Commit Graph

1471 Commits

Author SHA1 Message Date
Peter Steinberger
2986447935 fix: improve gmail tailscale errors 2026-01-07 03:10:35 +01:00
Peter Steinberger
467d4e17fe feat: add sandbox scope default 2026-01-07 02:52:41 +01:00
Peter Steinberger
15b7560a9b refactor: reuse gateway output helpers 2026-01-07 01:43:02 +00:00
Peter Steinberger
b88c4e9d20 chore: clean up lint and scratchpad 2026-01-07 01:28:46 +00:00
Peter Steinberger
bc9a3ce32a refactor: unify outbound delivery formatting 2026-01-07 01:26:09 +00:00
Peter Steinberger
3fbe2963b3 test: align outbound normalization 2026-01-07 01:22:55 +00:00
Peter Steinberger
8ba6473462 style: fix send json indent 2026-01-07 01:21:29 +00:00
Peter Steinberger
dd78c26e6d style: format direct send json 2026-01-07 01:21:03 +00:00
Peter Steinberger
2ce5df3efc style: align outbound delivery formatting 2026-01-07 01:20:40 +00:00
Peter Steinberger
aefaed159b refactor: normalize outbound payload delivery 2026-01-07 01:19:47 +00:00
Peter Steinberger
f171d509bb refactor: centralize outbound target validation 2026-01-07 01:16:39 +00:00
Sash Zats
f1643a5b8d Heartbeat: resolve main session key for session store 2026-01-06 20:14:30 -05:00
Peter Steinberger
f5938f8114 refactor: unify outbound delivery 2026-01-07 01:13:04 +00:00
Peter Steinberger
1ae5e9a26b feat: add docs search command 2026-01-07 02:03:06 +01:00
Sash Zats
eb8d7a19af Cron: enqueue system events in main session 2026-01-06 19:55:03 -05:00
Erik
cd4e2023ab fix(agent): capture compaction retry AbortError for model fallback
Wrap waitForCompactionRetry() in try/catch to capture AbortError
that was escaping and bypassing the model fallback mechanism.

When a timeout fires, session.abort() causes both session.prompt()
and waitForCompactionRetry() to throw AbortError. Previously only
the prompt error was captured, allowing the compaction error to
escape to model-fallback.ts where it was immediately re-thrown
(line 199: isAbortError check), bypassing fallback model attempts.

Fixes #313
2026-01-07 01:44:37 +01:00
Peter Steinberger
0116184b1c docs: recommend WSL2 for Windows installs 2026-01-07 01:21:36 +01:00
Peter Steinberger
62112d9978 feat: add onboarding doc links 2026-01-07 01:19:31 +01:00
Peter Steinberger
19c95d0ff7 fix(auth): serialize profile stats updates 2026-01-07 01:06:51 +01:00
Peter Steinberger
96d72ff91e fix(auth): lock auth profile updates 2026-01-07 01:00:47 +01:00
Muhammed Mukhthar CM
eb5f758f6b fix(auth): improve multi-account round-robin rotation and 429 handling
This commit fixes several issues with multi-account OAuth rotation that
were causing slow responses and inefficient account cycling.

## Changes

### 1. Fix usageStats race condition (auth-profiles.ts)

The `markAuthProfileUsed`, `markAuthProfileCooldown`, `markAuthProfileGood`,
and `clearAuthProfileCooldown` functions were using a stale in-memory store
passed as a parameter. Long-running sessions would overwrite usageStats
updates from concurrent sessions when saving.

**Fix:** Re-read the store from disk before each update to get fresh
usageStats from other sessions, then merge the update.

### 2. Capture AbortError from waitForCompactionRetry (pi-embedded-runner.ts)

When a request timed out, `session.abort()` was called which throws an
`AbortError`. The code structure was:

```javascript
try {
  await session.prompt(params.prompt);
} catch (err) {
  promptError = err;  // Catches AbortError here
}
await waitForCompactionRetry();  // But THIS also throws AbortError!
```

The second `AbortError` from `waitForCompactionRetry()` escaped and
bypassed the rotation/fallback logic entirely.

