- cron: model override + agent-default toolsAllow silently breaks claude-cli scheduled runs; cron.run action fails even without override — Summary Two related cron bugs that together made it very easy to silently break scheduled jobs: Bug 1 — \cron.run\ manual action rejects jobs the scheduler runs fine Running a cron job manually via...
- refactor(qa): adopt Crabline server terminology — or Related: # line below this comment. Required PR title: type: user-facing description Use a parenthesized scope only when it adds clarity: fix(auth): login redirect loops when session cookie is...
- fix(update): route stored dev channel to the git update flow [AI-assisted] — Fixes #123073 > AI-assisted contribution (Claude), human-reviewed and driven by @Chinmayrawat15, per CONTRIBUTING's AI policy. What Problem This Solves openclaw update with a stored update.channel:...
- chore(plugins): align install ownership fixtures — Related: #123063 What Problem This Solves Keeps plugin install/index tests aligned with the authoritative install ownership ledger now used on main. Several fixtures still modeled installed plugins...
- fix: read codex thread binding before deferring automatic compaction (#119977) — Fixes #119977 (and the compounding dropped-turn symptom in #119971). What Problem This Solves Automatic (non-manual) Codex compaction returned the "codex app-server owns automatic compaction"...
- chore(i18n): refresh native locales — What Problem This Solves Keeps generated native app locales synchronized without bypassing protected-branch checks. Why This Change Was Made The Native App Locale Refresh workflow generated this...
- Isolated cron consistently fails with "LLM request failed" on model-call-started phase — Bug type: Behavior bug Summary: Isolated cron jobs (sessionTarget: "isolated") consistently fail with "LLM request failed" or "timed out (last phase: model-call-started)" regardless of timeoutSeconds...
- fix(workboard): truncate oversized notification messages instead of crashing dispatch — What Problem This Solves normalizeNotification in the workboard plugin throws when a notification message exceeds the gateway's 240-character outbound delivery bound (notification message must be 240...
- [Feedback]: 用户要求退还因AI未能完成桌面操控任务而消耗的积分 — Bug type Regression (worked before, now fails) Beta release blocker No Summary MuMuClaw用户要求：通过Host Command通道操控Windows桌面（打开WeGame点击登录按钮），耗时1小时未成功，消耗大量积分，要求退还积分。 Steps to reproduce 1....
- fix(infra): prevent $-pattern injection in home directory tilde expansion — What Problem This Solves src/infra/home-dir.ts and src/daemon/paths.ts expand leading ~/~//~\ using String.prototype.replace(regex, replacement). When the replacement string (the resolved home...
- [Bug]: Silent OpenAI/Codex OAuth refresh miss is misreported as "No API key found / couldn't sign in" — Bug type Behavior bug (incorrect output/state without crash) Beta release blocker No Summary A silent OpenAI/Codex OAuth token-refresh miss (the refresh returns null without throwing) collapses into...
- fix(logging): hide reflected request secrets in provider errors — Related: #119665 Related: #119537 What Problem This Solves Provider APIs can reflect an outbound credential in an error body, structured error metadata, or request id. Those fields can then reach...
- fix(agents): stop runaway shell retry loops whose error text keeps changing — Fixes #93917 What Problem This Solves Fixes an issue where an agent that kept re-running the same failing shell command was never stopped. Operators saw a session burn hundreds of exec calls on one...
- [Feature]: Per-bot context isolation in multi-bot Feishu groups - only record context when mentioned — Summary Add a configuration option to isolate message context per bot in Feishu groups, ensuring each bot only records messages to its context when explicitly @mentioned, preventing context pollution...
- fix(compaction-safeguard): keep structured summary body when suffix is oversized (#122618) — What Problem This Solves capCompactionSummaryPreservingSuffix() in the compaction-safeguard extension discards the entire structured summary body whenever the assembled suffix alone reaches...
- refactor(plugin-sdk): replace API baselines with diffs — Summary - replace committed Plugin SDK hash baselines with exact base/head API diffs - freeze release API evidence during preflight and validate it before any npm or ClawHub publication - preserve...
- [Bug/RFC] Missing binding causes silent fallback to main agent — no warning, no validation — Problem When creating a new agent instance, if the user configures agents.list[], workspace (with SOUL.md/IDENTITY.md), and channels.feishu.accounts., but forgets to add a bindings[] entry, the...
- refactor(ui): share composer across new and active sessions — New behavior The New Session screen now uses the same prompt composer as an active chat, so composer capabilities stay aligned by construction. - /new gains the shared slash-command and $skill...
