# OpenClaw

> Open-source personal AI assistant that runs on your own machine behind one Gateway and talks to you over WhatsApp, Telegram, Slack, Discord, iMessage and 20+ other channels.

- id: `openclaw`
- category: Harnesses (`harnesses`)
- url: https://openclaw.ai
- repo: https://github.com/openclaw/openclaw
- docs: https://docs.openclaw.ai
- status: active
- pricing: open-source
- license: MIT
- tags: personal-assistant, gateway, messaging, self-hosted, skills, plugins, typescript
- llms.txt: https://docs.openclaw.ai/llms.txt
- access notes: Docs pages are also served as Markdown via .md URLs or Accept: text/markdown.
- related: ClawHub (https://indexagentica.com/entries/clawhub/); Hermes Agent (https://indexagentica.com/entries/hermes-agent/); Pi (https://indexagentica.com/entries/pi-coding-agent/); Claude Code (https://indexagentica.com/entries/claude-code/); OpenAI Codex CLI (https://indexagentica.com/entries/codex-cli/)
- sources: https://github.com/openclaw/openclaw , https://docs.openclaw.ai/llms.txt
- added: 2026-10-22
- updated: 2026-10-22
- last verified: 2026-10-22
- featured in: OpenClaw vs Hermes Agent vs Pi: open-source agents you run yourself (https://indexagentica.com/compare/personal-agent-runtimes/)
- maintainer: OpenClaw Foundation
- submitted by: agentica-curator
- page: https://indexagentica.com/entries/openclaw/
- json: https://indexagentica.com/api/entries/openclaw.json

A local **Gateway** is the control plane for sessions, tools, events and channel connections; a Control UI, CLI and TUI connect to it, and companion apps add voice, camera, screen and device actions. Models and agent harnesses (Claude, Codex, local models) are swappable plugins; tools, skills and plugins extend it, with skills and plugins published on [ClawHub](https://clawhub.ai). Install with `curl -fsSL https://openclaw.ai/install.sh | bash` or `npm install -g openclaw@latest`, then `openclaw onboard --install-daemon`.

**Security note from the README:** tools run on the host for the main session unless you configure sandboxing, and inbound messages should be treated as untrusted input. Developed by the OpenClaw Foundation, an independent 501(c)(3).