Commit Graph

3788 Commits

Author SHA1 Message Date
Peter Steinberger
00399d6c75 test(release): repair beta validation blockers 2026-06-01 06:27:02 +01:00
Vincent Koc
296cd8c912 fix(plugin-sdk): isolate provider catalog projection failures (#88767)
* fix(plugin-sdk): isolate provider catalog projection failures

* fix(plugin-sdk): share safe provider catalog projection

* fix(cron): preserve raw null clear schema

* fix(plugin-sdk): copy provider catalog model rows safely

* fix(plugin-sdk): keep id-only catalog models

* fix(plugin-sdk): require readable provider catalog base url

* fix(ci): satisfy cron and matrix lint checks

---------

Co-authored-by: Peter Steinberger <steipete@gmail.com>
2026-06-01 00:35:19 -04:00
Peter Steinberger
5b79e81569 fix: harden CLI and plugin edge cases (#88896)
* fix: harden CLI and plugin edge cases

* fix: preserve explicit TTS provider credentials

* fix: preserve direct TTS credentials

* fix: type TTS credential hydration config

* fix: preserve scoped TTS channel credentials

* fix: pin hydrated TTS runtime config

* fix: satisfy TTS hydration lint

* fix: preserve inherited TTS provider keys

* fix: read resolved TTS provider keys
2026-06-01 00:30:12 -04:00
Ted Li
6cb06f5fbc fix(reply): preserve sessions_send external routes (#88803)
* fix(reply): preserve sessions_send external routes

* fix(reply): preserve inherited route thread ids

* fix(reply): keep sessions_send delivery single-owner

* fix(reply): satisfy dispatch route lint

* fix(reply): preserve inherited ACP route metadata

* test(reply): type inherited route event assertions

* test(ci): satisfy current lint rules

* fix(reply): avoid stale inherited route threads

* fix(reply): trust explicit inherited route threads

* fix(reply): require trusted route thread sources

---------

Co-authored-by: Peter Steinberger <steipete@gmail.com>
2026-05-31 23:43:05 -04:00
Vincent Koc
be967545c5 fix(plugins): fail closed on trusted policy errors
Fail closed when bundled trusted tool policy registry, registration, owner id, evaluation, or decision reads fail, so malformed trusted-policy state cannot crash diagnostics or accidentally allow a tool call.

Route before-tool-call diagnostics through guarded trusted-policy readers and keep healthy no-op policy behavior unchanged.

Add focused host-hook contract and before-tool-call e2e coverage for the new fail-closed paths.

PR: #88394
2026-05-31 22:57:38 -04:00
Andy Ye
921598442a fix(hooks): expose inbound reply metadata before dispatch
Fixes #88521.

Expose finalized inbound reply metadata on plugin-visible hook payloads so before_dispatch and message hooks can implement reply-aware behavior without channel-specific workarounds.
2026-05-31 22:15:17 -04:00
Dallin Romney
4b56c44c02 test: consolidate plugin registration contracts (#88824) 2026-05-31 19:04:53 -07:00
Vincent Koc
570e2db252 fix(plugins): isolate cached tool runtime siblings 2026-05-31 21:05:23 -04:00
Vincent Koc
53990d5bbf fix(plugins): isolate web provider factory failures (#88807) 2026-05-31 21:04:18 -04:00
Brian
fda5254e99 fix: preserve npm plugin root on blocked install (#77237)
Preserve the active per-plugin managed npm project when npm-backed install validation blocks a candidate after npm has already mutated local state.

This snapshots package.json, package-lock.json, and node_modules before managed npm installs, restores that exact project state on failed validation, and rolls back staged npm-pack archives so blocked pack installs do not leave candidate debris.

Validation:
- OPENCLAW_VITEST_MAX_WORKERS=1 node scripts/run-vitest.mjs src/plugins/install.npm-spec.test.ts
- pnpm tsgo:core && pnpm tsgo:core:test
- node scripts/run-oxlint.mjs src/plugins/install.ts src/plugins/install.npm-spec.test.ts
- .agents/skills/autoreview/scripts/autoreview --mode local
- GitHub CI 26729255950
- Crabbox run_26e9f9f7591c

Thanks @zhuisDEV.

