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
2026-01-06 16:03:04 +01:00
2026-01-07 00:04:18 +01:00
2026-01-06 19:10:06 +00:00
2026-01-02 13:53:06 +02:00
2026-01-06 03:41:56 +01:00
2025-12-18 09:06:39 +01:00
2026-01-03 01:56:54 +00:00
2026-01-04 16:24:17 +01:00
2026-01-06 23:32:18 +00:00
2025-12-23 00:28:55 +00:00
2025-11-24 11:16:47 +01:00
2026-01-05 23:06:14 +01:00
2026-01-06 15:43:24 +01:00

🦞 CLAWDBOT — Personal AI Assistant

CLAWDBOT

EXFOLIATE! EXFOLIATE!

CI status GitHub release Discord MIT License

Clawdbot is a personal AI assistant you run on your own devices. It answers you on the providers you already use (WhatsApp, Telegram, Slack, Discord, Signal, iMessage, WebChat), can speak and listen on macOS/iOS/Android, and can render a live Canvas you control. The Gateway is just the control plane — the product is the assistant.

If you want a personal, single-user assistant that feels local, fast, and always-on, this is it.

Website · Docs · Getting Started: https://docs.clawd.bot/getting-started · Updating: https://docs.clawd.bot/updating · Showcase: https://docs.clawd.bot/showcase · FAQ: https://docs.clawd.bot/faq · Wizard: https://docs.clawd.bot/wizard · Nix: https://github.com/clawdbot/nix-clawdbot · Docker: https://docs.clawd.bot/docker · Discord: https://discord.gg/clawd

Preferred setup: run the onboarding wizard (clawdbot onboard). It walks through gateway, workspace, providers, and skills. The CLI wizard is the recommended path and works on macOS, Windows, and Linux. Works with npm, pnpm, or bun. New install? Start here: https://docs.clawd.bot/getting-started

Subscriptions (OAuth):

  • Anthropic (Claude Pro/Max)
  • OpenAI (ChatGPT/Codex)

Model note: while any model is supported, I strongly recommend Anthropic Pro/Max (100/200) + Opus 4.5 for longcontext strength and better promptinjection resistance. See Onboarding.

Models (selection + auth)

Do not download prebuilt binaries. Run from source.

Prefer pnpm for builds from source. Bun is optional for running TypeScript directly.

# Clone this repo
git clone https://github.com/clawdbot/clawdbot.git
cd clawdbot

pnpm install
pnpm ui:install
pnpm ui:build
pnpm build
pnpm clawdbot onboard

Note: pnpm clawdbot ... runs TypeScript directly (via tsx). pnpm build produces dist/ for running via Node / the packaged clawdbot binary.

Quick start (TL;DR)

Runtime: Node ≥22.

Full beginner guide (auth, pairing, providers): https://docs.clawd.bot/getting-started

pnpm clawdbot onboard

pnpm clawdbot gateway --port 18789 --verbose

# Dev loop (auto-reload on TS changes)
pnpm gateway:watch

# Send a message
pnpm clawdbot send --to +1234567890 --message "Hello from Clawdbot"

# Talk to the assistant (optionally deliver back to WhatsApp/Telegram/Slack/Discord)
pnpm clawdbot agent --message "Ship checklist" --thinking high

Upgrading? https://docs.clawd.bot/updating (and run clawdbot doctor).

If you run from source, prefer pnpm clawdbot … (not global clawdbot).

Security defaults (DM access)

Clawdbot connects to real messaging surfaces. Treat inbound DMs as untrusted input.

Full security guide: https://docs.clawd.bot/security

Default behavior on Telegram/WhatsApp/Signal/iMessage/Discord/Slack:

  • DM pairing (dmPolicy="pairing" / discord.dm.policy="pairing" / slack.dm.policy="pairing"): unknown senders receive a short pairing code and the bot does not process their message.
  • Approve with: clawdbot pairing approve --provider <provider> <code> (then the sender is added to a local allowlist store).
  • Public inbound DMs require an explicit opt-in: set dmPolicy="open" and include "*" in the provider allowlist (allowFrom / discord.dm.allowFrom / slack.dm.allowFrom).

Run clawdbot doctor to surface risky/misconfigured DM policies.

Highlights

Everything we built so far

Core platform

Providers

Apps + nodes

Tools + automation

Ops + packaging

How it works (short)

WhatsApp / Telegram / Slack / Discord / Signal / iMessage / WebChat
               │
               ▼
┌───────────────────────────────┐
│            Gateway            │  ws://127.0.0.1:18789
│       (control plane)         │  bridge: tcp://0.0.0.0:18790
└──────────────┬────────────────┘
               │
               ├─ Pi agent (RPC)
               ├─ CLI (clawdbot …)
               ├─ WebChat UI
               ├─ macOS app
               └─ iOS/Android nodes

Key subsystems

Tailscale access (Gateway dashboard)

Clawdbot can auto-configure Tailscale Serve (tailnet-only) or Funnel (public) while the Gateway stays bound to loopback. Configure gateway.tailscale.mode:

  • off: no Tailscale automation (default).
  • serve: tailnet-only HTTPS via tailscale serve (uses Tailscale identity headers by default).
  • funnel: public HTTPS via tailscale funnel (requires shared password auth).

Notes:

  • gateway.bind must stay loopback when Serve/Funnel is enabled (Clawdbot enforces this).
  • Serve can be forced to require a password by setting gateway.auth.mode: "password" or gateway.auth.allowTailscale: false.
  • Funnel refuses to start unless gateway.auth.mode: "password" is set.
  • Optional: gateway.tailscale.resetOnExit to undo Serve/Funnel on shutdown.

Details: Tailscale guide · Web surfaces

Remote Gateway (Linux is great)

Its perfectly fine to run the Gateway on a small Linux instance. Clients (macOS app, CLI, WebChat) can connect over Tailscale Serve/Funnel or SSH tunnels, and you can still pair device nodes (macOS/iOS/Android) to execute devicelocal actions when needed.

  • Gateway host runs the bash tool and provider connections by default.
  • Device nodes run devicelocal actions (system.run, camera, screen recording, notifications) via node.invoke. In short: bash runs where the Gateway lives; device actions run where the device lives.

Details: Remote access · Nodes · Security

macOS permissions via the Gateway protocol

The macOS app can run in node mode and advertises its capabilities + permission map over the Gateway WebSocket (node.list / node.describe). Clients can then execute local actions via node.invoke:

  • system.run runs a local command and returns stdout/stderr/exit code; set needsScreenRecording: true to require screen-recording permission (otherwise youll get PERMISSION_MISSING).
  • system.notify posts a user notification and fails if notifications are denied.
  • canvas.*, camera.*, screen.record, and location.get are also routed via node.invoke and follow TCC permission status.

Elevated bash (host permissions) is separate from macOS TCC:

  • Use /elevated on|off to toggle persession elevated access when enabled + allowlisted.
  • Gateway persists the persession toggle via sessions.patch (WS method) alongside thinkingLevel, verboseLevel, model, sendPolicy, and groupActivation.

Details: Nodes · macOS app · Gateway protocol

Agent to Agent (sessions_* tools)

  • Use these to coordinate work across sessions without jumping between chat surfaces.
  • sessions_list — discover active sessions (agents) and their metadata.
  • sessions_history — fetch transcript logs for a session.
  • sessions_send — message another session; optional replyback pingpong + announce step (REPLY_SKIP, ANNOUNCE_SKIP).

Details: Session tools

Skills registry (ClawdHub)

ClawdHub is a minimal skill registry. With ClawdHub enabled, the agent can search for skills automatically and pull in new ones as needed.

https://ClawdHub.com

Chat commands

Send these in WhatsApp/Telegram/Slack/WebChat (group commands are owner-only):

  • /status — health + session info (group shows activation mode)
  • /new or /reset — reset the session
  • /compact — compact session context (summary)
  • /think <level> — off|minimal|low|medium|high
  • /verbose on|off
  • /restart — restart the gateway (owner-only in groups)
  • /activation mention|always — group activation toggle (groups only)

macOS app (optional)

The Gateway alone delivers a great experience. All apps are optional and add extra features.

If you plan to build/run companion apps, initialize submodules first:

git submodule update --init --recursive
./scripts/restart-mac.sh

macOS (Clawdbot.app) (optional)

  • Menu bar control for the Gateway and health.
  • Voice Wake + push-to-talk overlay.
  • WebChat + debug tools.
  • Remote gateway control over SSH.

Note: signed builds required for macOS permissions to stick across rebuilds (see docs/mac/permissions.md).

iOS node (optional)

  • Pairs as a node via the Bridge.
  • Voice trigger forwarding + Canvas surface.
  • Controlled via clawdbot nodes ….

Runbook: iOS connect.

Android node (optional)

  • Pairs via the same Bridge + pairing flow as iOS.
  • Exposes Canvas, Camera, and Screen capture commands.
  • Runbook: Android connect.

Agent workspace + skills

  • Workspace root: ~/clawd (configurable via agent.workspace).
  • Injected prompt files: AGENTS.md, SOUL.md, TOOLS.md.
  • Skills: ~/clawd/skills/<skill>/SKILL.md.

Configuration

Minimal ~/.clawdbot/clawdbot.json (model + defaults):

{
  agent: {
    model: "anthropic/claude-opus-4-5"
  }
}

Full configuration reference (all keys + examples).

Security model (important)

  • Default: tools run on the host for the main session, so the agent has full access when its just you.
  • Group/channel safety: set agent.sandbox.mode: "non-main" to run nonmain sessions (groups/channels) inside persession Docker sandboxes; bash then runs in Docker for those sessions.
  • Sandbox defaults: allowlist bash, process, read, write, edit, sessions_list, sessions_history, sessions_send, sessions_spawn; denylist browser, canvas, nodes, cron, discord, gateway.

Details: Security guide · Docker + sandboxing · Sandbox config

WhatsApp

  • Link the device: pnpm clawdbot login (stores creds in ~/.clawdbot/credentials).
  • Allowlist who can talk to the assistant via whatsapp.allowFrom.
  • If whatsapp.groups is set, it becomes a group allowlist; include "*" to allow all.

Telegram

  • Set TELEGRAM_BOT_TOKEN or telegram.botToken (env wins).
  • Optional: set telegram.groups (with telegram.groups."*".requireMention); when set, it is a group allowlist (include "*" to allow all). Also telegram.allowFrom or telegram.webhookUrl as needed.
{
  telegram: {
    botToken: "123456:ABCDEF"
  }
}

Slack

  • Set SLACK_BOT_TOKEN + SLACK_APP_TOKEN (or slack.botToken + slack.appToken).

Discord

  • Set DISCORD_BOT_TOKEN or discord.token (env wins).
  • Optional: set commands.native, commands.text, or commands.useAccessGroups, plus discord.dm.allowFrom, discord.guilds, or discord.mediaMaxMb as needed.
{
  discord: {
    token: "1234abcd"
  }
}

Signal

  • Requires signal-cli and a signal config section.

iMessage

  • macOS only; Messages must be signed in.
  • If imessage.groups is set, it becomes a group allowlist; include "*" to allow all.

WebChat

  • Uses the Gateway WebSocket; no separate WebChat port/config.

Browser control (optional):

{
  browser: {
    enabled: true,
    controlUrl: "http://127.0.0.1:18791",
    color: "#FF4500"
  }
}

Docs

Use these when youre past the onboarding flow and want the deeper reference.

Advanced docs (discovery + control)

Operations & troubleshooting

Deep dives

Workspace & skills

Platform internals

Email hooks (Gmail)

Gmail Pub/Sub wiring (gcloud + gogcli), hook tokens, and auto-watch behavior are documented here.

Gateway auto-starts the watcher when hooks.enabled=true and hooks.gmail.account is set; clawdbot hooks gmail run is the manual daemon wrapper if you dont want auto-start.

clawdbot hooks gmail setup --account you@gmail.com
clawdbot hooks gmail run

Clawd

Clawdbot was built for Clawd, a space lobster AI assistant. 🦞
by Peter Steinberger and the community.

Community

See CONTRIBUTING.md for guidelines, maintainers, and how to submit PRs.
AI/vibe-coded PRs welcome! 🤖

Thanks to all clawtributors:

steipete thewilloftheshadow mcinteerj joshp123 joaohlisboa petter-b mukhtharcm dan-dr Nachx639 jeffersonwarrior mbelinky julianengel CashWilliams omniwired jverdi Syhids meaningfool rafaelreis-r wstock vsabavat scald sreekaransrinath ratulsarna osolmaz conhecendocontato hrdwdmrbl jayhickey jamesgroat gtsifrikas djangonavarro220 azade-c andranik-sahakyan adamgall jalehman jarvis-medmatic mneves75 regenrek tobiasbischoff MSch obviyus dbhurley Asleep123 Iamadig imfing kitze nachoiacovino VACInc cash-echo-bot claude kiranjd pcty-nextgen-service-account minghinmatthewlam ngutman onutc oswalpalash snopoke ManuelHettich loukotal hugobarauna AbhisekBasu1 emanuelst dantelex

Description
No description provided
Readme 149 MiB
Languages
TypeScript 82.5%
Swift 13.5%
Kotlin 1.9%
Shell 0.8%
CSS 0.5%
Other 0.8%