { isSenderAllowed } from "./monitor.js";
import { isSenderAllowed } from "./sender-allow.js";
import {
isGoogleChatSpaceTarget,
isGoogleChatUserTarget,
@@ -14,14 +14,13 @@ const mocks = vi.hoisted(() => ({
response: await fetch(params.url, params.init),
release: async () => {},
})),
verifySignedJwtWithCertsAsync: vi.fn(),
verifyIdToken: vi.fn(),
getGoogleChatAccessToken: vi.fn().mockResolvedValue("token"),
}));
vi.mock("../runtime-api.js", async () => {
const actual = await vi.importActual<typeof import("../runtime-api.js")>("../runtime-api.js");
vi.mock("openclaw/plugin-sdk/ssrf-runtime", () => {
return {
...actual,
fetchWithSsrFGuard: mocks.fetchWithSsrFGuard,
};
});
@@ -30,6 +29,7 @@ vi.mock("google-auth-library", () => ({
GoogleAuth: function GoogleAuth() {},
OAuth2Client: class {
verifyIdToken = mocks.verifyIdToken;
verifySignedJwtWithCertsAsync = mocks.verifySignedJwtWithCertsAsync;
},
}));
@@ -295,4 +295,34 @@ describe("verifyGoogleChatRequest", () => {
reason: "unexpected add-on principal: principal-2",
});
});
it("fetches Chat certs through the guarded fetch for project-number tokens", async () => {
const release = vi.fn();
mocks.fetchWithSsrFGuard.mockClear();
mocks.fetchWithSsrFGuard.mockResolvedValueOnce({
response: new Response(JSON.stringify({ "kid-1": "cert-body" }), { status: 200 }),
release,
});
mocks.verifySignedJwtWithCertsAsync.mockReset().mockResolvedValue(undefined);
await expect(
verifyGoogleChatRequest({
bearer: "token",
audienceType: "project-number",
audience: "123456789",
}),
).resolves.toEqual({ ok: true });
expect(mocks.fetchWithSsrFGuard).toHaveBeenCalledWith({
url: "https://www.googleapis.com/service_accounts/v1/metadata/x509/chat@system.gserviceaccount.com",
auditContext: "googlechat.auth.certs",
});
expect(mocks.verifySignedJwtWithCertsAsync).toHaveBeenCalledWith(
"token",
{ "kid-1": "cert-body" },
"123456789",
["chat@system.gserviceaccount.com"],
);
expect(release).toHaveBeenCalledOnce();
});
});

View File

@@ -1,7 +1,10 @@
import { normalizeLowercaseStringOrEmpty } from "openclaw/plugin-sdk/text-runtime";
import type { ResolvedGoogleChatAccount } from "./accounts.js";
import { findGoogleChatDirectMessage } from "./api.js";
function normalizeLowercaseStringOrEmpty(value: unknown): string {
return typeof value === "string" ? value.trim().toLowerCase() : "";
}
export function normalizeGoogleChatTarget(raw?: string | null): string | undefined {
const trimmed = raw?.trim();
if (!trimmed) {

View File

@@ -1,9 +1,11 @@
export {
resolveIMessageAttachmentRoots as resolveInboundAttachmentRoots,
resolveIMessageRemoteAttachmentRoots as resolveRemoteInboundAttachmentRoots,
} from "./src/media-contract.js";
export {
DEFAULT_IMESSAGE_ATTACHMENT_ROOTS,
resolveIMessageAttachmentRoots as resolveInboundAttachmentRoots,
resolveIMessageAttachmentRoots,
resolveIMessageRemoteAttachmentRoots as resolveRemoteInboundAttachmentRoots,
resolveIMessageRemoteAttachmentRoots,
} from "./src/media-contract.js";
} from "./media-contract-api.js";
export {
__testing as imessageConversationBindingTesting,
createIMessageConversationBindingManager,
} from "./src/conversation-bindings.js";

View File

@@ -0,0 +1,7 @@
export {
DEFAULT_IMESSAGE_ATTACHMENT_ROOTS,
resolveIMessageAttachmentRoots,
resolveIMessageAttachmentRoots as resolveInboundAttachmentRoots,
resolveIMessageRemoteAttachmentRoots,
resolveIMessageRemoteAttachmentRoots as resolveRemoteInboundAttachmentRoots,
} from "./src/media-contract.js";

View File

@@ -8,6 +8,7 @@ import {
type ProviderRuntimeModel,
} from "openclaw/plugin-sdk/plugin-entry";
import { CUSTOM_LOCAL_AUTH_MARKER } from "openclaw/plugin-sdk/provider-auth";
import { lmstudioMemoryEmbeddingProviderAdapter } from "./memory-embedding-adapter.js";
import {
LMSTUDIO_DEFAULT_API_KEY_ENV_VAR,
LMSTUDIO_LOCAL_API_KEY_PLACEHOLDER,
@@ -52,6 +53,7 @@ export default definePluginEntry({
name: "LM Studio Provider",
description: "Bundled LM Studio provider plugin",
register(api: OpenClawPluginApi) {
api.registerMemoryEmbeddingProvider(lmstudioMemoryEmbeddingProviderAdapter);
api.registerProvider({
id: PROVIDER_ID,
label: "LM Studio",

