Our take
Activeregular commits and releases- Activity
- 90
- Community
- 74
- Issues
- 84
- Pull requests
- 78
+2,363 stars in the last 30 days
- Releases more often than 93% of the projects we track
- Merges more pull requests than 84% of the projects we track
- Closes issues faster than 81% of the projects we track
OpenClaw is an actively maintained, configurable personal-assistant platform with broad channel support, but operating it safely requires care around inbound messages and host-level tools.
Good fit if
- You want an assistant connected to several messaging services and are comfortable managing a local Gateway.
- You value keeping assistant state and credentials on your hardware and choosing your own model provider.
- You need an actively developed TypeScript project with a large contributor base and frequent releases.
Look elsewhere if
- You want a hosted assistant that does not require you to install and operate the Gateway.
- You need tools to run in a sandbox by default, without configuring sandboxing yourself.
- You need support for a runtime outside the documented Node.js versions.
All health signals
| Last commit | 2026-10-10 (today) |
|---|---|
| Commits, last 90 days | 100+ |
| Releases, last 12 months | 100+ (latest v2026.9.9, 2026-10-08) |
| Contributors | 100+ (top contributor: 60% of commits) |
| Issues closed, last 90 days | 13+ (typically closed in 0 days) |
| Pull requests merged, last 90 days | 75 (typically merged in 0 days) |
| Project age | 11 months |
Checked on 2026-10-10 with the GitHub API.
Project history
Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.
Stars
394,404Forks
82,891Change since Nov 13, 2026: -4 stars·+6 forks
Overview
OpenClaw is a self-hosted AI assistant that connects model providers and tools to messaging services and companion apps. A local Gateway coordinates sessions, tools, events, and channel connections, so you can interact with the assistant through services you already use rather than a single dedicated chat interface.
It is aimed at people who want to keep assistant state, memory, and credentials on their own hardware while choosing their model provider and channels. The same Gateway can serve a personal setup or a shared team deployment, but tools may run directly on the host unless sandboxing is configured.
Key Features
- Connects to messaging channels including Discord, Slack, Telegram, WhatsApp, iMessage, and others.
- Provides a Gateway control plane for sessions, tools, events, and channel connections.
- Offers a Control UI, CLI, and terminal UI for operating the Gateway.
- Supports hosted and local model providers, with models and agent harnesses designed to be swappable.
- Extends assistant capabilities through tools, skills, and plugins.
- Companion apps and device nodes can provide voice, camera, screen, Canvas, and device-local actions on supported platforms.
- Stores state, memory, and credentials on the operator's hardware, with optional diagnostics export and opt-in anonymous feature statistics.
Use Cases
- Individuals can run an assistant on a personal computer and reach it through their existing messaging apps.
- Teams can use a shared Gateway deployment when they want a common assistant setup across multiple users.
- Developers can combine different model providers with plugins and tools without replacing the channel integrations.
- Users with supported companion devices can add voice or device-local actions to a chat-based assistant.
What you need
Detected in the repository
- Node.js >=24.16.0 <25 || >=26.1.0 (from package.json)
- A Dockerfile, so it can run in a container; a Compose file sets up the related services
- A test suite and automated checks on GitHub Actions
License in plain words
MITpermissive
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the copyright and license notice
- Share your changes: no
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
Install on macOS, Linux, or WSL2, then follow the onboarding wizard:
curl -fsSL https://openclaw.ai/install.sh | bash
See the README and getting started guide for Windows installation, package installation, channel setup, and configuration details.
Alternatives
- deepchat: DeepChat is a desktop client for model chats and agent workflows, while OpenClaw runs a configurable assistant through messaging channels and device companions.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| openclaw | TypeScript | MIT | 392k | Active |
| deepchat | TypeScript | Apache-2.0 | 6.4k | Active |
Considerations
- The documented Node.js requirement is
>=24.16.0 <25 || >=26.1.0. Docker and Compose deployment paths are available, but they are not required for a local install. - In the main session, tools run on the host unless sandboxing is configured. The project advises treating inbound messages as untrusted and consulting its security, exposure, and sandboxing guides before connecting other users or exposing the Gateway remotely.
- Assistant requests go to the model provider and chat platforms you configure. The README describes a daily version check by default, with an option to disable it, and says anonymous feature statistics are opt-in.
- The repository is a pnpm workspace. Its README says plain
npm installat the repository root is unsupported. - Frequent commits and releases indicate active development, but teams should account for operating a configurable system with multiple integrations and review its security guidance before deployment.
Found this useful?
Share it with someone who would like openclaw.