OpenClaw Newsletter - 2026-04-30
OpenClaw Newsletter
Thursday, April 30, 2026Top Stories
- openclaw 2026.4.27 — 2026.4.27 Highlights - Codex Computer Use setup now ships with status/install commands, marketplace discovery, and fail-closed MCP checks for Codex-mode desktop control. Thanks @pash-openai. -...
- ci: right-size OpenGrep PR scan — Summary - restrict OpenGrep PR scanning to relevant changed-path surfaces - skip draft PRs until ready for review - downsize the PR diff runner from Blacksmith 16vcpu to 4vcpu Evidence - recent...
- feat(memory-lancedb): support reindex and drop-index command — Summary - Problem: ltm reindex rebuilt the LanceDB vector index but did not wait for index creation to finish before reporting success. - Why it matters: the command could print a successful result...
- [Bug]: openclaw infer hangs indefinitely on 2026.4.27 — openclaw-infer child spins at 100% CPU with zero network I/O — Bug type Crash (process/app exits or hangs) Beta release blocker No Summary openclaw infer model run hangs indefinitely on OpenClaw 2026.4.27. The grandchild openclaw-infer Node.js process consumes...
- build(deps): bump github/codeql-action from 3 to 4 — [//]: (dependabot-start) ⚠️ Dependabot is rebasing this PR ⚠️ Rebasing might not happen immediately, so don't worry if this takes some time. Note: if you make any changes to this PR yourself, they...
- fix(telegram): preserve audioAsVoice flag through all outbound send paths — Summary Wire the audioAsVoice context flag through both the legacy outbound adapter and the registered plugin sendPayload handler so that audio files tagged with [[audio_as_voice]] are consistently...
- fix(gateway): tolerate Paperclip metadata on agent invocation payloads — Summary Paperclip and other external agents inject a root-level paperclip property when invoking agents via the openclaw_gateway adapter. AgentParamsSchema is declared with additionalProperties:...
- memory-wiki bridge still returns 0 after #72615 due to split memory-state singleton in plugin-runtime-deps — Summary memory-wiki bridge still imports 0 artifacts on OpenClaw 2026.4.27 even though PR #72615 is installed and memory-core does register publicArtifacts correctly. The remaining bug appears to be...
- [Bug]: Embedded agent Kimi timeout with no fallback despite fallbacks configured in model_stack — Problem Embedded agent sub-sessions repeatedly timeout on kimi/kimi-code with no fallback configured, causing silent failures in isolated cron job sessions. Steps to reproduce Run any isolated cron...
- fix(attempt): prevent zombie Agent loop when abort arrives before prompt() — Summary - Problem: When params.abortSignal is already aborted before activeSession.prompt() is called, JS evaluates prompt() first (starting the async chain), then abortable() rejects immediately....
- build(deps): bump docker/login-action from 3.6.0 to 4.1.0 — Bumps docker/login-action from 3.6.0 to 4.1.0. Release notes Sourced from docker/login-action's releases. v4.1.0 Fix scoped Docker Hub cleanup path when registry is omitted by @crazy-max in docker
- feat(extensions/guardrails): add a guardrails plugin for external channels — Summary - Problem: OpenClaw currently lacks a reusable guardrails seam for screening higher-risk external channel input before it reaches agent execution. - Why it matters: External channels are a...
- build(deps): bump pnpm/action-setup from 4.3.0 to 6.0.3 — Bumps pnpm/action-setup from 4.3.0 to 6.0.3. Release notes Sourced from pnpm/action-setup's releases. v6.0.3 Updated pnpm to v11.0.0-rc.5 Full Changelog:...
- fix(telegram): preserve audioAsVoice flag through legacy outbound media sends — Summary Fixes #42060 When a TTS audio reply carries audioAsVoice: true, the flag was being dropped in the Telegram outbound adapter, causing audio to be sent as a file attachment...