Co-authored-by: Brian <95547369+zhuisDEV@users.noreply.github.com>
2026-05-31 20:59:32 -04:00
Peter Steinberger
5443baa852 Persist plugin install index in SQLite (#88794)
* refactor: persist plugin install index in sqlite

* fix: merge legacy plugin index records into sqlite

* test: update plugin index sqlite fixtures

* fix: migrate custom plugin install indexes

* test: update plugin index sentinel

* fix: exclude migrated plugin index archives

* fix: read post-upgrade plugin index from sqlite

* fix: migrate legacy plugin index before agent runs

* fix: respect disabled persisted plugin registry reads

* test: type plugin install record fixtures

* fix: simplify plugin index record reader type

* test: fix sqlite plugin index CI fallout

* test: mock provider normalization in agent command tests

# Conflicts:
#	src/commands/agent-command.test-mocks.ts

* build: remove unused ui three dependency
2026-05-31 20:51:33 -04:00
Peter Steinberger
27dde7a4d6 chore(lint): enable stricter error rules 2026-06-01 01:12:21 +01:00
Vincent Koc
160aad6fb3 fix(agents): preserve exact custom provider models 2026-06-01 01:50:30 +02:00
Vincent Koc
1b69e7a005 fix(plugin-sdk): keep llm core alias on source graph 2026-06-01 00:20:10 +01:00
Vincent Koc
6df3fd5730 fix(gateway): list commands from gateway plugin registry 2026-06-01 00:20:10 +01:00
Feelw00
01193dea26 fix: make task persistence failures explicit
Preserve task and TaskFlow durability by persisting before in-memory registry mutation and surfacing explicit persistence failures instead of reporting fake success.

Adds non-throwing try-create runtime helpers while keeping existing throwing public create APIs compatible. Maintainer follow-up keeps task/TaskFlow sync repair bounded, prevents split task/delivery-state writes, and keeps CI green on the current base.

Thanks @Feelw00.
2026-06-01 00:12:28 +01:00
Peter Steinberger
22cb7fb6b7 chore(lint): enable no-promise-executor-return 2026-05-31 23:06:13 +01:00
Peter Steinberger
b653d94918 chore(lint): enable no-useless-assignment 2026-05-31 22:40:48 +01:00
waterblue
03dec8bb3a fix(openai): avoid replay ids when Responses store is disabled
Avoid replaying prior OpenAI Responses reasoning/message/function-call item ids when the outgoing request disables store, while preserving encrypted reasoning and normalized summary arrays for stateless replay. Keep explicit store-enabled OpenAI wrapper paths opted into item-id replay, and cover shared/simple Responses, ChatGPT/Codex Responses, and GitHub Copilot sanitizer behavior.

Regression tests cover store-disabled id omission, encrypted reasoning preservation, idless Copilot reasoning replay, and direct builder payloads. Local proof included focused Vitest, broad lint, broad test-types, bundled-extension lint, plugin boundary checks, autoreview clean, and live OpenAI Responses gpt-5.5 proof.

Co-authored-by: hang <zhanghang02@gmail.com>
2026-05-31 22:17:32 +01:00
Peter Steinberger
b0679d1f13 refactor(channels): store inbound queues in SQLite 2026-05-31 21:15:29 +01:00
Peter Steinberger
2df95c0b10 chore(lint): enable no-misused-promises 2026-05-31 20:42:13 +01:00
Peter Steinberger
ccb50f89da fix(plugins): clarify loader failure guidance 2026-05-31 15:12:22 -04:00
Peter Steinberger
304e2c83c0 chore(lint): enable stricter oxlint rules 2026-05-31 18:59:02 +01:00
zhang-guiping
b6e9473e9f fix(auth): skip Anthropic API keys for usage status
Fixes #85124.

Anthropic standard API keys no longer resolve as provider usage auth for `openclaw status --usage`, so valid inference keys are not sent to Anthropic's OAuth usage endpoint and surfaced as misleading invalid bearer-token errors.

The provider usage-auth SDK result now has an explicit handled/no-token shape so provider hooks can suppress generic fallback without widening the OAuth helper contract. Docs, Plugin SDK API baseline, and extension package-boundary cache inputs were updated with the new contract.