View File

@@ -0,0 +1,35 @@
import {
sanitizeEmbeddingCacheHeaders,
type MemoryEmbeddingProviderAdapter,
} from "openclaw/plugin-sdk/memory-core-host-engine-embeddings";
import {
createLmstudioEmbeddingProvider,
DEFAULT_LMSTUDIO_EMBEDDING_MODEL,
} from "./src/embedding-provider.js";
export const lmstudioMemoryEmbeddingProviderAdapter: MemoryEmbeddingProviderAdapter = {
id: "lmstudio",
defaultModel: DEFAULT_LMSTUDIO_EMBEDDING_MODEL,
transport: "remote",
authProviderId: "lmstudio",
allowExplicitWhenConfiguredAuto: true,
create: async (options) => {
const { provider, client } = await createLmstudioEmbeddingProvider({
...options,
provider: "lmstudio",
fallback: "none",
});
return {
provider,
runtime: {
id: "lmstudio",
cacheKeyData: {
provider: "lmstudio",
baseUrl: client.baseUrl,
model: client.model,
headers: sanitizeEmbeddingCacheHeaders(client.headers, ["authorization"]),
},
},
};
},
};

View File

@@ -21,6 +21,9 @@
"groupHint": "Self-hosted open-weight models"
}
],
"contracts": {
"memoryEmbeddingProviders": ["lmstudio"]
},
"configSchema": {
"type": "object",
"additionalProperties": false,

View File

@@ -1,20 +1,21 @@
import { formatErrorMessage } from "../../infra/errors.js";
import type { SsrFPolicy } from "../../infra/net/ssrf.js";
import { createSubsystemLogger } from "../../logging/subsystem.js";
import { createSubsystemLogger } from "openclaw/plugin-sdk/logging-core";
import {
buildRemoteBaseUrlPolicy,
createRemoteEmbeddingProvider,
normalizeEmbeddingModelWithPrefixes,
type MemoryEmbeddingProvider,
type MemoryEmbeddingProviderCreateOptions,
} from "openclaw/plugin-sdk/memory-core-host-engine-embeddings";
import { resolveMemorySecretInputString } from "openclaw/plugin-sdk/memory-core-host-secret";
import { formatErrorMessage, type SsrFPolicy } from "openclaw/plugin-sdk/ssrf-runtime";
import { LMSTUDIO_DEFAULT_EMBEDDING_MODEL, LMSTUDIO_PROVIDER_ID } from "./defaults.js";
import { ensureLmstudioModelLoaded } from "./models.fetch.js";
import { resolveLmstudioInferenceBase } from "./models.js";
import {
buildLmstudioAuthHeaders,
ensureLmstudioModelLoaded,
LMSTUDIO_DEFAULT_EMBEDDING_MODEL,
LMSTUDIO_PROVIDER_ID,
resolveLmstudioInferenceBase,
resolveLmstudioProviderHeaders,
resolveLmstudioRuntimeApiKey,
} from "../../plugin-sdk/lmstudio-runtime.js";
import { normalizeEmbeddingModelWithPrefixes } from "./embeddings-model-normalize.js";
import { createRemoteEmbeddingProvider } from "./embeddings-remote-provider.js";
import type { EmbeddingProvider, EmbeddingProviderOptions } from "./embeddings.types.js";
import { buildRemoteBaseUrlPolicy } from "./remote-http.js";
import { resolveMemorySecretInputString } from "./secret-input.js";
} from "./runtime.js";
const log = createSubsystemLogger("memory/embeddings");
@@ -47,7 +48,7 @@ function hasAuthorizationHeader(headers: Record<string, string> | undefined): bo
/** Resolves API key (real or synthetic placeholder) from runtime/provider auth config. */
async function resolveLmstudioApiKey(
options: EmbeddingProviderOptions,
options: MemoryEmbeddingProviderCreateOptions,
): Promise<string | undefined> {
try {
return await resolveLmstudioRuntimeApiKey({
@@ -65,8 +66,8 @@ async function resolveLmstudioApiKey(
/** Creates the LM Studio embedding provider client and preloads the target model before return. */
export async function createLmstudioEmbeddingProvider(
options: EmbeddingProviderOptions,
): Promise<{ provider: EmbeddingProvider; client: LmstudioEmbeddingClient }> {
options: MemoryEmbeddingProviderCreateOptions,
): Promise<{ provider: MemoryEmbeddingProvider; client: LmstudioEmbeddingClient }> {
const providerConfig = options.config.models?.providers?.lmstudio;
const providerBaseUrl = providerConfig?.baseUrl?.trim();
const isFallbackActivation = options.fallback === "lmstudio" && options.provider !== "lmstudio";

Some files were not shown because too many files have changed in this diff Show More