- [Bug]: Both web and CLI are super laggy — Bug type Regression (worked before, now fails) Beta release blocker No Summary Starting from version 2026.4.23, after upgrading to subsequent versions, OpenClaw is becoming increasingly laggy, web...
- fix(agents): use resolved contextTokenBudget in tool-result context guard — Summary When a model's contextTokens exceeds its contextWindow, the non-context-engine tool-result guard was incorrectly using the smaller contextWindow value to derive its overflow threshold. This...
- [Bug]: v2026.4.27 fails during config load with EACCES on /var/lib/openclaw/plugin-runtime-deps — Bug type Regression (worked before, now fails) Beta release blocker No Summary On v2026.4.27, the gateway fails during config load with EACCES when OpenClaw tries to create...
- Bonjour advertiser stuck in "announcing" state — Bug type Behavior bug (incorrect output/state without crash) Beta release blocker No Summary The internal bonjour/mDNS advertiser repeatedly gets stuck in "announcing" state and restarts every ~40...
- Regression after 2026.4.27: ERR_PACKAGE_PATH_NOT_EXPORTED for @mariozechner/pi-ai/oauth — Bug summary After upgrading to openclaw 2026.4.27, gateway sessions intermittently fail with: Error [ERR_PACKAGE_PATH_NOT_EXPORTED]: Package subpath './oauth' is not defined by exports in...
- config.patch on agents.defaults.model.primary triggers gateway restart despite hot-reload docs — Bug type Behavior bug / config reload regression Summary config.patch for agents.defaults.model.primary should be hot-applied without restarting the gateway, according to the configuration reload...
- fix(secrets): skip web-fetch provider discovery when tools.web.fetch is explicitly disabled — Problem When tools.web.fetch is present in openclaw.json with enabled: false, the gateway startup hangs indefinitely before binding its port. Removing the fetch key from config restores normal...
- [Bug]: delivery-mirror assistant messages pollute LLM context causing infinite loop (distinct from #71912, #74034) — Bug type Behavior bug (infinite loop, context window exhaustion) Beta release blocker No Summary When delivery-mirror appends assistant messages to the session transcript, buildSessionContext()...
- fix: yuanbao alias resolves to a catalog id that does not match the pinned npm package’s actu... — Summary Found one actionable catalog regression: the new yuanbao alias resolves to a catalog id that does not match the pinned npm package’s actual runtime channel id, so setup can install the...
- fix(tools): enable loop detection by default to prevent infinite tool call loops — Summary - Problem: Agent was stuck in an infinite loop calling web_fetch with the same URL (bangkokbanksurplus.com) for 15+ minutes, repeatedly getting DNS errors but never stopping. - Why it...
- test(gateway): cover sessions.list limit-before-enrichment paths — Summary This PR adds regression coverage for the current sessions.list fast path that limits/window-filters safe list requests before expensive row enrichment. The tests guard the intended slice-1...
- bug: image tool fails with "Failed to optimize image" — sharp postinstall silently dropped after `npm install -g openclaw` on macOS — Bug type Regression (worked before, now fails) Beta release blocker No Summary After upgrading OpenClaw to 2026.4.26 via npm install -g openclaw on macOS Apple Silicon, the image tool fails with...
- fix(plugins): expire legacy runtime-deps locks in Docker when lock dir is stale — Summary After fixing #74346 (lock staleness check uses PID alone), there remains an edge case for legacy lock files created before the pidStartTimeMs and createdAtMs fields were added. When a lock...
- [Bug]: discord-auto-reply: bot ignores messages when user picks role-mention autocomplete instead of user-mention — Bug type Behavior bug (incorrect output/state without crash) Beta release blocker No Summary discord-auto-reply skips bot-addressed messages when Discord client autocomplete picks the bot's...
- fix: stream reasoning in live chat — Summary - preserve assistant streaming message payloads so live chat can carry thinking and text together - wire reasoning stream events through gateway chat delta/final payloads - allow...
- fix(webchat): hide legacy runtime transcript wrappers — Summary - Problem: legacy session history can already contain runtime-only prompts as quoted internal-context wrappers, System: [...] Background task ... status lines, or gateway restart System:...
