thinkfleet-engine
Advanced tools
ThinkFleet Engine is an open-source AI agent runtime with a plugin SDK. It connects to 15+ messaging channels (WhatsApp, Telegram, Slack, Discord, Google Chat, Signal, iMessage, Microsoft Teams, WebChat, and more), supports 340+ skills, and runs on your own devices — self-hosted and fully under your control.
The engine is a standalone product: install it, point it at your LLM provider, and you have a fully functional AI assistant. For teams and businesses, the optional ThinkFleet SaaS platform adds managed credentials, billing, multi-agent orchestration, and a hosted control plane via a plugin that snaps into the engine.
thinkfleet-engine (open source, MIT)
+ Plugin SDK: registerTool(), registerChannel(), registerHook(), registerService(), ...
+ 15+ built-in channels, 340+ skills, browser/canvas/cron/media tools
+ Gateway WebSocket control plane
@thinkfleet/saas-connector (optional, proprietary plugin)
+ Managed API key vault (100+ providers)
+ Token budget enforcement
+ Multi-agent orchestration (crews)
+ OAuth marketplace (800+ services via Composio)
+ Virtual card provisioning
Docs · Getting Started · Plugin SDK · Channels · Docker
Preferred setup: run the onboarding wizard (thinkfleet-engine onboard). It walks through gateway, workspace, channels, and skills. Works on macOS, Linux, and Windows (via WSL2).
Works with npm, pnpm, or bun.
Subscriptions (OAuth):
Model note: while any model is supported, I strongly recommend Anthropic Pro/Max (100/200) + Opus 4.5 for long‑context strength and better prompt‑injection resistance. See Onboarding.
Runtime: Node ≥22.
npm install -g thinkfleet-engine@latest
# or: pnpm add -g thinkfleet-engine@latest
thinkfleet-engine onboard --install-daemon
The wizard installs the Gateway daemon (launchd/systemd user service) so it stays running.
Legacy CLI name thinkfleet remains available as a compatibility alias.
Runtime: Node ≥22.
Full beginner guide (auth, pairing, channels): Getting started
thinkfleet-engine onboard --install-daemon
thinkfleet-engine gateway --port 18789 --verbose
# Send a message
thinkfleet-engine message send --to +1234567890 --message "Hello"
# Talk to the assistant
thinkfleet-engine agent --message "Ship checklist" --thinking high
Upgrading? Updating guide (and run thinkfleet-engine doctor).
vYYYY.M.D or vYYYY.M.D-<patch>), npm dist-tag latest.vYYYY.M.D-beta.N), npm dist-tag beta (macOS app may be missing).main, npm dist-tag dev (when published).Switch channels (git + npm): thinkfleet-engine update --channel stable|beta|dev.
Details: Development channels.
Prefer pnpm for builds from source.
git clone https://github.com/rrader26/thinkfleet-engine.git
cd thinkfleet-engine
pnpm install
pnpm ui:build # auto-installs UI deps on first run
pnpm build
pnpm thinkfleet-engine onboard --install-daemon
# Dev loop (auto-reload on TS changes)
pnpm gateway:watch
Note: pnpm thinkfleet-engine ... runs TypeScript directly (via tsx). pnpm build produces dist/ for running via Node.
ThinkFleet Engine connects to real messaging surfaces. Treat inbound DMs as untrusted input.
Full security guide: Security
Default behavior on Telegram/WhatsApp/Signal/iMessage/Microsoft Teams/Discord/Google Chat/Slack:
dmPolicy="pairing" / channels.discord.dm.policy="pairing" / channels.slack.dm.policy="pairing"): unknown senders receive a short pairing code and the bot does not process their message.thinkfleet-engine pairing approve <channel> <code> (then the sender is added to a local allowlist store).dmPolicy="open" and include "*" in the channel allowlist (allowFrom / channels.discord.dm.allowFrom / channels.slack.dm.allowFrom).Run thinkfleet-engine doctor to surface risky/misconfigured DM policies.
main for direct chats, group isolation, activation modes, queue modes, reply-back. Group rules: Groups.WhatsApp / Telegram / Slack / Discord / Google Chat / Signal / iMessage / BlueBubbles / Microsoft Teams / Matrix / Zalo / Zalo Personal / WebChat
│
▼
┌───────────────────────────────┐
│ Gateway │
│ (control plane) │
│ ws://127.0.0.1:18789 │
└──────────────┬────────────────┘
│
├─ Pi agent (RPC)
├─ CLI (thinkfleet-engine…)
├─ WebChat UI
├─ macOS app
└─ iOS / Android nodes
location.get, notifications, plus macOS‑only system.run/system.notify.ThinkFleet Engine 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 (ThinkFleet Engine enforces this).gateway.auth.mode: "password" or gateway.auth.allowTailscale: false.gateway.auth.mode: "password" is set.gateway.tailscale.resetOnExit to undo Serve/Funnel on shutdown.Details: Tailscale guide · Web surfaces
It’s 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 device‑local actions when needed.
system.run, camera, screen recording, notifications) via node.invoke.
In short: exec runs where the Gateway lives; device actions run where the device lives.Details: Remote access · Nodes · Security
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 you’ll 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:
/elevated on|off to toggle per‑session elevated access when enabled + allowlisted.sessions.patch (WS method) alongside thinkingLevel, verboseLevel, model, sendPolicy, and groupActivation.Details: Nodes · macOS app · Gateway protocol
sessions_list — discover active sessions (agents) and their metadata.sessions_history — fetch transcript logs for a session.sessions_send — message another session; optional reply‑back ping‑pong + announce step (REPLY_SKIP, ANNOUNCE_SKIP).Details: Session tools
ThinkFleet Skills Hub is a minimal skill registry. With the Skills Hub enabled, the agent can search for skills automatically and pull in new ones as needed.
Send these in WhatsApp/Telegram/Slack/Google Chat/Microsoft Teams/WebChat (group commands are owner-only):
/status — compact session status (model + tokens, cost when available)/new or /reset — reset the session/compact — compact session context (summary)/think <level> — off|minimal|low|medium|high|xhigh (GPT-5.2 + Codex models only)/verbose on|off/usage off|tokens|full — per-response usage footer/restart — restart the gateway (owner-only in groups)/activation mention|always — group activation toggle (groups only)The Gateway alone delivers a great experience. All apps are optional and add extra features.
If you plan to build/run companion apps, follow the platform runbooks below.
Note: signed builds required for macOS permissions to stick across rebuilds (see docs/mac/permissions.md).
thinkfleet-engine nodes ….Runbook: iOS connect.
~/thinkfleet (configurable via agents.defaults.workspace).AGENTS.md, SOUL.md, TOOLS.md.~/thinkfleet/skills/<skill>/SKILL.md.Minimal ~/.thinkfleet/thinkfleet.json (model + defaults):
{
agent: {
model: "anthropic/claude-opus-4-5"
}
}
Full configuration reference (all keys + examples).
agents.defaults.sandbox.mode: "non-main" to run non‑main sessions (groups/channels) inside per‑session Docker sandboxes; bash then runs in Docker for those sessions.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
pnpm thinkfleet-engine channels login (stores creds in ~/.thinkfleet/credentials).channels.whatsapp.allowFrom.channels.whatsapp.groups is set, it becomes a group allowlist; include "*" to allow all.TELEGRAM_BOT_TOKEN or channels.telegram.botToken (env wins).channels.telegram.groups (with channels.telegram.groups."*".requireMention); when set, it is a group allowlist (include "*" to allow all). Also channels.telegram.allowFrom or channels.telegram.webhookUrl as needed.{
channels: {
telegram: {
botToken: "123456:ABCDEF"
}
}
}
SLACK_BOT_TOKEN + SLACK_APP_TOKEN (or channels.slack.botToken + channels.slack.appToken).DISCORD_BOT_TOKEN or channels.discord.token (env wins).commands.native, commands.text, or commands.useAccessGroups, plus channels.discord.dm.allowFrom, channels.discord.guilds, or channels.discord.mediaMaxMb as needed.{
channels: {
discord: {
token: "1234abcd"
}
}
}
signal-cli and a channels.signal config section.channels.imessage.groups is set, it becomes a group allowlist; include "*" to allow all.msteams config section.msteams.allowFrom; group access via msteams.groupAllowFrom or msteams.groupPolicy: "open".Browser control (optional):
{
browser: {
enabled: true,
color: "#FF4500"
}
}
Use these when you’re past the onboarding flow and want the deeper reference.
ThinkFleet Engine was built for Molty, a space lobster AI assistant. 🦞 by Peter Steinberger and the community.
See CONTRIBUTING.md for guidelines, maintainers, and how to submit PRs. AI/vibe-coded PRs welcome! 🤖
Special thanks to Mario Zechner for his support and for pi-mono.
Thanks to all clawtributors:
![google-labs-jules[bot]](https://avatars.githubusercontent.com/in/842251?v=4&s=48 "google-labs-jules[bot]")
![dependabot[bot]](https://avatars.githubusercontent.com/in/29110?v=4&s=48 "dependabot[bot]")
![blacksmith-sh[bot]](https://avatars.githubusercontent.com/in/807020?v=4&s=48 "blacksmith-sh[bot]")
FAQs
Open-source AI agent runtime engine with plugin SDK
The npm package thinkfleet-engine receives a total of 7 weekly downloads. As such, thinkfleet-engine popularity was classified as not popular.
We found that thinkfleet-engine demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
OpenAI rotated macOS signing certificates after a malicious Axios package reached its CI pipeline in a broader software supply chain attack.
Security News
Open source is under attack because of how much value it creates. It has been the foundation of every major software innovation for the last three decades. This is not the time to walk away from it.
Security News
Socket CEO Feross Aboukhadijeh breaks down how North Korea hijacked Axios and what it means for the future of software supply chain security.