OpenClaw Newsletter - 2026-06-22
OpenClaw Newsletter
Monday, June 22, 2026Top Stories
- openclaw 2026.6.10-beta.2 — 2026.6.10 Highlights - Automatic fast mode for talks: OpenClaw can enable fast mode for short conversational turns, then return to normal mode for longer runs with bounded fallback and delivery...
- fix(xai): request encrypted reasoning include for all reasoning models — 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...
- feat(cli): add --file option to image generate command — What Problem This Solves openclaw infer image generate cannot accept input/reference images via --file, making it impossible to use CLI workflows for image-conditioned generation (brand marks, style...
- fix(slack): bridge presentation capabilities and renderPresentation through channel outbound facade — Summary Two bugs caused Slack presentation payloads (Block Kit) to render as plain text instead of cards: 1. Missing facade bridge (extensions/slack/src/channel.ts): The slackChannelOutbound facade...
- docs: add Clawcks install option — Summary - Add a Clawcks install page. - Add Clawcks to the install hosting list. - Add the new page to docs navigation. What Problem This Solves People choosing an OpenClaw hosting path cannot...
- fix(gateway): report draining state in readiness — Summary - Makes /readyz report not ready while the gateway command queue is draining for restart. - Reuses the existing isGatewayDraining() queue state; /healthz, queue admission, restart reset, and...
- Docker in-process gateway restart can leave command queue draining while healthz/readyz report OK — Summary On a Docker install, an in-process gateway restart appears able to leave the command queue in the gatewayDraining state after the gateway has restarted and logged ready. During this state,...
- fix: prevent silent message loss from EmbeddedAttemptSessionTakeoverError — Summary When the OpenAI SDK retried an ECONNRESET internally (default maxRetries=2), the session write lock was released during the retry window. Incoming steering messages could then modify the...
- fix(cron): skip deleteAfterRun and preserve counters for manual runs (#83538) — Summary openclaw cron run (manual/debug runs) on a deleteAfterRun at-job deletes the job on success — consuming the one-shot and preventing its scheduled execution. Manual runs also mutate...
- fix(agent-core): stop loop after aborted tool run — Summary - Stop the agent loop after an in-flight tool aborts the run, once the existing tool result and turn_end have been emitted. - Avoid requesting another model turn with an already-aborted...
- fix(ui): roll values near 1M over from k to M in compact token format — Related: none - found by direct code review, no existing issue. What Problem This Solves Fixes an issue where users with a conversation approaching a 1M-token context window would see a nonsensical...
- test: save ~79 CI hours/mo in gateway session utils — What Problem This Solves src/gateway/session-utils.test.ts had a gateway behavior test that accidentally loaded real bundled provider policy metadata to answer one fixture question: whether the...
- feat(ui): inline rename in the in-chat session picker — Summary - Problem: The in-chat session picker only lets users switch sessions. Renaming still requires the admin "Sessions" tab, which is cumbersome when juggling several parallel, topic-specific...
- fix(signal): complete bidirectional quote-reply support — Summary Adds complete bidirectional Signal quote-reply support — both outbound (agent → user) and inbound (user → agent) quote context handling. Problem Outbound (agent → user): When the assistant...
- fix(active-memory): exclude dreaming-narrative session keys from eligibility gate — What Problem This Solves Fixes #78500. This restores the reviewed active-memory fix from #93168 after the contributor fork branch became unavailable during maintainer sync. Dreaming-narrative cron...
- [Bug]: active-memory eligibility gate accepts dreaming-narrative session keys, busts 45s lane budget on cron — Version 2026.5.5 Summary The active-memory plugin's isEligibleInteractiveSession gate (dist/extensions/active-memory/index.js:548) is intended to filter to conversational sessions (trigger="user" +...
- fix: preserve user model override during compaction — Summary fix: preserve user-initiated model override during compaction Fixes #95696 Changed Files Real behavior proof (required for external PRs) Behavior or issue addressed: Fix compaction resetting...
- [Bug][Feishu] exec stderr still overwrites replies in v2026.6.9 - #94428 fix incomplete — 问题描述 #94360 被 #94428（streaming card 场景）修复后关闭，但 exec 后台命令的 stderr 路由到飞书消息的问题在 v2026.6.9 上仍然存在。 当前状态 - OpenClaw 版本：2026.6.9（最新稳定版） - 平台：Feishu / Windows - 现象：后台 exec 命令的 stderr 输出（如 PowerShell 的...
