Free & Open Source
$0
MIT license, self-hosted
OpenClaw
Your own personal AI assistant. Any OS. Any platform. The lobster way. 🦞
An open-source, self-hosted personal AI assistant from Peter Steinberger and community. Runs on macOS, iOS, and Android and speaks on the channels you already use — WhatsApp, Telegram, Slack, Discord, iMessage, and more. Highly extensible: skills are Markdown files, with a Codex harness extension and MCP support for coding workflows.
+ Pros
- Fully self-hosted and MIT-licensed — complete data privacy, no code sent to third-party servers
- Multi-channel inbox: WhatsApp, Telegram, Slack, Discord, iMessage, Signal, Teams, Matrix, and 20+ more
- Skills as plain Markdown files — easy to create, share, and version-control custom capabilities
- Provider-agnostic: works with Claude, GPT, Gemini, and local models
- 100K+ GitHub stars — largest self-hosted personal AI assistant community
− Cons
- Requires self-hosting setup — not suitable for non-technical users
- No managed cloud option — infrastructure, updates, and maintenance are the user's responsibility
- Multi-channel integrations require individual API keys and configuration per platform
- Younger project — fewer built-in integrations than commercial tools like Cursor or Copilot
Pricing
OpenClaw is an open-source personal AI assistant you run on your own devices. Started in late 2025 by Peter Steinberger and a fast-growing community, it crossed 100,000 GitHub stars within its first week and quickly became the reference implementation for self-hosted agents. The framing is deliberate: not a chatbot, not a coding tool, but a single personal agent that lives on the channels you already use — WhatsApp, Telegram, Slack, Discord, Signal, iMessage, Microsoft Teams, Matrix, and many more — with native presence on macOS, iOS, and Android.
Although primarily a personal assistant, OpenClaw is highly relevant to coding workflows through its skills system: each skill is a folder with a SKILL.md file (Markdown frontmatter + instructions), and the project ships extensions for the OpenAI Codex harness and other coding tools. Custom skills can review GitHub PRs, run tests, manage your issue tracker, or orchestrate other CLIs. Multi-agent routing lets one OpenClaw gateway run several specialized agents — for code, research, ops — in parallel, each with its own workspace and memory.
Key Features
• Self-hosted personal AI assistant, MIT-licensed
• Multi-channel inbox: WhatsApp, Telegram, Slack, Discord, iMessage, Signal, Teams, Matrix, and ~20 more
• Native presence on macOS, iOS, and Android — voice in and out
• Skills as Markdown files (SKILL.md) inside ~/.openclaw/workspace/skills
• Codex harness extension for OpenAI Codex integration
• Multi-agent routing: one gateway, many isolated agents
• Persistent memory in plain Markdown + SQLite
• MCP server support and BYO LLM (Claude, GPT, Gemini, local)
• Built-in audit tool (`openclaw doctor`) and permission policies
Version History
Faster Gateway and replies with reduced startup scans; core transcript capture for meeting summaries; Telegram forum topics, iMessage attachment roots; realtime Talk inspection.
Discord voice follow-users with multi-user handoff; xAI device-code OAuth login; OpenRouter provider-level routing policy.
Mac app redesign with Settings cards; CLI tool plugin SDK (build/validate/init); browser modal dialog handling; Android realtime Talk Mode.
xAI Grok OAuth for SuperGrok subscribers; CLI cron run --wait with timeout; setup wizard localization (EN/ZH).
Signature Snippet
Create a skill at ~/.openclaw/workspace/skills/github-pr-reviewer/SKILL.md with simple Markdown instructions. From any connected channel — say a Telegram chat — send "review github.com/me/repo/pull/42" and OpenClaw fetches the diff, analyses it, and replies with a structured review.More in this Space
Vix
Vix is a Go-native, open-source (AGPL-3.0) AI coding agent that slashes token costs by 40-50% using a stem agent architecture and Tree-sitter virtual filesystem. It rethinks the plan/execute loop — keeping LLM cache warm across Explore, Plan, and Execute phases — while shipping Programmable Workflows, Whiteboard Mode with voice AI, MCP server support, and a self-evolving agent that writes its own scheduled jobs and watchers.
Late — High-Leverage AI Agent Orchestration
Orchestrate an entire AI dev team on 5GB VRAM using ephemeral subagents, exact-match diffs, and a zero-dependency Go binary. Works with any OpenAI-compatible model — local or cloud.
Paca
Paca is a free, open-source, self-hosted Scrum board where AI agents work as equal teammates — assigned to sprints, picking up tasks, and collaborating on BDD specs alongside humans. Built as an alternative to Jira and Linear, it treats AI agents as first-class Scrum members.