What it does
OpenClaw is an open-source AI assistant that runs on your own computer and replies in the messaging services you already use, including Discord, iMessage, Slack, Teams, Telegram and WhatsApp, with 20+ more listed. A single Gateway runs it, either as a personal assistant on a laptop or as a shared team deployment. State, memory and credentials stay on your hardware. Models and agent harnesses, including Claude, Codex and local models, are plugins you can swap, and it works with hosted and local model providers. Tools, skills and plugins extend what the assistant can do, with plugins shared through ClawHub. The README says there is no paid tier, hosted service or token, and that the project is stewarded by the OpenClaw Foundation, an independent 501(c)(3). Install is a shell script for macOS, Linux and WSL2, a PowerShell script for Windows, or an npm package, followed by an onboarding wizard. It also advises treating inbound messages as untrusted and reading the security and sandboxing guides before connecting other users or exposing the Gateway.
Capabilities
- Connects to 20+ messaging services such as WhatsApp, Telegram, Slack, Discord, Teams and iMessage
- One Gateway for a personal setup or a shared team deployment
- Swappable models and agent harnesses, hosted or local
- Tools, skills and plugins, with plugins shared through ClawHub
- Native apps for macOS, iOS, Android, Windows and Linux
- Control UI, CLI and TUI that connect to the Gateway
When to use it
- You want a personal or team assistant that lives in your existing chat apps
- You need memory and credentials to stay on hardware you control
- You want to choose or switch between hosted and local models
- You are comfortable with a command-line install and an onboarding wizard