Thanks @zhangguiping-xydt.

Proof:
- node scripts/run-vitest.mjs src/infra/provider-usage.auth.normalizes-keys.test.ts src/infra/provider-usage.auth.plugin.test.ts extensions/anthropic/index.test.ts
- pnpm plugin-sdk:api:check
- pnpm plugin-sdk:check-exports
- git diff --check origin/main...HEAD
- pnpm docs:list
- pnpm run test:extensions:package-boundary:compile
- autoreview clean: no accepted/actionable findings
- PR CI rollup green: 131 success, 22 skipped, 1 neutral, 0 failures

Co-authored-by: 张贵萍0668001030 <zhang.guiping@xydigit.com>
2026-05-31 18:26:03 +01:00
Peter Steinberger
33c246dbba refactor: move plugin state slices to sqlite
* refactor: move plugin state slices to sqlite

* fix: keep legacy plugin state migration out of runtime

* fix: add doctor migrations for plugin sqlite state

* fix: preserve teams feedback learning migration keys

* fix: merge teams legacy feedback learnings

* fix: guard doctor imports against plugin state caps

* fix: leave lossy teams learning filenames unmigrated

* fix: preserve teams feedback learning scope

* fix: load plugin doctor contracts from package dist

* fix: satisfy plugin state migration gates
2026-05-31 18:09:27 +01:00
Peter Steinberger
12d4dda1bb perf(plugins): avoid duplicate provider hook load probes
Avoid duplicate provider hook load probes.

Summary:
- Route provider hook-list resolution through the existing provider resolver skip path instead of pre-checking provider load state separately.
- Preserve the provider runtime in-flight/reentrant guard because existing tests prove it prevents cached misses and nested provider-load recursion.

Verification:
- node scripts/run-vitest.mjs src/plugins/providers.runtime.consult-current-snapshot.test.ts
- node scripts/run-vitest.mjs src/plugins/provider-runtime.test.ts
- node scripts/run-vitest.mjs src/plugins/providers.test.ts
- pnpm exec oxfmt --check src/plugins/providers.runtime.ts src/plugins/provider-hook-runtime.ts
- git diff --check
- pnpm changed:lanes --json
- autoreview --mode local --prompt-file /tmp/provider-hotpath-cleanup-review.md
- Live E2E: https://github.com/openclaw/openclaw/actions/runs/26718818705
2026-05-31 18:08:13 +01:00
Masato Hoshino
3ef02ca818 fix(plugins): reuse current metadata snapshot in provider hot paths
Refactor provider metadata lookup so hot paths consult the current process snapshot before falling back to a metadata load.

Centralize provider metadata lookup in the provider runtime and update the focused tests/mocks that exercise embedded-agent and provider loading paths.

Verification:
- node scripts/run-vitest.mjs src/plugins/providers.runtime.consult-current-snapshot.test.ts
- node scripts/run-vitest.mjs src/agents/embedded-agent-runner/run/attempt.cwd-split.test.ts
- node scripts/run-vitest.mjs src/plugins/providers.test.ts
- autoreview --mode branch --base origin/main
- CPU profile loop: current-snapshot resolve 0.459 us/call vs warm direct metadata load 131.493 us/call
- GitHub CI on 728bd53510

Co-authored-by: masatohoshino <g515hoshino@gmail.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-05-31 17:48:58 +01:00
Jerry-Xin
4e84d0eaa5 fix(auto-reply): track memory flush failure exhaustion
Add durable memoryFlush failure metadata and lifecycle events so provider failures during memory flush no longer leave a session with no recorded recovery state.

After three consecutive non-abort flush failures, mark the current compaction cycle as exhausted so later messages can proceed without deleting transcript history. Successful flushes clear the failure metadata, and plugin session-entry slot reservations now protect the new fields.

Release-note: memoryFlush sessions can now fail open after repeated provider-side flush failures instead of retrying indefinitely before normal replies.

Refs #85645

Co-authored-by: 忻役 <xinyi@mininglamp.com>
2026-05-31 17:47:12 +01:00
litang9
d446c26acb feat(deepseek): show provider balance in usage status
Show DeepSeek API-key account balance in status/auth-status usage surfaces by adding a summary-only provider usage snapshot path, a DeepSeek balance fetcher, SDK/docs coverage, and focused regression tests.

Maintainer verification accepted the additive provider-usage/status contract and the DeepSeek balance visibility boundary for authenticated status surfaces.

Proof:
- Live DeepSeek balance proof via 1Password-backed DEEPSEEK_API_KEY against https://api.deepseek.com/user/balance; key and balance amount redacted.
- GitHub CI run 26717953383 passed on the current head.
- Real behavior proof run 26718215605 passed after the PR body was refreshed.
- Local clean PR clone: git diff --check; node --max-old-space-size=8192 --import tsx scripts/generate-plugin-sdk-api-baseline.ts --check; node scripts/run-vitest.mjs run src/agents/bash-tools.exec.path.test.ts.

Co-authored-by: Alex Tang <tangli1987118@hotmail.com>
Co-authored-by: litang9 <141409885+litang9@users.noreply.github.com>
2026-05-31 17:35:41 +01:00
Peter Steinberger
7b78941ea5 refactor: clean up ACP package metadata and helpers (#88659)
* refactor: derive acp core package subpath maps

* refactor: split acp manager task and timeout helpers

* refactor: split acp translator presentation helpers

* fix: keep packaged acp core plugin aliases

* ci: split gateway control plane runtime shard
2026-05-31 15:53:14 +01:00
Peter Steinberger
82a0ba8c4c fix(plugins): remove redundant proxy assertion 2026-05-31 15:02:44 +01:00
Peter Steinberger
d99c824ac1 fix(plugins): delegate wrapped tool properties 2026-05-31 15:02:44 +01:00
Peter Steinberger
f62a22ce56 fix(plugins): preserve wrapped tool descriptors 2026-05-31 15:02:44 +01:00
Peter Steinberger
643633c1e5 fix(plugins): scope tool callbacks during materialization 2026-05-31 15:02:44 +01:00
Peter Steinberger
ec8cb8bcbf feat: add MCP code-mode namespace (#88636)
* feat: add MCP code-mode namespace

* fix: unblock mcp namespace ci gates
2026-05-31 15:02:19 +01:00
Peter Steinberger
7dea283756 refactor: expand acp core package (#88618)
* refactor: expand acp core package

* chore: drop acp core package symlink

* fix: keep acp core dependency graph stable

* fix: add acp core tsconfig subpaths

* fix: sync acp core boundary path artifacts

* fix: use kysely for cron run-log queries

* fix: resolve acp core subpaths in loaders
2026-05-31 14:48:57 +01:00
Vincent Koc
cc290050b4 fix(doctor): diagnose malformed provider catalogs
Move malformed static provider catalog diagnostics into `openclaw doctor` instead of adding fallback behavior to runtime projection.

Doctor now validates full provider registrations for malformed static catalog hooks, result containers, provider keys, model arrays, model iteration, model ids/names, invalid catalog order, and proxy/access errors. Runtime unified text provider catalog projection remains strict on the typed provider catalog contract.

Verification:
- `node scripts/run-vitest.mjs src/flows/doctor-core-checks.runtime.test.ts src/flows/doctor-core-checks.test.ts src/flows/doctor-health-contributions.test.ts src/flows/doctor-health-conversion-plan.test.ts src/plugin-sdk/provider-entry.test.ts`
- `node_modules/.bin/oxfmt --check src/plugins/provider-catalog-unified-text.ts src/flows/doctor-core-checks.ts src/flows/doctor-core-checks.test.ts src/flows/doctor-core-checks.runtime.ts src/flows/doctor-core-checks.runtime.test.ts src/flows/doctor-health-contributions.ts src/flows/doctor-health-contributions.test.ts src/flows/doctor-health-conversion-plan.ts src/flows/doctor-health-conversion-plan.test.ts`
- `node scripts/run-oxlint.mjs src/flows/doctor-core-checks.runtime.ts src/flows/doctor-core-checks.runtime.test.ts src/plugins/provider-catalog-unified-text.ts`
- `pnpm tsgo:test`
- `git diff --check origin/main...HEAD`
- `.agents/skills/autoreview/scripts/autoreview --mode branch --base origin/main --prompt-file /tmp/provider-catalog-doctor-review-context.txt`
- GitHub PR checks green on head `876fdda5a352b0f15bfbe2abe9be43ebada7c596`

