OpenClaw
OpenClaw is a self-hosted AI assistant that runs on your own computer and answers you inside the chat apps you already use: WhatsApp, Telegram, Slack, Discord, iMessage, Signal, Teams and more than 20 others. A single local Gateway manages sessions, tools and channel connections, and the same setup works as a personal assistant on a laptop or as a shared assistant for a team.
It is a good fit for developers and technically comfortable users who want an always-available assistant without handing their data to a hosted service. State, memory and credentials stay on your hardware, and the model layer is pluggable, so you can use Claude, Codex or local models. Typical setups include messaging your assistant from your phone while away from your desk, running one shared assistant in a team Slack or Discord, and extending it with community skills and plugins from ClawHub.
Getting started: the official installer (a shell script for macOS, Linux and WSL2, or a PowerShell script on Windows) provisions Node.js and launches an onboarding wizard that checks model access, creates the workspace and configures the Gateway. If you already manage Node.js 24.16 or newer, you can install the npm package and run `openclaw onboard --install-daemon` instead. `openclaw dashboard` then opens the Control UI, where you can send a first test message.
Features
Why choose it
Trade-offs
Maintenance check
- Last commit
- 2026-10-24
- Commits, last 90 days
- 100+
- Releases, last 90 days
- 30
- Latest release
- v2026.10.7 (2026-10-24)
- Open issues
- 7808
- Recent issues with at least one comment
- 8/8
Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.
Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.
Most discussed open issues
- #143524 [Bug]: Agent SQLite WAL grows to 1.4–2.8 GB in days despite wal_autocheckpoint=1000; blocks gateway startup (Windows, 2026.9.2/9.3)134 comments · 2026-09-09
- #116201 Realtime voice work can retain unbounded provider and consult state59 comments · 2026-07-30
- #149361 Umbrella: WebUI performance and stability51 comments · 2026-09-15
- #160959 [Bug]: Gateway blocks for minutes while capturing large external plugins (2026.9.6 regression)50 comments · 2026-09-29
Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.
Who it fits
- You want a self-hosted personal AI assistant that lives inside 20+ messaging apps you already use (WhatsApp, Telegram, Slack, Discord, iMessage, Signal, Teams).
- You want your state, memory, and credentials kept on your own machine, with optional companion apps on macOS, iOS, Android, Windows, and Linux for voice and screen actions.
- You want a zero-ops cloud-hosted chatbot without managing a local Node.js 24+ Gateway and API keys.
- You only want a desktop multi-provider chat GUI on your computer rather than messaging-channel gateways; Cherry Studio is simpler for desktop chat.
How it differs
Both are MIT-licensed personal AI assistants connected to chat platforms. OpenClaw runs a Node.js Gateway supporting 20+ global messaging channels plus native desktop/mobile companion apps; CowAgent (chatgpt-on-wechat) is written in Python with deep support for WeChat, Feishu, DingTalk, and WeCom.
vs deer-flow★ 84.5kOpenClaw is a personal AI assistant centred on multi-channel messaging and device companion nodes; DeerFlow 2.0 is a LangGraph super-agent runtime centred on sandboxed code execution, sub-agents, and deep research.
Compatibility
Quick start
Use cases
Alternatives
Featured in these lists
Comments
No comments yet. Be the first!