**Fix:** Wrap `waitForCompactionRetry()` in its own try/catch to capture
the error as `promptError`, enabling proper timeout handling.

Root cause analysis and fix proposed by @erikpr1994 in #313.

Fixes #313

### 3. Fail fast on 429 rate limits (pi-ai patch)

The pi-ai library was retrying 429 errors up to 3 times with exponential
backoff before throwing. This meant a rate-limited account would waste
30+ seconds retrying before our rotation code could try the next account.

**Fix:** Patch google-gemini-cli.js to:
- Throw immediately on first 429 (no retries)
- Not catch and retry 429 errors in the network error handler

This allows the caller to rotate to the next account instantly on rate limit.

Note: We submitted this fix upstream (https://github.com/badlogic/pi-mono/pull/504)
but it was closed without merging. Keeping as a local patch for now.

## Testing

With 6 Antigravity accounts configured:
- Accounts rotate properly based on lastUsed (round-robin)
- 429s trigger immediate rotation to next account
- usageStats persist correctly across concurrent sessions
- Cooldown tracking works as expected

## Before/After

**Before:** Multiple 429 retries on same account, 30-90s delays
**After:** Instant rotation on 429, responses in seconds
2026-01-07 00:56:32 +01:00
VAC
ff200e3993 fix(discord): handle voice messages with empty content
Discord voice messages have empty `content` with the audio in attachments.
The nullish coalescing operator (`??`) doesn't fall through on empty strings,
so voice messages were being dropped as 'empty content'.

Changed to logical OR (`||`) so empty string falls through to media placeholder.
2026-01-06 23:35:30 +00:00
Peter Steinberger
7214cf39ec fix: prefer home linuxbrew paths 2026-01-07 00:18:07 +01:00
Peter Steinberger
b57d36f49c fix(sessions_spawn): hard-fail invalid model overrides 2026-01-06 23:17:35 +00:00
Azade
0429a4b63b test(sessions_spawn): add test for model parameter 2026-01-06 23:17:35 +00:00
Azade
274f408e6f feat(sessions_spawn): add model parameter for sub-agent model override 2026-01-06 23:17:35 +00:00
Peter Steinberger
02c9cf0ff4 chore: remove duplicate daemon runtime imports 2026-01-07 00:14:08 +01:00
Peter Steinberger
dd0d23cd96 test(commands): add /stop native regression 2026-01-06 23:11:57 +00:00
Peter Steinberger
e0efcda77f fix(commands): wire /stop across chat commands 2026-01-06 23:11:57 +00:00
Nacho Iacovino
0df7c3addf feat(telegram): add /stop command to abort running agent
Adds a /stop command that:
- Can interrupt a running agent session mid-execution
- Works in both DMs and group chats (including forum topics)
- Uses grammy's bot.command() to run before the main message handler
- Returns status: stopped, stop requested, or nothing running