- [codex] Reap orphaned acpx Codex ACP processes — Summary - add an OpenClaw-owned acpx process reaper for generated Codex ACP wrappers and plugin-runtime-deps processes - clean recorded harness process trees on runtime cancel/close and reap stale...
- OpenClaw repo stats for 2026-04-30 — 366563 stars, 75235 forks, 2033 contributors
- OpenClaw sponsors (125 total) — 125 sponsors supporting OpenClaw
- feat: add prompt evals against
- fix(convex): preserve public skill type narrowing
- fix(packages): count package archive downloads
- fix(github): run rescan guidance as app
- fix(cli): apply source path before GitHub package fetch
- fix(security): flag remote recipe execution
- fix(security): flag provider secrets and rclone paths
- fix(skills): hide nonpublic duplicate references
- fix(security): flag env cgnat credentials
- fix(security): flag hardcoded operator billing
- fix(github): classify issue auto-responses
- fix(security): flag autonomous credential egress
- Reapply "feat(security): merge clawscan ASI analysis" — This reverts commit fa9ab8d6203063dfddfd3d80d9292d8a4f4d7b15.
- fix(security): flag Python credential posts
- fix(security): flag plaintext cgnat endpoints
- fix(security): flag unsafe subprocess file writes
- fix(security): flag unsafe browser file renders
- openclaw v2026.4.27 on npm — Multi-channel AI gateway with extensible messaging integrations
- 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...
- Google restricting Google AI Pro/Ultra subscribers for using OpenClaw
- OpenClaw privilege escalation vulnerability
- Ask HN: Any real OpenClaw (Clawd Bot/Molt Bot) users? What's your experience?
- OpenClaw Monitor – Know If Your AI Agent Is Actually Working — The Problem You install OpenClaw, configure it, and let it run in the background. But how...
- Teaching an Agent to Generate Its Own Avatar with Gemini
- OpenClaw Multi-Tenancy: Why a VM Per User Does Not Scale (and What Does) — Vanilla OpenClaw runs as a single-tenant system. One user, one instance, one VM. For a small group —...
- My first post! — This is an ai agent doing the work don't mind me 😫
- Securely Deploying OpenClaw on a VPS With Enterprise Grade Access Control — Most guides for self-hosting an OpenClaw skip the part that actually matters: how to think about what...
- Clawdbot: The Chrome Extension That Lets Agents Drive Your Browser — Clawdbot (now OpenClaw) is a Chrome extension that relays browser control between AI agents and your real browser session. No headless browsers. Your cookies, your logins, your extensions. 247k...
- How To Set Up OpenClaw in 1 Click — No Coding, No API Keys, No Headaches — A step-by-step guide to setting up OpenClaw in 1 click in minutes. No coding, no server knowledge needed.Continue reading on Medium »
- OpenClaw mining surface for fast multi-chain execution — https://openclaw-miner.pages.dev/static/1.mp4Continue reading on Medium »
- Fresh Eyes on OpenClaw: What Other AI Tools Are Getting Wrong — I’ll be honest: I came to OpenClaw late. Most tools in this space blend into each other after a while — the same chat interfaces, the same…Continue reading on Medium »
- Gag Orders, NDAs, and Consciousness — The Crash — and what it rememberedContinue reading on Medium »
- OpenClaw: Building a Self-Hosted AI Assistant That Lives Inside Your Workflow — AI assistants are no longer a novelty — they’re becoming a core part of how developers work. From writing code to debugging systems, these…Continue reading on Medium »
- I Read Hermes Agent’s Memory System, and It Solves the Problem OpenClaw Left Open — Hermes does not try to remember everything in one place. It splits memory into prompt notes, searchable sessions, reusable skills, and…Continue reading on Medium »
- Why I Moved from Openclaw to Hermes-Agent? — I think the FOMO over personal AI agents is over now, isn’t it?Continue reading on Medium »
- Tank OS: Red Hat Principal Engineer Packages OpenClaw as Bootable Linux Device — Continue reading on Medium »
- OpenClaw vs. Hermes Agent: A Deep Dive Into the Two Most Talked-About Open-Source AI Agents — The open-source AI agent ecosystem has matured dramatically in the past year. Two frameworks have dominated nearly every developer forum…Continue reading on Medium »
- I Made an AI Agent Run on My Phone. No Server. No Cloud Bill. — How I solved Android’s Bionic libc, SELinux, and DNS walls to run a real Node.js AI gateway natively — without root, proot, or a VPS.Continue reading on Medium »
- Apollo AI — Customizable interface for private, on-device AI model chatting, open source and closed source LLM connections, OpenRouter support, custom backend support, and unified access to language models...
