OpenClaw OpenClaw blog
Security writeups, release context, model/runtime notes, and the occasional lobster-shaped aside.
Review, revise, and apply proposed skills before they change how OpenClaw agents work.
Gideon Adegbesan Every ClawHub skill now ships with a Skill Card documenting what the skill does and where it came from, and is scanned by SkillSpector for hidden instructions and other agentic risks
OpenClaw is adding opt-in auto mode for Enterprise-ready host exec guardrails: policy runs first, low-risk misses get reviewed, and humans stay in the loop.
A release sweep across February through May shows faster agent turns, fewer dependencies, and a cleaner package shape.
The security roadmap for making OpenClaw a powerful personal assistant runtime users can understand, observe, and trust.
OpenClaw now runs OpenAI agent turns through the native Codex app-server harness by default, while bringing the lessons back to every model.
What happened around the 2026.4.24 and 2026.4.29 releases, why the direction was right, and what we are changing now.
The work you do not see behind the world's most-watched open source personal AI agent.
ClawHub skills are now scanned by VirusTotal's threat intelligence platform—bringing industry-leading security to the AI agent ecosystem.
The journey from Clawd to Moltbot to OpenClaw—and why this name is here to stay.