Also fixes session key lookup in pi-embedded-runner to use sessionKey
instead of sessionId, ensuring /stop finds the correct active run.
2026-01-06 23:11:57 +00:00
Peter Steinberger
5bc3f13b46 feat: colorize models auth key labels 2026-01-07 00:10:01 +01:00
Peter Steinberger
b357746e1c feat: add richer color to models output 2026-01-07 00:07:50 +01:00
Peter Steinberger
79f813e18e style: format lint offenders 2026-01-07 00:04:44 +01:00
Peter Steinberger
0ad74ee941 test: mock select prompt in doctor tests 2026-01-07 00:01:50 +01:00
Peter Steinberger
55278c1c71 feat: add daemon runtime prompts 2026-01-06 23:51:00 +01:00
Peter Steinberger
c920ee1166 Merge branch 'pr-335-merge' 2026-01-06 23:45:35 +01:00
Peter Steinberger
5939363eed fix: include telegram group sender in envelope headers 2026-01-06 22:34:02 +00:00
Peter Steinberger
8d50d08936 style: format daemon runtime changes 2026-01-06 23:29:38 +01:00
Peter Steinberger
8911a79d7f docs: rewrite cron jobs guide and heartbeat notes 2026-01-06 22:28:42 +00:00
Peter Steinberger
707f7918bc feat: add gateway daemon runtime selector 2026-01-06 23:27:58 +01:00
Peter Steinberger
18c43fe462 fix: bootstrap linuxbrew for skills 2026-01-06 23:27:38 +01:00
Peter Steinberger
39d2ba78b7 fix(cli): harden pairing provider parse 2026-01-06 22:17:18 +00:00
Peter Steinberger
0931a65ab2
fix: auto-recover from Gemini session corruption
Auto-merge after checks.
2026-01-06 22:12:05 +00:00
Peter Steinberger
fec7f37271 merge upstream/main 2026-01-06 23:09:01 +01:00
Peter Steinberger
86b56b2308 fix: harden gemini session reset 2026-01-06 23:06:01 +01:00
Peter Steinberger
2771001720 fix(state): auto-migrate legacy agent dir 2026-01-06 22:04:23 +00:00
Peter Steinberger
7aa7fa79d0 feat: update heartbeat defaults 2026-01-06 21:54:42 +00:00
Peter Steinberger
dba09058f5 fix(agents): default agent dir to multi-agent path 2026-01-06 21:54:42 +00:00
Peter Steinberger
96164b5955 fix: improve socket error handling 2026-01-06 22:43:29 +01:00
Emanuel Stadler
fb17a32283 feat: enhance error handling for socket connection errors
- Added `isError` property to `EmbeddedPiRunResult` and reply items to indicate error states.
- Updated error handling in `runReplyAgent` to provide more informative messages for specific socket connection errors.
2026-01-06 22:19:37 +01:00
James Groat
9b6e2478f5 fix(browser): add profile param to tabs routes and browser-tool
- tabs.ts now uses getProfileContext like other routes
- browser-tool threads profile param through all actions
- add tests for profile query param on /tabs endpoints
- update docs with browser tool profile parameter
2026-01-06 21:54:46 +01:00
James Groat
40758b16a9 fix(browser-cli): rename --profile to --browser-profile to avoid conflict with global --profile flag 2026-01-06 21:54:46 +01:00
minghinmatthewlam
2dd6b3aeb2
fix: write auth profiles to multi-agent path during onboarding
- Onboarding now writes auth profiles under ~/.clawdbot/agents/main/agent so the gateway sees credentials on first start.
- Hardened onboarding test to ignore legacy env vars.

Thanks @minghinmatthewlam!
2026-01-06 20:53:18 +00:00
Peter Steinberger
84c8209158 fix(slack): clear assistant thread status after replies 2026-01-06 21:41:30 +01:00
Shadow
8ebc789d25 Slack: send assistant thread status while typing 2026-01-06 21:34:52 +01:00
Shadow
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
Peter Steinberger
1bf44bf30c feat(models): show auth overview 2026-01-06 20:07:04 +00:00
Peter Steinberger
118c1e1042 fix: keep oauth profile stable 2026-01-06 19:43:28 +00:00
Peter Steinberger
67bda21811 fix: preserve markdown fences when chunking 2026-01-06 20:23:41 +01:00
Peter Steinberger
31dbc62bdd fix(telegram): prevent stuck typing after tool runs 2026-01-06 18:56:43 +00:00
Peter Steinberger
369af5fc58 style(agents): format usage helper 2026-01-06 19:54:50 +01:00
Peter Steinberger
d07e78855c fix(workspace): align clawd + bootstrap 2026-01-06 19:54:50 +01:00
Abhi
bdf597eb95
fix(telegram): stop typing after tool results (#322)
Thanks @AbhisekBasu1.
2026-01-06 18:54:08 +00:00
Peter Steinberger
2f24ea492b fix: restore Anthropic token accounting 2026-01-06 18:52:01 +00:00
Peter Steinberger
9fb37cbf93 style: format whatsapp inbound allowlist 2026-01-06 18:33:37 +00:00
Peter Steinberger
dbfa316d19 feat: multi-agent routing + multi-account providers 2026-01-06 18:33:37 +00:00
Peter Steinberger
c47aff5244 fix(onboard): clarify DM policy keys 2026-01-06 18:09:21 +01:00
Peter Steinberger
4933905366 fix(onboard): configure DM policies 2026-01-06 18:09:21 +01:00
Onur
6cf3570c5b
feat(agent): add skipBootstrap config to skip bootstrap file creation (#292) 2026-01-06 11:02:51 -06:00
Peter Steinberger
b081f45b17 fix(onboard): explain DM pairing defaults 2026-01-06 17:58:06 +01:00
Muhammed Mukhthar CM
4bb53e19f9
fix(build): import tool-display.json instead of fs.readFileSync (#312) 2026-01-06 10:55:02 -06:00
Peter Steinberger
967cef80bc fix(security): lock down inbound DMs by default 2026-01-06 17:51:56 +01:00
Peter Steinberger
11b6fddcbf
Merge pull request #283 from Oncomatic/jarvis/telegram-media-error-notify
fix(telegram): notify user when media exceeds size limit
2026-01-06 15:18:52 +00:00
Peter Steinberger
3ff17b70ea chore: changelog for #293 2026-01-06 15:32:06 +01:00
Palash Oswal
b91012b697
fix(cli): don't force localhost gateway url in remote mode
Fixes remote gateway setup (remote mode) by not overriding url; adds regression tests. Thanks @oswalpalash.
2026-01-06 14:30:45 +00:00
Simon Kelly
5aa1ed2c96
fix(slack): use named import for @slack/bolt App class (#299)
* fix(slack): use named import for @slack/bolt App class

The default import `import bolt from '@slack/bolt'` followed by
`const { App } = bolt` doesn't work correctly in Bun due to ESM/CJS
interop issues. The default export comes through as a function rather
than the module object.

Switching to a named import `import { App } from '@slack/bolt'`
resolves the issue and allows the Slack provider to start successfully.

* fix(slack): align Bolt mock with named App export

---------

Co-authored-by: Peter Steinberger <steipete@gmail.com>
2026-01-06 14:22:14 +00:00
VAC
eadb923000 fix: auto-recover from Gemini session corruption
Detect the Gemini API error 'function call turn comes immediately after
a user turn or after a function response turn' which indicates corrupted
session history.

When detected:
- Delete the corrupted transcript file
- Remove the session entry from the store
- Return a user-friendly message asking them to retry

This prevents the error loop where every subsequent message fails with
the same error until manual intervention.

Fixes #296
2026-01-06 07:25:15 -05:00
Manuel Hettich
aae5926db9 fix(telegram): notify user when media exceeds size limit
When a file exceeds mediaMaxMb, send a friendly error message
instead of silently dropping the upload.

---
🤖 Authored by Jarvis (AI assistant)
2026-01-06 08:45:07 +00:00
Peter Steinberger
3693449d7e feat: sandbox session tool visibility 2026-01-06 08:40:30 +00:00
Peter Steinberger
5926a98c52 fix(configure): don’t write auth.order by default 2026-01-06 09:25:36 +01:00
Peter Steinberger
f2d353459f test(auth): stop prioritizing lastGood 2026-01-06 09:25:33 +01:00
Peter Steinberger
ed2075ce69 test(gateway): deflake cron finished event wait 2026-01-06 09:25:31 +01:00
Muhammed Mukhthar CM
9e49c762e0
fix(auth): prioritize round-robin over lastGood for multi-account rotation (#281)
* fix(auth): prioritize round-robin over lastGood for multi-account rotation

When multiple OAuth accounts are configured, the round-robin rotation was
not working because lastGood was always prioritized, defeating the sort by
lastUsed.

Changes:
- Remove lastGood prioritization in resolveAuthProfileOrder
- Always apply orderProfilesByMode (sorts by lastUsed, oldest first)
- Only respect configuredOrder when explicitly set in config
- preferredProfile still takes priority for explicit user choice

Tested with 2 Google Antigravity accounts - verified alternating usage.

Follow-up to PR #269.

* style: fix formatting
2026-01-06 08:16:35 +00:00
Peter Steinberger
cf1a1d107e fix: add OpenAI Codex OAuth to configure 2026-01-06 09:13:51 +01:00
Muhammed Mukhthar CM
42d1c2448e
fix(cron-tool): use generic object schema for job/patch to fix Claude via Antigravity (#280) 2026-01-06 02:13:09 -06:00
Peter Steinberger
c27dd75135 build(control-ui): prefer bun for UI build 2026-01-06 09:08:25 +01:00
Peter Steinberger
a279bcfeb1 feat: add sessions_spawn sub-agent tool 2026-01-06 08:41:45 +01:00
Peter Steinberger
952657d55c feat(tui): add /elev alias 2026-01-06 08:41:04 +01:00
Ayaan Zaidi
7a48b908e4
refactor: replace tsx with bun for TypeScript execution (#278) 2026-01-06 07:14:08 +00:00
Peter Steinberger
dbb51006cd feat: unify group policy allowlists 2026-01-06 06:40:42 +00:00
Peter Steinberger
51e8bbd2a8 style: normalize type definitions 2026-01-06 07:21:10 +01:00
Peter Steinberger
aa16b679ad fix: improve auth profile failover 2026-01-06 07:18:06 +01:00
Peter Steinberger
a7b5753dc4
Merge pull request #269 from mukhtharcm/feat/multi-account-roundrobin
feat: Multi-account OAuth with round-robin rotation
2026-01-06 06:13:19 +00:00
Peter Steinberger
b5c604b7b7 fix: require slash for control commands 2026-01-06 07:05:17 +01:00
Peter Steinberger
7d896b5f67 fix: doctor memory hint 2026-01-06 06:01:24 +00:00
Shadow
91cb2c02a7
fix: allow optional reply body 2026-01-05 23:47:33 -06:00
Shadow
69f285c5ca
chore: fixed CI 2026-01-05 23:36:48 -06:00
Peter Steinberger
b759cb6f37 feat(providers): normalize location parsing 2026-01-06 06:31:09 +01:00
Nacho Iacovino
255e77f530 feat(telegram): parse location and venue messages
- Add TelegramLocation, TelegramVenue, and TelegramMessageWithLocation types
- Add formatLocationMessage() to convert location/venue shares to text
- Add extractLocationData() for structured location access in ctxPayload
- Handle both raw location pins and venue shares (places with names)
- Include location in reply-to context for quoted messages

Location messages now appear as:
- [Location: lat, lon ±accuracy] for raw pins
- [Venue: Name - Address (lat, lon)] for places

ctxPayload includes LocationLat, LocationLon, LocationAccuracy,
VenueName, and VenueAddress fields for programmatic access.
2026-01-06 06:31:09 +01:00
Muhammed Mukhthar CM
18c7795ee0 feat: treat timeout as rate limit for profile rotation
Antigravity rate limits cause requests to hang indefinitely rather than
returning 429 errors. This change detects timeouts and treats them as
potential rate limits:

- Added timedOut flag to track timeout-triggered aborts
- Timeout now triggers profile cooldown + rotation
- Logs: "Profile X timed out (possible rate limit). Trying next account..."

This ensures automatic failover when Antigravity hangs due to rate limiting.
2026-01-06 05:20:01 +00:00