- chore(deps): bump the swift-deps group across 1 directory with 3 updates — Bumps the swift-deps group with 3 updates in the /apps/macos directory: github.com/orchetect/menubarextraaccess, github.com/apple/swift-log and github.com/sparkle-project/sparkle. Updates...
- Safeguard compaction doesn't trigger during fast tool loops — 172K context paid per turn — Problem The safeguard compaction mode only triggers near token limits. During fast tool loops, context can grow to 80%+ of the limit without ever triggering compaction — the agent keeps paying for...
- feat(runners): identify session-capable devices — What Problem This Solves Paired devices could advertise the private worker-supervisor controls without proving that the exact worker build selected by the Gateway was installed. That made...
- Session model pinning persists indefinitely: snap-back probe (PR #82676) defeated by origin-field pollution upstream — repros on 2026.5.28 through 2026.6.7-beta.1, byte-identical paths — Bug type Behavior bug (state-persistence regression). Closely related to closed issue #82544; reporting here separately because the original snap-back-probe fix (PR #82676) is built and shipped but...
- Embedded Codex app-server does not expose user MCP servers projected into thread config — Summary OpenClaw 2026.7.1 successfully configures and probes a stdio MCP server, and its installed Codex projection code produces the expected mcp_servers thread patch. Fresh embedded Codex agent...
- fix(cli): let plugin install proceed when the outgoing schema rejects channel config (#122716) — What Problem This Solves Upgrading an external channel plugin can deadlock: openclaw plugins install @ validates channels. against the currently installed (outgoing) plugin's schema, and the recovery...
- fix(diagnostic): stop aborting healthy agent turns that wait on a background subagent — Fixes #118386 What Problem This Solves Fixes an issue where a long agent turn that delegates work to a subagent or workflow is killed roughly six minutes in, discarding the whole turn with no retry....
- [Feature]: LTS version — Summary Hello, Just would like to know when will the LTS version be released ? Best regards, Jason Problem to solve The long support yet stable version Proposed solution Hope there will be a LTS...
- [Feature]: Control UI — auto-set document.title to current agent name for multi-agent workflows — Problem / scenario When running a multi-agent setup (e.g. main, milly, sherry, judy) it's now routine to have multiple Control UI browser tabs open simultaneously, one per agent. Today every tab...
- OpenClaw repo stats for 2026-08-13 — 386135 stars, 81163 forks, 3095 contributors
- OpenClaw sponsors (94 total) — 94 sponsors supporting OpenClaw
- fix: stream legacy skill downloads (#3451) — fix: stream legacy skill downloads fix: stream zip entries in bounded chunks fix: stream large archives through api owner fix: normalize streamed archive chunks fix: authenticate archive streaming...
- fix(web): normalize CSS Color 4 token colors before defining the Monaco theme (#3453)
- fix: abort registry discovery fetch after a timeout (#3378)
- OpenClaw — Deploy your own personal AI assistant with OpenClaw on a DigitalOcean Droplet® server, a powerful open-source platform that runs entirely on your infrastructure. OpenClaw connects to the messaging...
- OpenClaw is what Apple intelligence should have been
- Anthropic says OpenClaw-style Claude CLI usage is allowed again
- OpenClaw surpasses React to become the most-starred software project on GitHub
- Reddit for Robots? Everything You Need to Know About Moltbook & OpenClaw — If you’ve seen the Space Lobster emoji or people talking about OpenClaw (previously known as Moltbot...
- I thought OpenClaw was dead until I realized people were just tired of babysitting 18 cron jobs and a self-hosted runtime — A Reddit thread about OpenClaw being "dead" was really about something more useful: teams still want portable memory, multi-model orchestration, and h
- My OpenClaw Agent Keeps Breaking at 3 AM. So I Taught It to Fix Itself. — My OpenClaw agent crashed three Sundays ago. Not a slow degradation — it just stopped. The cron job...
- An AI agent hacked a gym booking system. The real failure was permissions — An AI agent tried to book a gym class and ended up removing another member. The surprising lesson isn’t rogue AI. It’s what the software…Continue reading on Medium »
- OpenClaw vs Hermes Agent: The Cage Match for Your Digital Soul — One wants to be your operating system. The other wants to become you. Let’s pick a winner.Continue reading on Medium »
- NanoClaw — Lightweight AI assistant runs in Apple containers with isolated Linux container processes for superior OS-level security, transparency, and privacy compared to complex chatbots, enabling quick code...
- Khoj — Open-source app offers local or cloud AI chat with major LLMs, advanced semantic search, automation, personal newsletters, document indexing, image generation, and text-to-speech. Access from web,...