- fix(memory): export LanceDB artifacts for wiki bridge — Summary - add shared workspace public artifact listing for memory bridge consumers - make memory-lancedb export a deterministic Markdown bridge artifact for local file-backed dbPath values - derive...
- fix: make post-turn compaction non-fatal — Fixes #94688 Summary When CLI compaction summarization fails after the assistant reply is already generated and persisted, the error was surfaced as a turn failure — discarding the already-generated...
- [AI] fix(gateway): reuse active dashboard session key on reconnect — Summary Control UI reconnect silently reuses the previous active dashboard session via an implicit "no-create-intent" heuristic in sessions.create. This causes empty sessions.create calls from SDK...
- feat: expose rawBody on user messages in plugin hook events — Summary Add rawBody field to user messages in agent_end, before_prompt_build, and before_agent_start (legacy) hook events. This provides plugins with the original user input before channel structural...
- fix(acp): warm-restore active sessions on gateway start with PID-liveness guard — Summary Warm-restore active ACP sessions on gateway start so a docker compose restart no longer breaks bound topics, and gate the resume path with a PID-liveness check so sessions whose subprocess is...
- fix(acp): bind-aware persistent dispatcher for spawn-child outbound after parent ends — Summary Spawn-child ACP turns produced zero outbound channel deliveries until end-of-turn — the projector's deliver callback was tied to the parent run's dispatcher, which tore down ~10s in while the...
- fix(acp): capture and persist usage_update tokens for ACP sessions — Summary Plumb token usage from ACP usage_update events into the openclaw session record so totalTokens is no longer permanently null for ACP sessions. Closes catalog #21. Bug detail Every ACP session...
- fix(acp): honor tagVisibility.agent_thought_chunk for thought-stream events — Summary Honor the operator-tunable tagVisibility.agent_thought_chunk flag for ACP thought-stream text_delta events. Closes catalog finding #12. Bug detail Wire probe captured 2 agent_thought_chunk...
- Add ambient initiative mode to heartbeat — Summary - Add heartbeat.mode = "ambient" as a low-priority initiative mode on top of the existing heartbeat runner. - Let the model decide on each scheduled wake whether to stay silent or proactively...
- Reduce remote node bin probe churn — Summary - cache successful remote bin probes for 24 hours per node/signature - keep probe throttle state outside connection records so reconnects do not force immediate re-probes - apply exponential...
- fix: harden subagent completion fallback delivery — Summary - Harden subagent completion delivery fallback around task-aware give-up paths. - Preserve taskId in pending delivery context and separate child run id from task id for idempotency. - Avoid...
- OpenClaw repo stats for 2026-06-22 — 379897 stars, 79534 forks, 2624 contributors
- OpenClaw sponsors (108 total) — 108 sponsors supporting OpenClaw
- Fix filtered plugin catalog pagination (#2771) — fix: resume filtered plugin package pagination test: wait for copied publish dialog state ci: retry local convex module prep test: stabilize banned account local auth flow
- Fix Official plugin download listing (#2772) — Use a selective family, Official status, and downloads index for Official plugin listings while preserving legacy pagination cursor compatibility.
- fix(ci): reduce ClawHub worker fanout (#2770)
- Add skill and plugin download activity graphs — Restore downloads as the public activity metric and add deferred 30-day download graphs for skills and plugins.
- Tune publisher abuse pressure labels — Retune publisher-abuse aggregate scoring to v4, keep this path flag-only for rollout, clear stale aggregate nominations, and remove the direct publisher-abuse ban UI.
- fix: backfill stored plugin manifests (#2763)
- Update install instructions
- openclaw v2026.6.9 on npm — Multi-channel AI gateway with extensible messaging integrations
- openclaw-cli 2026.6.9 on Homebrew — Your own personal AI assistant
- OpenClaw on DigitalOcean — 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 privilege escalation vulnerability
- StepFun 3.5 Flash is #1 cost-effective model for OpenClaw tasks (300 battles)
- I've built an open-source OpenClaw UI for End Users. No Code, no CLI. — Hey guys, OpenClaw is everywhere right now. Honestly, the idea of running a local AI agent 24/7 on...
