# FastClaw

> A lightweight AI agent runtime written in Go — single binary, any LLM, multi-agent, sandbox, cloud-ready.


## Home

- [Home](https://fastclaw.ai): FastClaw is an agent factory: create, manage and run AI agents — each with its own personality, memory, skills and tools — and put them on the web, your API, or nine chat apps.

## Features

- [Features](https://fastclaw.ai/features): Agents, models, channels, sandboxes, skills, an OpenAI-compatible API and a scheduler — all in a single Go binary.
- [An agent factory: many agents, each with its own soul](https://fastclaw.ai/features/agents): Create and run many AI agents from one FastClaw install — each with its own SOUL.md personality, memory, skills, model and sharing settings.
- [Bring any LLM — and switch per agent](https://fastclaw.ai/features/models): Run FastClaw agents on OpenAI, Anthropic, DeepSeek, Ollama, OpenRouter, Groq, Mistral or any OpenAI-compatible API, with per-agent overrides and retries.
- [Put your AI agent in every chat app your users already use](https://fastclaw.ai/features/channels): Bind any FastClaw agent to Telegram, Discord, Slack, Feishu, WeCom, WeChat, WhatsApp, LINE or iMessage — with per-chat session isolation and owner pairing.
- [Let agents run real code — inside a sandbox](https://fastclaw.ai/features/sandbox): Built-in exec, file and web tools, run in an E2B cloud sandbox or a Docker container, with automatic skill and workspace hydration and file sync.
- [Teach agents new skills in one click](https://fastclaw.ai/features/skills): Install skills from ClawHub, skills.sh or GitHub, connect MCP servers, add JSON-RPC plugins and let agents read Notion, GitHub and Linear.
- [An OpenAI-compatible API for your agents](https://fastclaw.ai/features/api): Drive FastClaw agents from any app: OpenAI-compatible chat completions, agent management, per-end-user isolation, usage reporting and ACP.
- [Agents that work while you sleep](https://fastclaw.ai/features/scheduler): FastClaw agents schedule their own jobs — one-off reminders, intervals or cron expressions — and deliver results to the chat that asked.

## Use cases

- [Use cases](https://fastclaw.ai/use-cases): See how SaaS teams, developers, companies, community owners and agencies run their agents on FastClaw.
- [AI agent backend for SaaS builders](https://fastclaw.ai/use-cases/for-saas-builders): Give every customer their own AI agent without building a runtime. FastClaw handles agents, end-user isolation, sandboxed tools and usage metering over an API.
- [Self-hosted AI assistant for developers](https://fastclaw.ai/use-cases/for-developers): Run your own AI agents on your laptop or server with one Go binary. Any model, local or hosted, reachable from the terminal, the web or your chat apps.
- [Internal AI agents for teams on Slack, Feishu and WeCom](https://fastclaw.ai/use-cases/for-teams): Deploy shared AI agents inside the chat tools your team already uses. Self-hosted, with per-person sessions, shared skills and scheduled reports.
- [AI chatbot for Telegram, Discord and LINE communities](https://fastclaw.ai/use-cases/for-communities): Run an AI agent for your community on Telegram, Discord or LINE. Every member gets their own session, and untrusted chatters stay inside a sandbox.
- [AI agent platform for agencies building bots for clients](https://fastclaw.ai/use-cases/for-agencies): Build an agent once, fork it for each client, and deliver it on WeChat, WhatsApp or the web — all from one FastClaw deployment with per-client quotas.

## Compare

- [Compare](https://fastclaw.ai/compare): Side-by-side comparisons with other agent tools — what each does best, and when to pick which.
- [FastClaw vs OpenClaw: which agent runtime fits you?](https://fastclaw.ai/compare/fastclaw-vs-openclaw): OpenClaw is a personal AI assistant you run yourself. FastClaw is a Go agent runtime for many users and apps. Compare install, multi-tenancy, API, license.
- [The OpenClaw alternative for multi-user and app-embedded agents](https://fastclaw.ai/compare/openclaw-alternative): Looking for an OpenClaw alternative? FastClaw is a single-binary Go runtime for serving agents to many users and apps — and it reads OpenClaw skills.
- [FastClaw vs Dify: agent runtime or LLM app platform?](https://fastclaw.ai/compare/fastclaw-vs-dify): Dify is a visual platform for LLM workflows and RAG apps. FastClaw is a lightweight runtime for autonomous agents in chat apps and products. Compare the two.

