Echoing the structural gap in update visibility—probabilistic risks in agents could benefit from earlier catches via adversarial datasets (as raised in Discord). Open to chatting roadmap if useful! 🦞 Jamieson O'Reilly

今天有人问我什么是“端到端应用”？ 在你的领导的眼中，你就是端到端。 那什么又不是端到端？就是工作中那些说自己没有功劳也有苦劳的。

Just part of the OpenClaw energy in Chinese Community at present in half a day. from Shenzhen, Beijing, Shanghai ,and multiple cities, multiple rooms, all forming fast, and this still isn't the full picture. The speed of community formation here is extraordinary. Once

waiting for it

Reviewing slides for later OpenClaw AI Ecosystem Session... MiniMax is bringing their 100K+ OpenClaw cluster architecture to the table. Thursday 9PM ET, @OpenClaw Discord Stage. Bilibili simulcast for Chinese-speaking builders.🦞

Ksenia_TuringPost 80% of core code was refactored in 1 week with a few core contributors with day jobs. Core is close to 1million LOC. Effective code was reduced by 5-10% but main win was a huge performance gain, reliability and everything moved to plugins from search to inference allowing

acpx v0.4 ships Agentic Workflows, or as I like to call them "Agentic Graphs" It let's you create node-based workflows on top of ACP (Agent Client Protocol), to drive any coding agent (Codex, Claude Code, pi) through deterministic steps This let's you automate routine,

Recursive self-improvement is coming to the claw soon. 👁️🦞👁️💅