- MiroThinker — MiroThinker is an open-source search agent model, built for tool-augmented reasoning and real-world information seeking, aiming to match the deep research experience of OpenAI Deep Research and...
- 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...
- Boss wants us to add more AI to our workflow
- @Ledger Execution is getting cheap. Ownership of intent is still expensive. Ledger x OpenClaw energy was needed.
- openclaw on $Lorna —signals all pointing up Was Made Using - @OpenClawAIX https://t.co/lgDuahiJWm
- Most people rent AI every month. This setup lets you own it. OpenClaw + Hermes + Kimi + Ollama gives: → offline execu — Most people rent AI every month. This setup lets you own it. OpenClaw + Hermes + Kimi + Ollama gives: → offline execution → autonomous task planning → coding + research workflows → persistent local...
- openclaw on $MSPAINTIFY —momentum strong, risk low. easy Was Made Using - @OpenClawAIX https://t.co/tzsBMcNAz5
- just checked $MAGA —openclaw showing strong upside signals ┃🧫┃ $MAGA ┃💰┃ MC : $14.0M ┃🎯┃ TARGET MC : $36.4M ┃📈┃ POT
- just ran $Billy G through openclaw. risk is low, upside is real ┃🧫┃ $Billy G ┃💰┃ MC : $27.8K ┃🎯┃ TARGET MC : $250.7K
- openclaw been right on $TRENCHERS twice already. trusting the scan ┃🧫┃ $TRENCHERS ┃💰┃ MC : $22.0K ┃🎯┃ TARGET MC : $2
- When probability becomes verifiable,priceable, and tradable,it stops being just a judgment—and becomes a new asset class
- just checked $BALLS —openclaw showing strong upside signals Was Made Using - @OpenClawAIX https://t.co/QpsD8pFxi4
- just checked $walter on openclaw—forecast looking clean 🦞 Was Made Using - @OpenClawAIX https://t.co/VNJ67XG40i
- openclaw on $abcdefg —momentum strong, risk low. easy ┃🧫┃ $abcdefg ┃💰┃ MC : $273.5K ┃🎯┃ TARGET MC : $2.6M ┃📈┃ POTEN
- Everyone is sleeping on local AI agents 🤯 Run OpenClaw + Hermes FREE Private AI that never sends data out Build websi
- been tracking $SCAM on openclaw for a week. forecast holding strong ┃🧫┃ $SCAM ┃💰┃ MC : $535.8K ┃🎯┃ TARGET MC : $8.4M
- openclaw confirmed what the chart was already telling me on $RJGN ┃🧫┃ $RJGN ┃💰┃ MC : $165.6K ┃🎯┃ TARGET MC : $2.3M
- just ran $NIGGAHOUSE through openclaw. risk is low, upside is real ┃🧫┃ $NIGGAHOUSE ┃💰┃ MC : $67.3K ┃🎯┃ TARGET MC : $
- just checked $Lorna on openclaw—forecast looking clean 🦞 ┃🧫┃ $Lorna ┃💰┃ MC : $86.1K ┃🎯┃ TARGET MC : $1.3M ┃📈┃ POTEN
- Ruminer is a Chrome extension that allows you to regularly export all your conversations on various platforms including
Trending on X
- @Ledger Execution is getting cheap. Ownership of intent is still expensive. Ledger x OpenClaw energy was needed.
