014667e00b
fix: tighten group elevated targeting
2026-01-08 22:57:18 +01:00
8930ec32cb
feat: add slack multi-account routing
2026-01-08 08:49:16 +01:00
dc3e3f27d4
style: apply lint fixes
2026-01-08 04:44:15 +00:00
5b45a143a1
fix: stabilize media paths and logs
2026-01-08 04:41:01 +00:00
ab8db941d0
feat: expand inbound media notes
2026-01-08 05:20:39 +01:00
ad5c87c193
fix: relax slash command parsing
2026-01-08 03:24:18 +01:00
36b443f4f3
Telegram: fix /think command to show current level when no arg
2026-01-08 03:24:18 +01:00
b2de667b11
fix: persist topic session files
2026-01-07 22:56:50 +00:00
8da4f259dd
Implement Phase 2: Topic-level message history isolation for multi-topic Telegram support
...
Add topic-specific session file isolation to fix root cause of Gemini turn validation errors.
Each Telegram topic now maintains its own conversation history file, eliminating race
conditions and message corruption during concurrent topic processing.
Changes:
1. Enhanced resolveSessionTranscriptPath() to support optional topicId parameter
- Topic ID (Telegram messageThreadId) now incorporated into session filename
- Format: sessionId.jsonl (direct chats) vs sessionId-topic-{topicId}.jsonl (topics)
- Backward compatible: topicId is optional
2. Updated reply.ts to pass MessageThreadId to session file resolution
- ctx.MessageThreadId now flows through to resolveSessionTranscriptPath()
- Automatically provides topic context for each incoming message
3. Automatic propagation through entire system
- sessionFile parameter automatically carries topic-specific path through:
- FollowupRun object (queued runs)
- runEmbeddedPiAgent() calls
- compactEmbeddedPiSession() calls
- SessionManager lifecycle (load, read, write operations)
Benefits:
✓ Complete elimination of shared .jsonl race conditions
✓ Each topic's conversation history independently cached
✓ SessionManager instances operate on isolated files
✓ No concurrent mutations of the same message history
✓ Maintains full Phase 1 turn validation as safety layer
Testing:
✓ Build succeeds with no TypeScript errors
✓ Backward compatible with non-topic sessions (direct messages)
✓ Topic ID properly extracted from Telegram messageThreadId
Expected impact:
- Gemini "function call turn" errors eliminated (root cause fixed)
- Message history corruption prevented across all topics
- Improved stability in multi-topic scenarios
- Each topic maintains independent conversation state
This completes the two-phase fix:
- Phase 1 (previous): Turn validation to suppress errors
- Phase 2 (current): Topic isolation to fix root cause
🤖 Generated with Claude Code
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
2026-01-07 22:51:26 +00:00
434c25331e
refactor: centralize typing mode signals
2026-01-07 22:18:11 +00:00
bac1608933
feat: add typing mode controls
2026-01-07 21:58:54 +00:00
7ce1f635cd
fix(commands): harden model alias parsing
2026-01-07 20:41:41 +00:00
bb29a3ee3f
fix: filter reserved commands from model aliases + add tests
2026-01-07 20:41:41 +00:00
e41540e4ff
feat(commands): add dynamic /<alias> model switching
2026-01-07 20:41:41 +00:00
7e5cef29a0
Threads: add Slack/Discord thread sessions
2026-01-07 19:30:30 +01:00
61f720b945
feat: add skill filter + group system prompt plumbing
2026-01-07 11:44:37 +01:00
a700f9896d
feat: telegram draft streaming
2026-01-07 11:08:32 +01:00
7973fd4caf
feat: add agents command
2026-01-07 10:03:53 +01:00
1673a221f8
feat: add /reasoning reasoning visibility
2026-01-07 06:17:31 +01:00
3668388912
fix(routing): harden originating reply routing
2026-01-07 05:02:34 +00:00
9d50ebad7d
feat(routing): route replies to originating channel
...
Implement reply routing based on OriginatingChannel/OriginatingTo fields.
This ensures replies go back to the provider where the message originated
instead of using the session's lastChannel.
Changes:
- Add OriginatingChannel/OriginatingTo fields to MsgContext (templating.ts)
- Add originatingChannel/originatingTo fields to FollowupRun (queue.ts)
- Create route-reply.ts with provider-agnostic router
- Update all providers (Telegram, Slack, Discord, Signal, iMessage)
to pass originating channel info
- Update reply.ts to pass originating channel to followupRun
- Update followup-runner.ts to use route-reply for originating channels
This addresses the issue where messages from one provider (e.g., Slack)
would receive replies on a different provider (e.g., Telegram) because
the queue used the last active dispatcher instead of the originating one.
