OpenClaw
Open-source self-hosted personal AI agent framework with a public skill registry (ClawHub) that runs across multiple messaging channels.
Last release Jun 10, 2026 · Last commit Jun 7, 2026 · ★ 377,307
Overview
OpenClaw (formerly Clawd/Moltbot) is an open-source, locally-running AI assistant that connects to messaging channels like WhatsApp, Telegram, Discord, Slack, Signal, and iMessage. Skills are distributed as a SKILL.md plus supporting files via the ClawHub registry, and crypto runtimes such as Bankr publish OpenClaw-compatible skills. It is a general-purpose agent framework, not crypto-native; onchain capability comes from third-party skills. Widely-cited star and skill counts could not be confirmed from official sources and are not asserted here.
Details
- Chains
- Unknown
- Agent types
- Personal Assistant, Agent Framework, Skill Runtime
- Open source
- Yes
- Pricing
- Free / self-hosted (open source)
- Last checked
- 2026-06-10
Marketplace Signals
- ▸ Functional agent
Verification status reflects evidence reviewed by Sato Hub. Self-reported claims are never presented as verified.
Recent Activity
All news →Releases and posts from OpenClaw, pulled from its GitHub and X.
openclaw v2026.6.10-alpha.1
openclaw 2026.6.10-alpha.1
openclaw v2026.6.9-alpha.3
openclaw 2026.6.9-alpha.3
OpenClaw 2026.6.5 🦞 🔎 Parallel web search bundled 🛡️ security.installPolicy for skill/plugin installs 🎙️ Matrix voice + thread fixes 🧠 Anthropic/MCP recovery hardened Less chaos. More claws. http
OpenClaw 2026.6.5 🦞 🔎 Parallel web search bundled 🛡️ security.installPolicy for skill/plugin installs 🎙️ Matrix voice + thread fixes 🧠 Anthropic/MCP recovery hardened Less chaos. More claws. https://t.co/eXevfEsApf
openclaw 2026.6.5-beta.6
Highlights QQBot now strips model reasoning/thinking scaffolding before native delivery, preventing raw content from leaking into channel replies. ( #89913 , #90132 ) Thanks @openperf . MCP tool results now coerce resource_link , resource , audio , malformed image, and future non-text/image blocks at the materialize boundary, preventing Anthropic 400s and poisoned session history after a tool retu
openclaw v2026.6.9-alpha.1
openclaw 2026.6.9-alpha.1
OpenClaw 2026.6.5-beta.3
2026.6.5 Highlights QQBot now strips model reasoning/thinking scaffolding before native delivery, preventing raw content from leaking into channel replies. ( #89913 , #90132 ) Thanks @openperf . MCP tool results now coerce resource_link , resource , audio , malformed image, and future non-text/image blocks at the materialize boundary, preventing Anthropic 400s and poisoned session history after a
Related Resources
ElizaOS
ActiveOpen-source TypeScript framework for building crypto-native multi-agent systems.
GOAT SDK
ActiveCrossmint's open-source library of onchain actions (tools) for AI agents across chains.
Solana Agent Kit
ActiveSendAI's open-source toolkit connecting AI agents to Solana protocol actions.
Join the Sato Hub Briefing
Get the weekly briefing on onchain agents, MCPs, skill repos, tools, and autonomous crypto infrastructure.