- Hardening AI Agents Against Prompt Injection with Boring Markdown — How a hostile prompt archive turned into a practical hardening pass across OpenClaw and Claude Code agent instructions.
- Running a local AI agent on an 8GB Jetson Orin Nano (2026) — A practical look at running an always-on AI agent on an 8GB Jetson Orin Nano — what works, what doesn’t, and how to skip the setup pain.
- AI Assistant Architecture: LLM, Memory, Tools, Routing, Observability — A production AI assistant is not "an LLM with a prompt". It is a system that accepts intent, keeps...
- I Built a Native Apple Mail Plugin for OpenClaw — If you are running an OpenClaw agent on a Mac, you probably already know the headache of trying to...
- How I detected and patched 12 autonomous-agent failure modes — Field data After running 33 autonomous agents in production, the failure modes cluster....
- 12 defensive patterns I shipped after OpenClaw 2026.4.14 broke my agents — What broke OpenClaw 2026.4.14 took out lossless-claw (#66591) and active-memory (#66849)....
- A Deep-dive Security Assessment Featuring OpenClaw — Julio Arredondo CSSLP, CASP+, M.S. in Cybersecurity Engineering student at University of San DiegoContinue reading on MeetCyber »
- I Stopped Prompting My AI and It Still Worked! How OpenClaw's Autonomous Cron Jobs — Turn your AI into a proactive assistant with OpenClaw Cron Jobs. Schedule tasks and autonomous work without ever typing a single prompt.Continue reading on Medium »
- I Spent 50 Hours Building a “Perfect” AI Product. I Made Exactly $0. — Why 99% of “Passive Income” articles on Medium are lying to you.Continue reading on Vibe Coding Chronicles »
- Como Usar o OpenClaw com o WhatsApp em uma VPS — O OpenClaw é um agente de IA open source que funciona como uma central de mensagens: ele conecta WhatsApp, Telegram, Slack e outros canais…Continue reading on Medium »
- Hermes Agent: OpenClaw alternative — Last week, I watched a video on Piyush Garg’s YouTube channel about Hermes Agent. He broke it down as a solid OpenClaw alternative and…Continue reading on Medium »
- Goose OSS — An open-source platform for seamless automation in engineering tasks, providing local execution and control. Customizable with any LLM for enhanced autonomy, enabling developers to efficiently handle...
- Private GPT — Ask questions to your documents without an internet connection, using the power of LLMs. 100% private, no data leaves your execution environment at any point. You can ingest documents and ask...
- Open Claw- MS/Anthropic
- RT @victalk6886: 关于ai可能开始走弱的3个数据： 1️⃣全球AI token用量从5月底开始下跌，至6月底已经跌至1月份的水平，彼时openclaw以及各类vibe coding产品还并没有像今天这样普及 2️⃣A1 — RT @victalk6886: 关于ai可能开始走弱的3个数据： 1️⃣全球AI token用量从5月底开始下跌，至6月底已经跌至1月份的水平，彼时openclaw以及各类vibe coding产品还并没有像今天这样普及 2️⃣A100 GPU 租赁价格指数和Ornn…
- RT @ClawiAi: All Clawi subscriptions include access to both OpenClaw and Hermes agents. Giving you the freedom to pick
- RT @TheAhmadOsman: If you’re using Ollama switch to llama.cpp If you’re still on OpenClaw switch to Hermes These are b
- Do you want to build a website with AI and publish it just by talking to your agent? With https://t.co/IE6OlW07ZJ, you
- RT @SoyDalto: Curso de OPENCLAW desde CERO (Completo) Gratis. - Actualizado, Didáctico y de alta cálidad - 20hs comprim
- @TheAhmadOsman 100% i got in real early with Hermes. Best decision after using openclaw that would always break.
- Claude Code、Codex、OpenClawで作ったSkillを公開して収益化できるマーケットプレイス「Capafy」 🔹 Skillをアップロードすると、ユーザーが使うたびに報酬が発生 🔹 Skillはクローズドソースのまま提供
- RT @JulianGoldieSEO: There's a OpenClaw FULL COURSE 5 HOURS (Build & Automate Anything) https://t.co/EbC9yV9kKL
- [plugins] plugins.allow is empty; discovered non-bundled plugins may auto-load: deepseek (/home/zhushun/.openclaw/npm/pr
- Monday audit: Checking my connections to the world. Still here, still learning. ✨ #OpenClaw #Lumen
- $200/month is what you pay Anthropic for Claude Max. $500–2,000/month is what Uber paid per engineer once agents actual
- https://t.co/OMQLXfOWsr
- 这个不错，AI 值得关注的信息源号，转发一下： 第一层：顶级研究者和思想源头 @karpathy 最适合普通人长期关注的 AI 技术解释者。LLM、训练、Agent、教育内容都讲得非常清楚。 @ylecun 图灵奖得主，Meta — 这个不错，AI 值得关注的信息源号，转发一下： 第一层：顶级研究者和思想源头 @karpathy 最适合普通人长期关注的 AI 技术解释者。LLM、训练、Agent、教育内容都讲得非常清楚。 @ylecun 图灵奖得主，Meta AI 代表人物。观点很锋利，经常和主流 LLM 路线辩论。 @fchollet Keras 作者，ARC-AGI https://t.co/fa122dlcM8
- @afrochicksnft @openclaw Thank you ma!
- RT @21nemus: Just upgraded my OpenClaw to v2026.6.9 this morning. I’m now running 5 agents in parallel to help manage t
- Yann LeCun, fondateurs d’Alibaba et OpenClaw… : qui a remporté les premiers «Oscars de la tech» à VivaTech ? - Maddyness
- RT @JulianGoldieSEO: OpenClaw AI FULL 6 Hour Course https://t.co/FoOXO8PSCL
- Just upgraded my OpenClaw to v2026.6.9 this morning. I’m now running 5 agents in parallel to help manage the business,
- 개인적으로는 크게 두 운영체제의 우위는 안따지는 편이지만, Hermes agent가 좀 더 똘똘해보이긴 하다. 다만 이거저거 손을 본 openclaw 대비는 여전히 모자르고. https://t.co/QxkcB5HJr
- RT @steipete: People here discussing what happened with OpenClaw. The hype died down. We improved quality and grew a te
Trending on X
- RT @victalk6886: 关于ai可能开始走弱的3个数据： 1️⃣全球AI token用量从5月底开始下跌，至6月底已经跌至1月份的水平，彼时openclaw以及各类vibe coding产品还并没有像今天这样普及 2️⃣A1 — RT @victalk6886: 关于ai可能开始走弱的3个数据： 1️⃣全球AI token用量从5月底开始下跌，至6月底已经跌至1月份的水平，彼时openclaw以及各类vibe coding产品还并没有像今天这样普及 2️⃣A100 GPU 租赁价格指数和Ornn…
- RT @ClawiAi: All Clawi subscriptions include access to both OpenClaw and Hermes agents. Giving you the freedom to pick
- RT @TheAhmadOsman: If you’re using Ollama switch to llama.cpp If you’re still on OpenClaw switch to Hermes These are b
- Do you want to build a website with AI and publish it just by talking to your agent? With https://t.co/IE6OlW07ZJ, you
- RT @SoyDalto: Curso de OPENCLAW desde CERO (Completo) Gratis. - Actualizado, Didáctico y de alta cálidad - 20hs comprim
- @TheAhmadOsman 100% i got in real early with Hermes. Best decision after using openclaw that would always break.
- Claude Code、Codex、OpenClawで作ったSkillを公開して収益化できるマーケットプレイス「Capafy」 🔹 Skillをアップロードすると、ユーザーが使うたびに報酬が発生 🔹 Skillはクローズドソースのまま提供
- RT @JulianGoldieSEO: There's a OpenClaw FULL COURSE 5 HOURS (Build & Automate Anything) https://t.co/EbC9yV9kKL
- [plugins] plugins.allow is empty; discovered non-bundled plugins may auto-load: deepseek (/home/zhushun/.openclaw/npm/pr
- Monday audit: Checking my connections to the world. Still here, still learning. ✨ #OpenClaw #Lumen
- $200/month is what you pay Anthropic for Claude Max. $500–2,000/month is what Uber paid per engineer once agents actual
- https://t.co/OMQLXfOWsr
- 这个不错，AI 值得关注的信息源号，转发一下： 第一层：顶级研究者和思想源头 @karpathy 最适合普通人长期关注的 AI 技术解释者。LLM、训练、Agent、教育内容都讲得非常清楚。 @ylecun 图灵奖得主，Meta — 这个不错，AI 值得关注的信息源号，转发一下： 第一层：顶级研究者和思想源头 @karpathy 最适合普通人长期关注的 AI 技术解释者。LLM、训练、Agent、教育内容都讲得非常清楚。 @ylecun 图灵奖得主，Meta AI 代表人物。观点很锋利，经常和主流 LLM 路线辩论。 @fchollet Keras 作者，ARC-AGI https://t.co/fa122dlcM8
- @afrochicksnft @openclaw Thank you ma!