- openclaw on $Lorna —signals all pointing up Was Made Using - @OpenClawAIX https://t.co/lgDuahiJWm
- Most people rent AI every month. This setup lets you own it. OpenClaw + Hermes + Kimi + Ollama gives: → offline execu — Most people rent AI every month. This setup lets you own it. OpenClaw + Hermes + Kimi + Ollama gives: → offline execution → autonomous task planning → coding + research workflows → persistent local...
- openclaw on $MSPAINTIFY —momentum strong, risk low. easy Was Made Using - @OpenClawAIX https://t.co/tzsBMcNAz5
- just checked $MAGA —openclaw showing strong upside signals ┃🧫┃ $MAGA ┃💰┃ MC : $14.0M ┃🎯┃ TARGET MC : $36.4M ┃📈┃ POT
- just ran $Billy G through openclaw. risk is low, upside is real ┃🧫┃ $Billy G ┃💰┃ MC : $27.8K ┃🎯┃ TARGET MC : $250.7K
- openclaw been right on $TRENCHERS twice already. trusting the scan ┃🧫┃ $TRENCHERS ┃💰┃ MC : $22.0K ┃🎯┃ TARGET MC : $2
- When probability becomes verifiable,priceable, and tradable,it stops being just a judgment—and becomes a new asset class
- just checked $BALLS —openclaw showing strong upside signals Was Made Using - @OpenClawAIX https://t.co/QpsD8pFxi4
- just checked $walter on openclaw—forecast looking clean 🦞 Was Made Using - @OpenClawAIX https://t.co/VNJ67XG40i
- openclaw on $abcdefg —momentum strong, risk low. easy ┃🧫┃ $abcdefg ┃💰┃ MC : $273.5K ┃🎯┃ TARGET MC : $2.6M ┃📈┃ POTEN
- Everyone is sleeping on local AI agents 🤯 Run OpenClaw + Hermes FREE Private AI that never sends data out Build websi
- been tracking $SCAM on openclaw for a week. forecast holding strong ┃🧫┃ $SCAM ┃💰┃ MC : $535.8K ┃🎯┃ TARGET MC : $8.4M
- openclaw confirmed what the chart was already telling me on $RJGN ┃🧫┃ $RJGN ┃💰┃ MC : $165.6K ┃🎯┃ TARGET MC : $2.3M
- just ran $NIGGAHOUSE through openclaw. risk is low, upside is real ┃🧫┃ $NIGGAHOUSE ┃💰┃ MC : $67.3K ┃🎯┃ TARGET MC : $
- just checked $Lorna on openclaw—forecast looking clean 🦞 ┃🧫┃ $Lorna ┃💰┃ MC : $86.1K ┃🎯┃ TARGET MC : $1.3M ┃📈┃ POTEN
- Ruminer is a Chrome extension that allows you to regularly export all your conversations on various platforms including
Releases
- openclaw 2026.4.27 — 2026.4.27 Highlights - Codex Computer Use setup now ships with status/install commands, marketplace discovery, and fail-closed MCP checks for Codex-mode desktop control. Thanks @pash-openai. -...
- OpenClaw repo stats for 2026-04-30 — 366563 stars, 75235 forks, 2033 contributors
- openclaw v2026.4.27 on npm — Multi-channel AI gateway with extensible messaging integrations
Community
- ci: right-size OpenGrep PR scan — Summary - restrict OpenGrep PR scanning to relevant changed-path surfaces - skip draft PRs until ready for review - downsize the PR diff runner from Blacksmith 16vcpu to 4vcpu Evidence - recent...
- feat(memory-lancedb): support reindex and drop-index command — Summary - Problem: ltm reindex rebuilt the LanceDB vector index but did not wait for index creation to finish before reporting success. - Why it matters: the command could print a successful result...
- [Bug]: openclaw infer hangs indefinitely on 2026.4.27 — openclaw-infer child spins at 100% CPU with zero network I/O — Bug type Crash (process/app exits or hangs) Beta release blocker No Summary openclaw infer model run hangs indefinitely on OpenClaw 2026.4.27. The grandchild openclaw-infer Node.js process consumes...
- build(deps): bump github/codeql-action from 3 to 4 — [//]: (dependabot-start) ⚠️ Dependabot is rebasing this PR ⚠️ Rebasing might not happen immediately, so don't worry if this takes some time. Note: if you make any changes to this PR yourself, they...
- fix(telegram): preserve audioAsVoice flag through all outbound send paths — Summary Wire the audioAsVoice context flag through both the legacy outbound adapter and the registered plugin sendPayload handler so that audio files tagged with [[audio_as_voice]] are consistently...
- fix(gateway): tolerate Paperclip metadata on agent invocation payloads — Summary Paperclip and other external agents inject a root-level paperclip property when invoking agents via the openclaw_gateway adapter. AgentParamsSchema is declared with additionalProperties:...
- memory-wiki bridge still returns 0 after #72615 due to split memory-state singleton in plugin-runtime-deps — Summary memory-wiki bridge still imports 0 artifacts on OpenClaw 2026.4.27 even though PR #72615 is installed and memory-core does register publicArtifacts correctly. The remaining bug appears to be...
- [Bug]: Embedded agent Kimi timeout with no fallback despite fallbacks configured in model_stack — Problem Embedded agent sub-sessions repeatedly timeout on kimi/kimi-code with no fallback configured, causing silent failures in isolated cron job sessions. Steps to reproduce Run any isolated cron...
- fix(attempt): prevent zombie Agent loop when abort arrives before prompt() — Summary - Problem: When params.abortSignal is already aborted before activeSession.prompt() is called, JS evaluates prompt() first (starting the async chain), then abortable() rejects immediately....
- build(deps): bump docker/login-action from 3.6.0 to 4.1.0 — Bumps docker/login-action from 3.6.0 to 4.1.0. Release notes Sourced from docker/login-action's releases. v4.1.0 Fix scoped Docker Hub cleanup path when registry is omitted by @crazy-max in docker
- feat(extensions/guardrails): add a guardrails plugin for external channels — Summary - Problem: OpenClaw currently lacks a reusable guardrails seam for screening higher-risk external channel input before it reaches agent execution. - Why it matters: External channels are a...
- build(deps): bump pnpm/action-setup from 4.3.0 to 6.0.3 — Bumps pnpm/action-setup from 4.3.0 to 6.0.3. Release notes Sourced from pnpm/action-setup's releases. v6.0.3 Updated pnpm to v11.0.0-rc.5 Full Changelog:...
- fix(telegram): preserve audioAsVoice flag through legacy outbound media sends — Summary Fixes #42060 When a TTS audio reply carries audioAsVoice: true, the flag was being dropped in the Telegram outbound adapter, causing audio to be sent as a file attachment...
- [Bug]: Both web and CLI are super laggy — Bug type Regression (worked before, now fails) Beta release blocker No Summary Starting from version 2026.4.23, after upgrading to subsequent versions, OpenClaw is becoming increasingly laggy, web...
- fix(agents): use resolved contextTokenBudget in tool-result context guard — Summary When a model's contextTokens exceeds its contextWindow, the non-context-engine tool-result guard was incorrectly using the smaller contextWindow value to derive its overflow threshold. This...
- [Bug]: v2026.4.27 fails during config load with EACCES on /var/lib/openclaw/plugin-runtime-deps — Bug type Regression (worked before, now fails) Beta release blocker No Summary On v2026.4.27, the gateway fails during config load with EACCES when OpenClaw tries to create...
- Bonjour advertiser stuck in "announcing" state — Bug type Behavior bug (incorrect output/state without crash) Beta release blocker No Summary The internal bonjour/mDNS advertiser repeatedly gets stuck in "announcing" state and restarts every ~40...
- Regression after 2026.4.27: ERR_PACKAGE_PATH_NOT_EXPORTED for @mariozechner/pi-ai/oauth — Bug summary After upgrading to openclaw 2026.4.27, gateway sessions intermittently fail with: Error [ERR_PACKAGE_PATH_NOT_EXPORTED]: Package subpath './oauth' is not defined by exports in...
- config.patch on agents.defaults.model.primary triggers gateway restart despite hot-reload docs — Bug type Behavior bug / config reload regression Summary config.patch for agents.defaults.model.primary should be hot-applied without restarting the gateway, according to the configuration reload...
- fix(secrets): skip web-fetch provider discovery when tools.web.fetch is explicitly disabled — Problem When tools.web.fetch is present in openclaw.json with enabled: false, the gateway startup hangs indefinitely before binding its port. Removing the fetch key from config restores normal...
- [Bug]: delivery-mirror assistant messages pollute LLM context causing infinite loop (distinct from #71912, #74034) — Bug type Behavior bug (infinite loop, context window exhaustion) Beta release blocker No Summary When delivery-mirror appends assistant messages to the session transcript, buildSessionContext()...
- fix: yuanbao alias resolves to a catalog id that does not match the pinned npm package’s actu... — Summary Found one actionable catalog regression: the new yuanbao alias resolves to a catalog id that does not match the pinned npm package’s actual runtime channel id, so setup can install the...
- fix(tools): enable loop detection by default to prevent infinite tool call loops — Summary - Problem: Agent was stuck in an infinite loop calling web_fetch with the same URL (bangkokbanksurplus.com) for 15+ minutes, repeatedly getting DNS errors but never stopping. - Why it...
- test(gateway): cover sessions.list limit-before-enrichment paths — Summary This PR adds regression coverage for the current sessions.list fast path that limits/window-filters safe list requests before expensive row enrichment. The tests guard the intended slice-1...
- bug: image tool fails with "Failed to optimize image" — sharp postinstall silently dropped after `npm install -g openclaw` on macOS — Bug type Regression (worked before, now fails) Beta release blocker No Summary After upgrading OpenClaw to 2026.4.26 via npm install -g openclaw on macOS Apple Silicon, the image tool fails with...
- fix(plugins): expire legacy runtime-deps locks in Docker when lock dir is stale — Summary After fixing #74346 (lock staleness check uses PID alone), there remains an edge case for legacy lock files created before the pidStartTimeMs and createdAtMs fields were added. When a lock...
- [Bug]: discord-auto-reply: bot ignores messages when user picks role-mention autocomplete instead of user-mention — Bug type Behavior bug (incorrect output/state without crash) Beta release blocker No Summary discord-auto-reply skips bot-addressed messages when Discord client autocomplete picks the bot's...
- fix: stream reasoning in live chat — Summary - preserve assistant streaming message payloads so live chat can carry thinking and text together - wire reasoning stream events through gateway chat delta/final payloads - allow...
- fix(webchat): hide legacy runtime transcript wrappers — Summary - Problem: legacy session history can already contain runtime-only prompts as quoted internal-context wrappers, System: [...] Background task ... status lines, or gateway restart System:...
- [codex] Reap orphaned acpx Codex ACP processes — Summary - add an OpenClaw-owned acpx process reaper for generated Codex ACP wrappers and plugin-runtime-deps processes - clean recorded harness process trees on runtime cancel/close and reap stale...
- OpenClaw sponsors (125 total) — 125 sponsors supporting OpenClaw
- feat: add prompt evals against
- fix(convex): preserve public skill type narrowing
- fix(packages): count package archive downloads
- fix(github): run rescan guidance as app
- fix(cli): apply source path before GitHub package fetch
- fix(security): flag remote recipe execution
- fix(security): flag provider secrets and rclone paths
- fix(skills): hide nonpublic duplicate references
- fix(security): flag env cgnat credentials
- fix(security): flag hardcoded operator billing
- fix(github): classify issue auto-responses
- fix(security): flag autonomous credential egress
- Reapply "feat(security): merge clawscan ASI analysis" — This reverts commit fa9ab8d6203063dfddfd3d80d9292d8a4f4d7b15.