## Integrations

- [Integrations](https://fastclaw.ai/integrations): Model providers, chat channels and runtime infrastructure that FastClaw supports out of the box.
- [Use OpenAI GPT models with FastClaw](https://fastclaw.ai/integrations/openai): Run FastClaw agents on OpenAI models — set a deployment-wide default, override per agent, and expose your agents through an OpenAI-compatible API.
- [Use Anthropic Claude models with FastClaw](https://fastclaw.ai/integrations/anthropic): Run FastClaw agents on Claude through Anthropic's native Messages API — with prompt caching, extended thinking and image input preserved.
- [Use DeepSeek models with FastClaw](https://fastclaw.ai/integrations/deepseek): Run FastClaw agents on DeepSeek — a built-in preset, thinking-mode support and per-agent model overrides for low-cost, capable agents.
- [Run local Ollama models as FastClaw agents](https://fastclaw.ai/integrations/ollama): Point FastClaw at Ollama to run AI agents on local, open-weight models — no API key, no data leaving your machine.
- [Use hundreds of models via OpenRouter with FastClaw](https://fastclaw.ai/integrations/openrouter): Connect FastClaw to OpenRouter and give every agent access to models from many vendors with a single API key.
- [Build a Telegram AI agent bot with FastClaw](https://fastclaw.ai/integrations/telegram): Turn any FastClaw agent into a Telegram bot in minutes: paste a BotFather token, pair your account, and chat in DMs or groups.
- [Build a Discord AI agent bot with FastClaw](https://fastclaw.ai/integrations/discord): Connect a FastClaw agent to Discord: answer DMs and @-mentions in servers, with per-channel sessions, memory and sandboxed tools.
- [Add an AI agent to Slack with FastClaw](https://fastclaw.ai/integrations/slack): Connect a FastClaw agent to Slack via Socket Mode — answer DMs and @-mentions in channels, no public URL required.
- [Build a Feishu (Lark) AI agent bot with FastClaw](https://fastclaw.ai/integrations/feishu): Create a Feishu bot by scanning a QR code and connect it to a FastClaw agent — long-connection mode, no public URL needed.
- [Build a WeCom AI agent bot with FastClaw](https://fastclaw.ai/integrations/wecom): Connect a FastClaw agent to a WeCom (企业微信) smart bot over a long connection — create it by QR code, no public URL needed.
- [Build a WeChat AI agent with FastClaw](https://fastclaw.ai/integrations/wechat): Bind a WeChat account to a FastClaw agent by scanning a QR code — direct messages go to the agent and replies come back in WeChat.
- [Build a WhatsApp AI agent with FastClaw](https://fastclaw.ai/integrations/whatsapp): Link a WhatsApp number to a FastClaw agent by scanning a QR code, just like WhatsApp Web — the agent replies from that number.
- [Build a LINE AI agent bot with FastClaw](https://fastclaw.ai/integrations/line): Connect a FastClaw agent to the LINE Messaging API — 1:1 chats, group @-mentions, media in and out, and signed webhooks.
- [Answer iMessage with an AI agent using FastClaw](https://fastclaw.ai/integrations/imessage): Run FastClaw on a Mac and let an agent answer iMessage direct messages sent to that Mac's Apple ID.
- [Sandbox AI agent code execution with E2B and FastClaw](https://fastclaw.ai/integrations/e2b): Run every FastClaw agent command in an isolated E2B cloud sandbox — skills and workspace hydrated in, files synced back after each call.
- [Run FastClaw agents in Docker sandboxes](https://fastclaw.ai/integrations/docker): Use Docker as FastClaw's sandbox backend to isolate agent commands on your own hardware — or deploy FastClaw itself with Docker Compose.
- [Scale FastClaw with PostgreSQL](https://fastclaw.ai/integrations/postgres): Swap FastClaw's default SQLite file for PostgreSQL to run multiple pods behind a load balancer, with automatic schema migrations.
- [Connect MCP servers to FastClaw agents](https://fastclaw.ai/integrations/mcp): Give FastClaw agents external tools through the Model Context Protocol — add http or stdio MCP servers per agent from the console.

## Free tools

- [Free tools](https://fastclaw.ai/tools): Small, free utilities that run entirely in your browser. No sign-up.
- [SOUL.md generator](https://fastclaw.ai/tools/soul-md-generator): Generate a SOUL.md personality file for your AI agent: role, voice, boundaries and rules. Works with FastClaw and any agent that reads SOUL.md.
- [API snippet generator](https://fastclaw.ai/tools/api-snippet-generator): Build ready-to-paste code for an OpenAI-compatible agent chat endpoint, with per-end-user isolation and streaming. Works with FastClaw's /v1/chat/completions.

## Pricing

- [Pricing](https://fastclaw.ai/pricing): FastClaw is source-available. Run it yourself at no cost, let us host it, or license it for a multi-tenant product.

## Blog

- [Blog](https://fastclaw.ai/blog): Guides, tutorials and deep dives from the FastClaw team.

## Blog Posts

- [Deploy your own Telegram AI bot in 15 minutes](https://fastclaw.ai/blog/deploy-a-telegram-ai-bot): Step-by-step: install FastClaw, connect a model provider, give your agent a personality, bind it to a Telegram bot and pair it with your account — no public URL or webhook required.
- [How to add AI agents to your SaaS without building an agent platform](https://fastclaw.ai/blog/add-ai-agents-to-your-saas): A practical guide to giving every customer their own AI agent: user-scope API keys, per-user agents, end-user isolation, streaming, files and usage-based billing with the FastClaw /v1 API.
- [What is an AI agent runtime? (And how it differs from a framework or a chat app)](https://fastclaw.ai/blog/what-is-an-agent-runtime): Frameworks help you write an agent. Chat apps let you talk to one. An agent runtime runs many agents for real users — sessions, tools, sandboxes, memory and channels included.