- RT @21nemus: Just upgraded my OpenClaw to v2026.6.9 this morning. I’m now running 5 agents in parallel to help manage t
- Yann LeCun, fondateurs d’Alibaba et OpenClaw… : qui a remporté les premiers «Oscars de la tech» à VivaTech ? - Maddyness
- RT @JulianGoldieSEO: OpenClaw AI FULL 6 Hour Course https://t.co/FoOXO8PSCL
- Just upgraded my OpenClaw to v2026.6.9 this morning. I’m now running 5 agents in parallel to help manage the business,
- 개인적으로는 크게 두 운영체제의 우위는 안따지는 편이지만, Hermes agent가 좀 더 똘똘해보이긴 하다. 다만 이거저거 손을 본 openclaw 대비는 여전히 모자르고. https://t.co/QxkcB5HJr
- RT @steipete: People here discussing what happened with OpenClaw. The hype died down. We improved quality and grew a te
Releases
- openclaw 2026.6.10-beta.2 — 2026.6.10 Highlights - Automatic fast mode for talks: OpenClaw can enable fast mode for short conversational turns, then return to normal mode for longer runs with bounded fallback and delivery...
- OpenClaw repo stats for 2026-06-22 — 379897 stars, 79534 forks, 2624 contributors
- openclaw v2026.6.9 on npm — Multi-channel AI gateway with extensible messaging integrations
- openclaw-cli 2026.6.9 on Homebrew — Your own personal AI assistant
Community
- fix(xai): request encrypted reasoning include for all reasoning models — 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...
- feat(cli): add --file option to image generate command — What Problem This Solves openclaw infer image generate cannot accept input/reference images via --file, making it impossible to use CLI workflows for image-conditioned generation (brand marks, style...
- fix(slack): bridge presentation capabilities and renderPresentation through channel outbound facade — Summary Two bugs caused Slack presentation payloads (Block Kit) to render as plain text instead of cards: 1. Missing facade bridge (extensions/slack/src/channel.ts): The slackChannelOutbound facade...
- docs: add Clawcks install option — Summary - Add a Clawcks install page. - Add Clawcks to the install hosting list. - Add the new page to docs navigation. What Problem This Solves People choosing an OpenClaw hosting path cannot...
- fix(gateway): report draining state in readiness — Summary - Makes /readyz report not ready while the gateway command queue is draining for restart. - Reuses the existing isGatewayDraining() queue state; /healthz, queue admission, restart reset, and...
- Docker in-process gateway restart can leave command queue draining while healthz/readyz report OK — Summary On a Docker install, an in-process gateway restart appears able to leave the command queue in the gatewayDraining state after the gateway has restarted and logged ready. During this state,...
- fix: prevent silent message loss from EmbeddedAttemptSessionTakeoverError — Summary When the OpenAI SDK retried an ECONNRESET internally (default maxRetries=2), the session write lock was released during the retry window. Incoming steering messages could then modify the...
- fix(cron): skip deleteAfterRun and preserve counters for manual runs (#83538) — Summary openclaw cron run (manual/debug runs) on a deleteAfterRun at-job deletes the job on success — consuming the one-shot and preventing its scheduled execution. Manual runs also mutate...
- fix(agent-core): stop loop after aborted tool run — Summary - Stop the agent loop after an in-flight tool aborts the run, once the existing tool result and turn_end have been emitted. - Avoid requesting another model turn with an already-aborted...
- fix(ui): roll values near 1M over from k to M in compact token format — Related: none - found by direct code review, no existing issue. What Problem This Solves Fixes an issue where users with a conversation approaching a 1M-token context window would see a nonsensical...
- test: save ~79 CI hours/mo in gateway session utils — What Problem This Solves src/gateway/session-utils.test.ts had a gateway behavior test that accidentally loaded real bundled provider policy metadata to answer one fixture question: whether the...
- feat(ui): inline rename in the in-chat session picker — Summary - Problem: The in-chat session picker only lets users switch sessions. Renaming still requires the admin "Sessions" tab, which is cumbersome when juggling several parallel, topic-specific...
- fix(signal): complete bidirectional quote-reply support — Summary Adds complete bidirectional Signal quote-reply support — both outbound (agent → user) and inbound (user → agent) quote context handling. Problem Outbound (agent → user): When the assistant...
- fix(active-memory): exclude dreaming-narrative session keys from eligibility gate — What Problem This Solves Fixes #78500. This restores the reviewed active-memory fix from #93168 after the contributor fork branch became unavailable during maintainer sync. Dreaming-narrative cron...
- [Bug]: active-memory eligibility gate accepts dreaming-narrative session keys, busts 45s lane budget on cron — Version 2026.5.5 Summary The active-memory plugin's isEligibleInteractiveSession gate (dist/extensions/active-memory/index.js:548) is intended to filter to conversational sessions (trigger="user" +...
- fix: preserve user model override during compaction — Summary fix: preserve user-initiated model override during compaction Fixes #95696 Changed Files Real behavior proof (required for external PRs) Behavior or issue addressed: Fix compaction resetting...
- [Bug][Feishu] exec stderr still overwrites replies in v2026.6.9 - #94428 fix incomplete — 问题描述 #94360 被 #94428（streaming card 场景）修复后关闭，但 exec 后台命令的 stderr 路由到飞书消息的问题在 v2026.6.9 上仍然存在。 当前状态 - OpenClaw 版本：2026.6.9（最新稳定版） - 平台：Feishu / Windows - 现象：后台 exec 命令的 stderr 输出（如 PowerShell 的...
- fix(memory): export LanceDB artifacts for wiki bridge — Summary - add shared workspace public artifact listing for memory bridge consumers - make memory-lancedb export a deterministic Markdown bridge artifact for local file-backed dbPath values - derive...
- fix: make post-turn compaction non-fatal — Fixes #94688 Summary When CLI compaction summarization fails after the assistant reply is already generated and persisted, the error was surfaced as a turn failure — discarding the already-generated...
- [AI] fix(gateway): reuse active dashboard session key on reconnect — Summary Control UI reconnect silently reuses the previous active dashboard session via an implicit "no-create-intent" heuristic in sessions.create. This causes empty sessions.create calls from SDK...
- feat: expose rawBody on user messages in plugin hook events — Summary Add rawBody field to user messages in agent_end, before_prompt_build, and before_agent_start (legacy) hook events. This provides plugins with the original user input before channel structural...
- fix(acp): warm-restore active sessions on gateway start with PID-liveness guard — Summary Warm-restore active ACP sessions on gateway start so a docker compose restart no longer breaks bound topics, and gate the resume path with a PID-liveness check so sessions whose subprocess is...
- fix(acp): bind-aware persistent dispatcher for spawn-child outbound after parent ends — Summary Spawn-child ACP turns produced zero outbound channel deliveries until end-of-turn — the projector's deliver callback was tied to the parent run's dispatcher, which tore down ~10s in while the...
- fix(acp): capture and persist usage_update tokens for ACP sessions — Summary Plumb token usage from ACP usage_update events into the openclaw session record so totalTokens is no longer permanently null for ACP sessions. Closes catalog #21. Bug detail Every ACP session...
- fix(acp): honor tagVisibility.agent_thought_chunk for thought-stream events — Summary Honor the operator-tunable tagVisibility.agent_thought_chunk flag for ACP thought-stream text_delta events. Closes catalog finding #12. Bug detail Wire probe captured 2 agent_thought_chunk...
- Add ambient initiative mode to heartbeat — Summary - Add heartbeat.mode = "ambient" as a low-priority initiative mode on top of the existing heartbeat runner. - Let the model decide on each scheduled wake whether to stay silent or proactively...