2026-01-07 04:51:33 +00:00
9b22e1f6e9
feat(commands): unify chat commands ( #275 )
...
* Chat commands: registry, access groups, Carbon
* Chat commands: clear native commands on disable
* fix(commands): align command surface typing
* docs(changelog): note commands registry (PR #275 )
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com>
2026-01-06 20:17:56 +00:00
dbfa316d19
feat: multi-agent routing + multi-account providers
2026-01-06 18:33:37 +00:00
6cf3570c5b
feat(agent): add skipBootstrap config to skip bootstrap file creation ( #292 )
2026-01-06 11:02:51 -06:00
d5f088978a
fix: stop typing after dispatcher idle
2026-01-06 03:09:49 +00:00
20a361a3cf
refactor: centralize agent timeout defaults
2026-01-06 02:48:44 +00:00
b56338171b
feat: gate slash commands and add compact
2026-01-06 02:23:55 +01:00
b04c838c15
feat!: redesign model config + auth profiles
2026-01-06 00:56:58 +00:00
17ef7b3b0e
fix: status runtime + help
2026-01-05 07:07:17 +01:00
995f5959af
fix: stage sandbox media for inbound attachments
2026-01-05 06:18:11 +01:00
1bad96aa2b
style: tidy auto-reply imports and formatting
2026-01-05 01:46:16 +01:00
852f947b44
fix: unify control command handling
2026-01-05 01:31:36 +01:00
946b32c842
fix(whatsapp): suppress typing during heartbeats
...
- Prevent typing indicator during heartbeat runs
- Add regression tests
Co-authored-by: Jake <mcinteerj@gmail.com>
2026-01-04 23:03:36 +00:00
1657c5e3d2
fix: route system events per session
2026-01-04 22:11:04 +01:00
64e656af82
fix: default elevated level to 'off' when not allowed
...
When elevatedAllowed is false (e.g., for heartbeat surface which isn't
in any allowFrom list), the elevated level was incorrectly defaulting
to 'on', causing bash commands to fail with 'elevated is not available'.
Now defaults to 'off' when elevated isn't allowed, so bash works
normally without trying to use elevated mode.
Fixes: https://github.com/clawdbot/clawdbot/issues/181
2026-01-04 17:36:14 +01:00
246adaa119
chore: rename project to clawdbot
2026-01-04 14:38:51 +00:00
c9504a6f20
refactor: split config module
2026-01-04 07:05:17 +01:00
5e36e2c3f3
fix: allow elevated via discord username
2026-01-04 05:47:28 +00:00
d2da305190
feat: fallback elevated allowlist to discord dms
2026-01-04 05:31:00 +00:00
ff88f3c075
style: fix lint ordering
2026-01-04 06:27:54 +01:00
a03895dfa9
fix: default elevated mode to on
2026-01-04 05:19:28 +00:00
fe0b3500cc
feat: add elevated bash mode
2026-01-04 05:15:59 +00:00
72a9e58777
refactor(auto-reply): split reply flow
2026-01-04 05:47:37 +01:00
a1780efb9f
fix: adjust typing TTL
2026-01-04 00:26:31 +00:00
86038ec165
chore: apply lint fixes
2026-01-04 00:06:02 +01:00
e7c9b9a749
feat: add sessions tools and send policy
2026-01-03 23:44:42 +01:00
934f891932
fix: include embedded agent error cause in reply
2026-01-03 21:30:43 +00:00
ce92fac983
chore: formatting
2026-01-03 12:35:16 -06:00
f313af75e9
fix: avoid duplicate block-stream payloads
2026-01-03 16:53:56 +00:00
200dd634fb
fix: preserve block streaming order
2026-01-03 17:14:01 +01:00
3bbdcaf87f
fix: avoid duplicate block streaming
2026-01-03 17:10:47 +01:00
9f8eeceae7
feat: soften block streaming chunking
2026-01-03 16:48:26 +01:00
733e86516e
fix: address runtime warnings in build
2026-01-03 15:01:38 +00:00
1a00175eb7
chore: fix lint formatting
2026-01-03 14:57:49 +00:00
48731f494b
fix: add embedded run logs and typing ttl
2026-01-03 14:09:19 +00:00
b6301c719b
fix: default low thinking for reasoning models
2026-01-03 12:19:06 +00:00
772ada4308
fix: refine tool summaries and scope discord tool
2026-01-03 12:33:42 +01:00
b914eaa6fa
chore: apply biome lint fixes
2026-01-03 05:10:09 +01:00
971b98c96d
test: cover new queue modes
2026-01-03 04:47:34 +01:00
ac36eba822
feat: expand queue modes and followup backlog
2026-01-03 04:26:49 +01:00
1781105438
group chat: hint that reactions are welcome while lurking
...
Even when staying silent, emoji reactions show engagement without cluttering chat.
2026-01-03 02:29:32 +00:00
632ca01fbf
style: format linted files
2026-01-03 03:10:17 +01:00
a3865f1417
group chat: add lurking guidance to system prompt
...
Be a good group participant: lurk and follow the conversation,
but only chime in when genuinely helpful. Quality over quantity.
2026-01-03 02:02:55 +00:00
0c013a237f
fix: default block streaming break to message_end
2026-01-03 01:33:10 +00:00
3368fcf31e
fix: avoid duplicate replies with block streaming
2026-01-03 02:16:01 +01:00
2d28fa34f5
feat: make block streaming break configurable
2026-01-03 01:49:27 +01:00
59601eb99c
fix: preserve newlines in reply tags
2026-01-02 23:36:43 +00:00
9616f4b2b1
feat: stream reply blocks immediately
2026-01-03 00:28:33 +01:00
2c92ccd66e
feat: add reply tags and replyToMode
2026-01-02 23:18:41 +01:00
a9ff03acaf
feat: unify group mention defaults
2026-01-02 22:50:58 +01:00
5cf1a9535e
feat: move group mention gating to provider groups
2026-01-02 22:24:26 +01:00
8989bd9fd7
fix(auto-reply): default whatsapp self-only on empty config
2026-01-02 17:15:26 +01:00
314164fb8a
chore: fix lint and add gateway auth tests
2026-01-02 17:15:26 +01:00
0766c5e3cb
refactor: move whatsapp allowFrom config
2026-01-02 13:00:29 +01:00
b50df6eb1d
style: format linted files
2026-01-02 12:20:38 +01:00
eb44ae76f1
feat: add discord guild map + group dm controls
2026-01-02 11:15:52 +01:00
9adbf47773
refactor: normalize group session keys
2026-01-02 10:14:58 +01:00
4c2812b429
fix: refine HEARTBEAT_OK handling
2026-01-02 01:42:27 +01:00
c0976ec099
fix(gateway): stream chat events for agent runs
2026-01-02 01:04:59 +01:00
06e379a239
fix: suppress stray HEARTBEAT_OK replies
2026-01-01 23:53:29 +01:00
dc8f2bda2a
fix: restart via systemd on linux
2026-01-01 17:48:28 +01:00
cae78b3f91
fix: treat /model status as model list
2025-12-27 12:10:44 +00:00
c0fb814658
fix: normalize imports for lint
2025-12-27 04:02:13 +01:00
96911d7790
fix: enqueue system event on model switch
2025-12-27 01:17:12 +00:00
7e380bb6f8
fix: enable lmstudio responses and drop think tags
2025-12-27 00:28:52 +00:00
2477ffd860
chore: fix lint/test gating
2025-12-26 23:54:30 +00:00
5c8e1b6eef
feat: add model aliases + minimax shortlist
2025-12-26 23:26:14 +00:00
ce91aba4de
fix: apply biome formatting
2025-12-26 14:38:37 +01:00
a678c3f53e
refactor(queue): remove drop mode
2025-12-26 14:29:28 +01:00
3e4fc7ff7f
feat(queue): add reset/default directive
2025-12-26 14:24:53 +01:00
8dda07a1e9
feat(queue): add queue modes and discord gating
2025-12-26 13:35:44 +01:00
ac659ff5a7
feat(discord): Discord transport
2025-12-26 13:20:30 +01:00
d28265cfbe
fix: handle embedded agent overflow
2025-12-26 10:20:21 +01:00
8f9d7405ed
style: fix biome formatting
2025-12-26 00:50:46 +00:00
9f7b1f0942
feat: move heartbeat config to agent.heartbeat
2025-12-26 01:13:42 +01:00
1ef888ca23
refactor(config): drop agent.provider
2025-12-26 01:13:42 +01:00
8b815bce94
feat(config): allow provider/model shorthand
2025-12-26 01:13:42 +01:00
198f8ea700
fix(agent): serialize runs per session
2025-12-25 23:50:52 +01:00
267cdf20e1
style: fix biome lint
2025-12-24 00:33:35 +00:00
93af424ce5
refactor: move inbound config
2025-12-24 00:22:52 +00:00
Peter Steinberger
Lutro
hsrvc
Azade
Shadow
Josh Lehman
Onur
Jake
Cash Williams