Co-authored-by: Vincent Koc <vincentkoc@ieee.org>
2026-05-31 14:48:15 +01:00
Peter Steinberger
7423e9cb66 refactor(openai): confine legacy codex repair to doctor
Confine retired OpenAI Codex identifiers to doctor repair and migration paths while keeping runtime OpenAI surfaces canonical.\n\nProof: focused Vitest; autoreview clean; AWS Crabbox check:changed run_3789cbe12413 (cbx_2c88b700810b) passed.
2026-05-31 14:03:17 +01:00
Peter Steinberger
f83886c12d chore(lint): trim remaining suppressions 2026-05-31 13:01:19 +01:00
Peter Steinberger
63c6252389 test(release): stabilize beta validation after rebase 2026-05-31 13:00:09 +01:00
Peter Steinberger
00d17e9df7 refactor: make OpenAI Codex legacy doctor-only (#88605) 2026-05-31 12:58:01 +01:00
WhatsSkiLL
22b8e1cf4f fix(plugins): scope startup metadata manifest reads
Limit plugin metadata snapshots to the channel, provider, and startup surfaces that need them, while preserving unscoped fallback for incomplete index data and provider runtime resolution.

Refs #70533.
Refs #84628.

Co-authored-by: IWhatsskill <IWhatsskill@users.noreply.github.com>
2026-05-31 11:58:56 +01:00
Peter Steinberger
9dc4c9ec2e fix: expose Feishu tools for named accounts 2026-05-31 11:36:48 +01:00
Peter Steinberger
77f1359612 refactor: extract media and ACP core packages (#88534)
* refactor: extract media and acp core packages

* refactor: remove relocated media and acp sources

* build: wire new core packages into dependency checks

* test: alias new core packages in vitest

* build: keep media sniffer runtime dependency

* docs: refresh plugin sdk api baseline

* fix: keep normalized proposal queries non-empty

* test: keep channel timer tests isolated

* fix: keep rebased plugin checks green

* fix: preserve sms numeric allowlist entries

* test: harden exec foreground timeout failure

* test: remove duplicate skill workshop assertion

* fix: remove channel config lint suppression

* test: refresh lint suppression allowlist
2026-05-31 11:30:33 +01:00
stain lu
4ca22b95bc test(plugins): cover Link agent wallet bundle shape (#75181)
* test(plugins): cover Link agent wallet bundle shape

* test(plugins): add bundle fixture helpers

* test(plugins): align Link manifest fixture expectation

---------

Co-authored-by: Peter Steinberger <steipete@gmail.com>
2026-05-31 10:43:12 +01:00
Peter Steinberger
f5eca3f84c chore(lint): enable object and reassignment rules 2026-05-31 09:32:52 +01:00
Andy Ye
0f1767a26a fix(telegram): support media message edits
Fixes #86161.

Route Telegram media-message edits through the Telegram caption/reply-markup APIs instead of always calling `editMessageText`. Button-only edits now update reply markup, explicit captions use `editMessageCaption`, and text edits can fall back to caption edits when Telegram reports the message has no editable text.

Also documents the edit behavior, adds regression coverage, tightens timer-spy cleanup for the affected agents test lane, and removes a stale loader helper from the current base that broke core typecheck.

Co-authored-by: Andy Ye <35905412+TurboTheTurtle@users.noreply.github.com>
2026-05-31 09:22:20 +01:00
Peter Steinberger
63a95930ca chore: remove stale unsafe assertion suppressions 2026-05-31 09:11:28 +01:00
Peter Steinberger
4471335d26 Revert "fix(ci): clean core unsafe assertions"
This reverts commit 88203c9b10.
2026-05-31 04:09:48 -04:00
Peter Steinberger
88203c9b10 fix(ci): clean core unsafe assertions 2026-05-31 03:58:58 -04:00