46f0bfc55b
Gateway: harden custom session-store discovery ( #44176 )
...
Merged via squash.
Prepared head SHA: 52ebbf5188
2026-03-12 16:44:46 +00:00
b77b7485e0
feat(push): add iOS APNs relay gateway ( #43369 )
...
* feat(push): add ios apns relay gateway
* fix(shared): avoid oslog string concatenation
# Conflicts:
# apps/shared/OpenClawKit/Sources/OpenClawKit/GatewayChannel.swift
* fix(push): harden relay validation and invalidation
* fix(push): persist app attest state before relay registration
* fix(push): harden relay invalidation and url handling
* feat(push): use scoped relay send grants
* feat(push): configure ios relay through gateway config
* feat(push): bind relay registration to gateway identity
* fix(push): tighten ios relay trust flow
* fix(push): bound APNs registration fields (#43369 ) (thanks @ngutman)
2026-03-12 18:15:35 +02:00
9342739d71
fix(providers): respect user-configured baseUrl for kimi-coding ( #36647 )
...
* fix(providers): respect user-configured baseUrl for kimi-coding
The kimi-coding provider was built exclusively from
`buildKimiCodingProvider()` defaults, ignoring any user-specified
`baseUrl` or other overrides in `openclaw.json` providers config.
This caused 404 errors when users configured a custom endpoint.
Now merge `explicitProviders["kimi-coding"]` on top of defaults,
matching the pattern used by ollama/vllm. User's `baseUrl`, `api`,
and `models` take precedence; env/profile API key still wins.
Fixes #36353
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* Tests: use Kimi implicit provider harness
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Vincent Koc <vincentkoc@ieee.org>
2026-03-12 12:14:07 -04:00
3e28e10c2f
Plugins: require explicit trust for workspace-discovered plugins ( #44174 )
...
* Plugins: disable implicit workspace plugin auto-load
* Tests: cover workspace plugin trust gating
* Changelog: note workspace plugin trust hardening
* Plugins: keep workspace trust gate ahead of memory slot defaults
* Tests: cover workspace memory-slot trust bypass
2026-03-12 12:12:41 -04:00
0a8fa0e001
Moonshot: respect explicit baseUrl for CN endpoint so platform.moonshot.cn keys authenticate ( #33637 ) ( #33696 )
...
* Moonshot: respect explicit baseUrl for CN endpoint so platform.moonshot.cn keys authenticate (#33637 )
* Moonshot: address review - remove dead constant, import canonical URLs (#33696 )
2026-03-12 12:10:38 -04:00
3fa91cd69d
feat: add sessions_yield tool for cooperative turn-ending ( #36537 )
...
Merged via squash.
Prepared head SHA: 75d9204c86
2026-03-12 08:46:47 -07:00
e6897c800b
Plugins: fix env-aware root resolution and caching ( #44046 )
...
Merged via squash.
Prepared head SHA: 6e8852a188
2026-03-12 15:31:31 +00:00
688e3f0863
Compaction Runner: emit transcript updates post-compact ( #25558 )
...
Merged via squash.
Prepared head SHA: 8a858436ed
2026-03-12 08:22:12 -07:00
8ad0ca309e
Subagents: stop retrying external completion timeouts ( #41235 ) ( #43847 )
...
* Changelog: add subagent announce timeout note
* Tests: cover subagent completion timeout no-retry
* Subagents: stop retrying external completion timeouts
* Config: update subagent announce timeout default docs
* Tests: use fake timers for subagent timeout retry guard
2026-03-12 11:03:06 -04:00
7844bc89a1
Security: require Feishu webhook encrypt key ( #44087 )
...
* Feishu: require webhook encrypt key in schema
* Feishu: cover encrypt key webhook validation
* Feishu: enforce encrypt key at startup
* Feishu: add webhook forgery regression test
* Feishu: collect encrypt key during onboarding
* Docs: require Feishu webhook encrypt key
* Changelog: note Feishu webhook hardening
* Docs: clarify Feishu encrypt key screenshot
* Feishu: treat webhook encrypt key as secret input
* Feishu: resolve encrypt key only in webhook mode
2026-03-12 11:01:00 -04:00
99170e2408
Hardening: normalize Unicode command obfuscation detection ( #44091 )
...
* Exec: cover unicode obfuscation cases
* Exec: normalize unicode obfuscation detection
* Changelog: note exec detection hardening
* Exec: strip unicode tag character obfuscation
* Exec: harden unicode suppression and length guards
* Exec: require path boundaries for safe URL suppressions
2026-03-12 10:57:49 -04:00
eff0d5a947
Hardening: tighten preauth WebSocket handshake limits ( #44089 )
...
* Gateway: tighten preauth handshake limits
* Changelog: note WebSocket preauth hardening
* Gateway: count preauth frame bytes accurately
* Gateway: cap WebSocket payloads before auth
2026-03-12 10:55:41 -04:00
48cbfdfac0
Hardening: require LINE webhook signatures ( #44090 )
...
* LINE: require webhook signatures in express handler
* LINE: require webhook signatures in node handler
* LINE: update express signature tests
* LINE: update node signature tests
* Changelog: note LINE webhook hardening
* LINE: validate signatures before parsing webhook bodies
* LINE: reject missing signatures before body reads
2026-03-12 10:50:36 -04:00
c965049dc6
fix(mattermost): pass mediaLocalRoots through reply delivery ( #44021 )
...
Merged via squash.
Prepared head SHA: 856f11f129
2026-03-12 20:13:51 +05:30
b0f717aa02
build: align Node 22 guidance with 22.16 minimum
2026-03-12 20:07:44 +05:30
0a8d2b6200
build: raise Node 22 compatibility floor to 22.16
2026-03-12 20:07:44 +05:30
deada7edd3
build: default to Node 24 and keep Node 22 compat
2026-03-12 20:07:44 +05:30
2f037f0930
Agents: adapt pi-ai oauth and payload hooks
2026-03-12 10:19:14 -04:00
f3be1c828c
fix(status): resolve context window by provider-qualified key, prefer max on bare-id collision, solve #35976 ( #36389 )
...
Merged via squash.
Prepared head SHA: f8cf752c59
2026-03-12 07:00:36 -07:00
ff47876e61
fix: carry observed overflow token counts into compaction ( #40357 )
...
Merged via squash.
Prepared head SHA: b99eed4329
2026-03-12 06:58:42 -07:00
f2e28fc30f
fix(telegram): allow fallback models in /model validation ( #40105 )
...
Merged via squash.
Prepared head SHA: de07585e03
2026-03-12 13:55:51 +01:00
4f620bebe5
fix(doctor): canonicalize gateway service entrypoint paths ( #43882 )
...
Merged via squash.
Prepared head SHA: 9f530d2a86
2026-03-12 12:39:22 +02:00
5acf6cae8e
fix: stop main-session UI replies inheriting channel routes
2026-03-12 15:39:34 +05:30
8ea79b64d0
fix: preserve sandbox write payload stdin ( #43876 )
...
Merged via squash.
Prepared head SHA: a10fd4b21c
2026-03-12 12:42:57 +03:00
f640326e31
fix(failover): add missing network errno patterns to text-based timeout classifier ( #42830 )
...
Merged via squash.
Prepared head SHA: 91761487e8
2026-03-12 12:34:44 +03:00
7c889e7113
Refactor: trim duplicate gateway/onboarding helpers and dead utils ( #43871 )
...
* Gateway: share input provenance schema
* Onboarding: dedupe top-level channel patching
* Utils: remove unused path helpers
* Protocol: refresh generated gateway models
2026-03-12 05:04:31 -04:00
cb7b38105f
Merge remote-tracking branch 'origin/vincentkoc-code/fix-terminal-table-width'
...
* origin/vincentkoc-code/fix-terminal-table-width:
Terminal: consume unsupported escape bytes in tables
Skills: normalize emoji presentation across outputs
Changelog: note terminal skills table fixes
Skills: use Terminal-safe emoji in list output
Terminal: stop shrinking CLI tables by one column
Terminal: refine table wrapping and width handling
Update CHANGELOG.md
Deps: patch file-type and hono
Tests: cover emoji table alignment
Terminal: wrap table cells by grapheme width
Terminal: measure grapheme display width
Tests: cover grapheme terminal width
Changelog: add unreleased March 9 entries
# Conflicts:
# CHANGELOG.md
# package.json
# pnpm-lock.yaml
# src/cli/skills-cli.format.ts
# src/terminal/table.test.ts
2026-03-12 04:56:21 -04:00
658bd54ecf
feat(llm-task): add thinking override
...
Co-authored-by: Xaden Ryan <165437834+xadenryan@users.noreply.github.com>
2026-03-12 19:21:35 +11:00
f37815b323
Gateway: block profile mutations via browser.request ( #43800 )
...
* Gateway: block profile mutations via browser.request
* Changelog: note GHSA-vmhq browser request fix
* Gateway: normalize browser.request profile guard paths
2026-03-12 04:21:03 -04:00
46a332385d
Gateway: keep spawned workspace overrides internal ( #43801 )
...
* Gateway: keep spawned workspace overrides internal
* Changelog: note GHSA-2rqg agent boundary fix
* Gateway: persist spawned workspace inheritance in sessions
* Agents: clean failed lineage spawn state
* Tests: cover lineage attachment cleanup
* Tests: cover lineage thread cleanup
2026-03-12 04:20:00 -04:00
97683071b5
Tests: extend exec allowlist glob coverage
2026-03-12 04:01:49 -04:00
9aeaa19e9e
Agents: clear invalidated Kimi tool arg repair ( #43824 )
2026-03-12 03:53:06 -04:00
c5ea6134d0
feat(ui): add chat infrastructure modules (slice 1/3 of dashboard-v2) ( #41497 )
...
* feat(ui): add chat infrastructure modules (slice 1 of dashboard-v2)
New self-contained chat modules extracted from dashboard-v2-structure:
- chat/slash-commands.ts: slash command definitions and completions
- chat/slash-command-executor.ts: execute slash commands via gateway RPC
- chat/slash-command-executor.node.test.ts: test coverage
- chat/speech.ts: speech-to-text (STT) support
- chat/input-history.ts: per-session input history navigation
- chat/pinned-messages.ts: pinned message management
- chat/deleted-messages.ts: deleted message tracking
- chat/export.ts: shared exportChatMarkdown helper
- chat-export.ts: re-export shim for backwards compat
Gateway fix:
- Restore usage/cost stripping in chat.history sanitization
- Add test coverage for sanitization behavior
These modules are additive and tree-shaken — no existing code
imports them yet. They will be wired in subsequent slices.
* Update ui/src/ui/chat/export.ts
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
* fix(ui): address review feedback on chat infra slice
- export.ts: handle array content blocks (Claude API format) instead
of silently exporting empty strings
- slash-command-executor.ts: restrict /kill all to current session's
subagent subtree instead of all sessions globally
- slash-command-executor.ts: only count truly aborted runs (check
aborted !== false) in /kill summary
* fix: scope /kill <id> to current session subtree and preserve usage.cost in chat.history
- Restrict /kill <id> matching to only subagents belonging to the current
session's agent subtree (P1 review feedback)
- Preserve nested usage.cost in chat.history sanitization so cost badges
remain available (P2 review feedback)
* fix(ui): tighten slash kill scoping
* fix(ui): support legacy slash kill scopes
* fix(ci): repair pr branch checks
* Gateway: harden chat abort and export
* UI: align slash commands with session tree scope
* UI: resolve session aliases for slash command lookups
* Update .gitignore
* Cron: use shared nested lane resolver
---------
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
Co-authored-by: Vincent Koc <vincentkoc@ieee.org>
2026-03-12 03:48:58 -04:00
ed0ec57a7b
fix: scope telegram polling restart to telegram errors ( #43799 )
...
* fix: scope telegram polling restart to telegram errors
* fix: make telegram error tagging best-effort
* fix: scope telegram polling restart to telegram errors (#43799 )
2026-03-12 13:14:17 +05:30
82e3ac21ee
Infra: tighten exec allowlist glob matching ( #43798 )
...
* Infra: tighten exec allowlist glob matching
* Changelog: note GHSA-f8r2 exec allowlist fix
2026-03-12 03:33:50 -04:00
d8ee97c466
Agents: recover malformed Anthropic-compatible tool call args ( #42835 )
...
* Agents: recover malformed anthropic tool call args
* Agents: add malformed tool call regression test
* Changelog: note Kimi tool call arg recovery
* Agents: repair toolcall end message snapshots
* Agents: narrow Kimi tool call arg repair
2026-03-12 03:28:22 -04:00
0bcb95e8fa
Models: enforce source-managed SecretRef markers in models.json ( #43759 )
...
Merged via squash.
Prepared head SHA: 4a065ef5d8
2026-03-12 02:22:52 -05:00
e8a162d3d8
fix(mattermost): prevent duplicate messages when block streaming + threading are active ( #41362 )
...
* fix(mattermost): prevent duplicate messages when block streaming + threading are active
Remove replyToId from createBlockReplyPayloadKey so identical content is
deduplicated regardless of threading target. Add explicit threading dock
to the Mattermost plugin with resolveReplyToMode reading from config
(default "all"), and add replyToMode to the Mattermost config schema.
Fixes #41219
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(mattermost): address PR review — per-account replyToMode and test clarity
Read replyToMode from the merged per-account config via
resolveMattermostAccount so account-level overrides are honored in
multi-account setups. Add replyToMode to MattermostAccountConfig type.
Rename misleading test to clarify it exercises shouldDropFinalPayloads
short-circuit, not payload key dedup.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* Replies: keep block-pipeline reply targets distinct
* Tests: cover block reply target-aware dedupe
* Update CHANGELOG.md
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Vincent Koc <vincentkoc@ieee.org>
2026-03-12 03:15:17 -04:00
6c196c913f
fix(cron): prevent duplicate proactive delivery on transient retry ( #40646 )
...
* fix(cron): prevent duplicate proactive delivery on transient retry
* refactor: scope skipQueue to retryTransient path only
Non-retrying direct delivery (structured content / thread) keeps the
write-ahead queue so recoverPendingDeliveries can replay after a crash.
Addresses review feedback from codex-connector.
* fix: preserve write-ahead queue on initial delivery attempt
The first call through retryTransientDirectCronDelivery now keeps the
write-ahead queue entry so recoverPendingDeliveries can replay after a
crash. Only subsequent retry attempts set skipQueue to prevent
duplicate sends.
Addresses second codex-connector review on ea5ae5c.
* ci: retrigger checks
* Cron: bypass write-ahead queue for direct isolated delivery
* Tests: assert isolated cron skipQueue invariants
* Changelog: add cron duplicate-delivery fix entry
---------
Co-authored-by: Vincent Koc <vincentkoc@ieee.org>
2026-03-12 03:01:19 -04:00
f3c00fce15
fix: prevent duplicate assistant messages in TUI ( fixes #35278 ) ( #35364 )
...
* fix: prevent duplicate assistant messages in TUI (fixes #35278 )
When startAssistant() is called multiple times with the same runId,
it was creating duplicate AssistantMessageComponent instances instead
of reusing the existing one. This caused messages to appear twice in
the terminal UI.
The fix checks if a component already exists for the runId before
creating a new one. If it exists, we update its text instead of
appending a duplicate component.
Test coverage includes verification that:
- Only one component is created when startAssistant is called twice
- The second text replaces the first
- Component count remains 1 (prevents regression)
Generated with [Claude Code](https://claude.ai/code )
via [Happy](https://happy.engineering )
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
* Changelog: add TUI duplicate-render fix entry
---------
Co-authored-by: 沐沐 <mumu@example.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Happy <yesreply@happy.engineering>
Co-authored-by: Vincent Koc <vincentkoc@ieee.org>
2026-03-12 02:59:42 -04:00
99ec687d7a
fix(agents): enforce sandboxed session_status visibility ( #43754 )
...
* agents: guard sandboxed session_status access
* test(agents): cover sandboxed session_status scope
* docs(changelog): credit session_status hardening
* agents: preflight sandboxed session_status checks
* test(agents): cover session_status existence oracle
* agents: preserve legacy session_status tree keys
* test(agents): cover legacy session_status tree keys
* Update CHANGELOG.md
2026-03-12 02:54:25 -04:00
12dc299cde
fix(imessage): dedupe reflected self-chat duplicates ( #38440 )
...
* iMessage: drop reflected self-chat duplicates
* Changelog: add iMessage self-chat echo dedupe entry
* iMessage: keep self-chat dedupe scoped to final group identity
* iMessage: harden self-chat cache
* iMessage: sanitize self-chat duplicate logs
* iMessage: scope group self-chat dedupe by sender
* iMessage: move self-chat cache identity into cache
* iMessage: hash full self-chat text
* Update CHANGELOG.md
2026-03-12 02:27:35 -04:00
d8d8dc7421
Infra: fail closed without device scope baseline
2026-03-12 01:42:12 -04:00
276ee259ca
Tests: clean up temp git helper directory
2026-03-12 01:42:12 -04:00
4f462facda
Infra: cap device tokens to approved scopes ( #43686 )
...
* Infra: cap device tokens to approved scopes
* Changelog: note device token hardening
2026-03-12 01:25:52 -04:00
2504cb6a1e
Security: escape invisible exec approval format chars ( #43687 )
...
* Infra: escape invisible exec approval chars
* Gateway: sanitize exec approval display text
* Tests: cover sanitized exec approval payloads
* Tests: cover sanitized exec approval forwarding
* Changelog: note exec approval prompt hardening
2026-03-12 01:20:04 -04:00
1dcef7b644
Infra: block GIT_EXEC_PATH in host env sanitizer ( #43685 )
...
* Infra: block GIT_EXEC_PATH in host env sanitizer
* Changelog: note host env hardening
2026-03-12 01:16:03 -04:00
18f15850e6
fix(browser): restore proxy attachment media size cap ( #43684 )
...
* browser: honor shared proxy file size cap
* test(browser): cover proxy file size cap
* docs(changelog): note browser proxy size cap fix
2026-03-12 01:04:31 -04:00
fbc1bd6f8e
fix: clear telegram polling cleanup timers
2026-03-12 09:36:04 +05:30
70abee69e9
fix(telegram): avoid polling restart hang after stall detection
2026-03-12 09:36:04 +05:30
Gustavo Madeira Santana
Nimrod Gutman
2233admin
Vincent Koc
chengzhichao-xydt
Jacob Riff
Rodrigo Uroz
Lyle
Altay
0x4C33
rabsef-bicrym
avirweb
Ayaan Zaidi
glitch
jnMetaCode
Xaden Ryan
Val Alexander
Josh Avant
Mathias Nagler
wangchunyue(王春跃)
lisitan
Ayaan Zaidi
Huang X