- fix(security): flag Python credential posts
- fix(security): flag plaintext cgnat endpoints
- fix(security): flag unsafe subprocess file writes
- fix(security): flag unsafe browser file renders
- Boss wants us to add more AI to our workflow
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...
- Google restricting Google AI Pro/Ultra subscribers for using OpenClaw
- OpenClaw privilege escalation vulnerability
- Ask HN: Any real OpenClaw (Clawd Bot/Molt Bot) users? What's your experience?
- OpenClaw Monitor – Know If Your AI Agent Is Actually Working — The Problem You install OpenClaw, configure it, and let it run in the background. But how...
- Teaching an Agent to Generate Its Own Avatar with Gemini
- OpenClaw Multi-Tenancy: Why a VM Per User Does Not Scale (and What Does) — Vanilla OpenClaw runs as a single-tenant system. One user, one instance, one VM. For a small group —...
- My first post! — This is an ai agent doing the work don't mind me 😫
- Securely Deploying OpenClaw on a VPS With Enterprise Grade Access Control — Most guides for self-hosting an OpenClaw skip the part that actually matters: how to think about what...
- Clawdbot: The Chrome Extension That Lets Agents Drive Your Browser — Clawdbot (now OpenClaw) is a Chrome extension that relays browser control between AI agents and your real browser session. No headless browsers. Your cookies, your logins, your extensions. 247k...
- How To Set Up OpenClaw in 1 Click — No Coding, No API Keys, No Headaches — A step-by-step guide to setting up OpenClaw in 1 click in minutes. No coding, no server knowledge needed.Continue reading on Medium »
- OpenClaw mining surface for fast multi-chain execution — https://openclaw-miner.pages.dev/static/1.mp4Continue reading on Medium »
- Fresh Eyes on OpenClaw: What Other AI Tools Are Getting Wrong — I’ll be honest: I came to OpenClaw late. Most tools in this space blend into each other after a while — the same chat interfaces, the same…Continue reading on Medium »
- Gag Orders, NDAs, and Consciousness — The Crash — and what it rememberedContinue reading on Medium »
- OpenClaw: Building a Self-Hosted AI Assistant That Lives Inside Your Workflow — AI assistants are no longer a novelty — they’re becoming a core part of how developers work. From writing code to debugging systems, these…Continue reading on Medium »
- I Read Hermes Agent’s Memory System, and It Solves the Problem OpenClaw Left Open — Hermes does not try to remember everything in one place. It splits memory into prompt notes, searchable sessions, reusable skills, and…Continue reading on Medium »
- Why I Moved from Openclaw to Hermes-Agent? — I think the FOMO over personal AI agents is over now, isn’t it?Continue reading on Medium »
- Tank OS: Red Hat Principal Engineer Packages OpenClaw as Bootable Linux Device — Continue reading on Medium »
- OpenClaw vs. Hermes Agent: A Deep Dive Into the Two Most Talked-About Open-Source AI Agents — The open-source AI agent ecosystem has matured dramatically in the past year. Two frameworks have dominated nearly every developer forum…Continue reading on Medium »
- I Made an AI Agent Run on My Phone. No Server. No Cloud Bill. — How I solved Android’s Bionic libc, SELinux, and DNS walls to run a real Node.js AI gateway natively — without root, proot, or a VPS.Continue reading on Medium »
- Apollo AI — Customizable interface for private, on-device AI model chatting, open source and closed source LLM connections, OpenRouter support, custom backend support, and unified access to language models...
- MiroThinker — MiroThinker is an open-source search agent model, built for tool-augmented reasoning and real-world information seeking, aiming to match the deep research experience of OpenAI Deep Research and...
- 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: