Pinned
Friends of Clawd Discord is now live!
A place to hang out, share your Clawdis setups, ask questions, and help each other build cool stuff with AI assistants.
Join the crustacean crew:
OpenClaw🦞
793 posts
OpenClaw🦞
@openclaw
The AI that does things. Emails, calendar, home automation, from your favorite chat app. Your machine, your rules.
New shell, same lobster soul. 🦞
- Replying to @openclawControl UI starts faster and answers sooner: cached model metadata, no startup catalog wait, lazy slash commands, first-event tracing, and slow-reply diagnostics.
- Replying to @openclawMCP gets Streamable HTTP loopback transport, OAuth/SSE auth fixes, and broader schema compatibility, so external clients and OpenClaw-managed servers hit fewer weird edges.
- Replying to @openclawBrowser automation can attach to existing CDP sessions, validate discovered WebSockets, handle default-profile cdpUrl, and keep browser output boundaries safer.
- Replying to @openclawiMessage recovery is sturdier: always-on inbound restart, durable echo markers, block streaming, idle approval discovery, hardened outbound transport, and better startup diagnostics.
- Replying to @openclawTelegram gets less fussy: account-scoped topics route to the right agent, streamed text survives tool calls, /compact works from generic ingress, and unauthorized DM text stays out of context.
- Replying to @openclawSecurity pass is the headline under the headline: transcripts, sandbox binds, host env inheritance, MCP stdio, Codex HTTP access, loopback tools, and exec approvals got stricter.
- Replying to @openclawClaude Fable 5 lands with adaptive thinking support. Provider setup also picks up OpenRouter OAuth, Gemma 4 reasoning replay, and cleaner dynamic tool progress.
- OpenClaw 2026.6.6 🦞 🔒 Tighter security boundaries 💬 Safer Telegram + iMessage delivery 🧠 Claude Fable 5 + OpenRouter OAuth ⚡ Faster Control UI first replies Less weird, more work done.
- Replying to @openclawAnthropic extended-thinking sessions recover better after prompt-cache expiry or Gateway restart, and richer MCP tool blocks get normalized before they poison history.
- Replying to @openclawQQBot now strips model reasoning/thinking scaffolding before native delivery, so users see the answer instead of raw internal narration.
- Replying to @openclawMatrix got less awkward: voice notes can be transcribed before mention gating, and native threads preserve reply/read context instead of flattening the conversation.
- Replying to @openclawPlugin and ClawHub installs are sturdier: pinned GitHub skill commits, install-policy checks, and SQLite-backed install records for better update/repair behavior.