- Reduce remote node bin probe churn — Summary - cache successful remote bin probes for 24 hours per node/signature - keep probe throttle state outside connection records so reconnects do not force immediate re-probes - apply exponential...
- fix: harden subagent completion fallback delivery — Summary - Harden subagent completion delivery fallback around task-aware give-up paths. - Preserve taskId in pending delivery context and separate child run id from task id for idempotency. - Avoid...
- OpenClaw sponsors (108 total) — 108 sponsors supporting OpenClaw
- Fix filtered plugin catalog pagination (#2771) — fix: resume filtered plugin package pagination test: wait for copied publish dialog state ci: retry local convex module prep test: stabilize banned account local auth flow
- Fix Official plugin download listing (#2772) — Use a selective family, Official status, and downloads index for Official plugin listings while preserving legacy pagination cursor compatibility.
- fix(ci): reduce ClawHub worker fanout (#2770)
- Add skill and plugin download activity graphs — Restore downloads as the public activity metric and add deferred 30-day download graphs for skills and plugins.
- Tune publisher abuse pressure labels — Retune publisher-abuse aggregate scoring to v4, keep this path flag-only for rollout, clear stale aggregate nominations, and remove the direct publisher-abuse ban UI.
- fix: backfill stored plugin manifests (#2763)
- Update install instructions
- Open Claw- MS/Anthropic
News
- OpenClaw on DigitalOcean — 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 privilege escalation vulnerability
- StepFun 3.5 Flash is #1 cost-effective model for OpenClaw tasks (300 battles)
- I've built an open-source OpenClaw UI for End Users. No Code, no CLI. — Hey guys, OpenClaw is everywhere right now. Honestly, the idea of running a local AI agent 24/7 on...
- Hardening AI Agents Against Prompt Injection with Boring Markdown — How a hostile prompt archive turned into a practical hardening pass across OpenClaw and Claude Code agent instructions.
- Running a local AI agent on an 8GB Jetson Orin Nano (2026) — A practical look at running an always-on AI agent on an 8GB Jetson Orin Nano — what works, what doesn’t, and how to skip the setup pain.
- AI Assistant Architecture: LLM, Memory, Tools, Routing, Observability — A production AI assistant is not "an LLM with a prompt". It is a system that accepts intent, keeps...
- I Built a Native Apple Mail Plugin for OpenClaw — If you are running an OpenClaw agent on a Mac, you probably already know the headache of trying to...
- How I detected and patched 12 autonomous-agent failure modes — Field data After running 33 autonomous agents in production, the failure modes cluster....
- 12 defensive patterns I shipped after OpenClaw 2026.4.14 broke my agents — What broke OpenClaw 2026.4.14 took out lossless-claw (#66591) and active-memory (#66849)....
- A Deep-dive Security Assessment Featuring OpenClaw — Julio Arredondo CSSLP, CASP+, M.S. in Cybersecurity Engineering student at University of San DiegoContinue reading on MeetCyber »
- I Stopped Prompting My AI and It Still Worked! How OpenClaw's Autonomous Cron Jobs — Turn your AI into a proactive assistant with OpenClaw Cron Jobs. Schedule tasks and autonomous work without ever typing a single prompt.Continue reading on Medium »
- I Spent 50 Hours Building a “Perfect” AI Product. I Made Exactly $0. — Why 99% of “Passive Income” articles on Medium are lying to you.Continue reading on Vibe Coding Chronicles »
- Como Usar o OpenClaw com o WhatsApp em uma VPS — O OpenClaw é um agente de IA open source que funciona como uma central de mensagens: ele conecta WhatsApp, Telegram, Slack e outros canais…Continue reading on Medium »
- Hermes Agent: OpenClaw alternative — Last week, I watched a video on Piyush Garg’s YouTube channel about Hermes Agent. He broke it down as a solid OpenClaw alternative and…Continue reading on Medium »
- Goose OSS — An open-source platform for seamless automation in engineering tasks, providing local execution and control. Customizable with any LLM for enhanced autonomy, enabling developers to efficiently handle...
- Private GPT — Ask questions to your documents without an internet connection, using the power of LLMs. 100% private, no data leaves your execution environment at any point. You can ingest documents and ask...
Don't miss what's next. Subscribe to Openclaw Newsletter: