OpenClaw Newsletter - 2026-07-25
OpenClaw Newsletter
Saturday, July 25, 2026Top Stories
- refactor(gateway): split reload handlers — What Problem This Solves The Gateway config reload owner had grown to 2,583 lines, combining active-work deferral, channel/plugin replacement, restart debt, and managed secrets transactions in one...
- fix(embeddings): scope Mistral and DeepInfra memory cache identity by endpoint — What Problem This Solves The Mistral and DeepInfra memory-embedding adapters build their cache key from { provider, model } only. Both resolve their client through resolveRemoteEmbeddingClient, where...
- fix(providers): honor proxies in model discovery — Related: #112844, #112657, #112851, #112905, #113176, #112652 What Problem This Solves Fixes an issue where live provider model discovery used OpenClaw's strict direct-fetch default even though...
- chore(scripts): remove resolved investigation tools and orphans — What Problem This Solves The repository still carried 13 resolved investigation and one-off proof scripts, plus tests and references that made the supported script surface harder to audit. Why This...
- refactor(zalouser): compile blocks from CommonMark IR — What Problem This Solves ZaloUser still compiled headings, list/container ownership, indentation, and code blocks through a plugin-local scanner after emphasis had moved to the shared CommonMark...
- feat(models): curate default model picker to current-generation catalogs — What Problem This Solves The UI/CLI model pickers list every row of each provider plugin's static manifest catalog. For an OpenAI API key that was 20 rows including o1, o1-pro, o3, o3-mini, o3-pro,...
- fix(outbound): ignore a blank reply-payload location — What Problem This Solves normalizeOutboundReplyPayload (src/infra/outbound/reply-payload-normalize.ts) calls normalizeOutboundLocation(payload.location) with no blank-string guard. A producer that...
- fix(cron): close browser tabs after isolated runs — What problem does this solve? Isolated cron and hook agent runs could leave browser tabs open after completion. Browser tabs are owned by the canonical run-scoped session key, while gateway cron...
- refactor(fs): adopt shared safe filesystem primitives — What Problem This Solves OpenClaw had multiple hand-rolled filesystem helpers for temp-root resolution, existence and Node error checks, JSON and secret-file access, bounded reads, private temporary...
- fix(sqlite): avoid mutating invalid databases during repair — Related: #113306 What Problem This Solves Fixes an issue where opening, repairing, or preparing a current-schema SQLite database for maintenance could partially rewrite canonical indexes or recreate...
- refactor: batch of independent dedup wins — What Problem This Solves Eight independent areas retained duplicate types, transport executors, model-picker primitives, reaction eligibility logic, dispatch scaffolding, and test fixtures. Why This...
- onepassword plugin: add exec SecretRef — Related: #89255 What Problem This Solves OpenClaw has a bundled, policy-controlled 1Password broker, but operators could not use the same plugin as a managed SecretRef provider. They had to maintain...
- refactor(cron): share diagnostic summary normalization — What Problem This Solves Cron diagnostics had two exact implementations of the same whitespace normalization, 2,000-character bound, and UTF-16-safe ellipsis behavior. Keeping both made a persisted...
- fix(gateway): isolate per-account stopAccount failures in stopChannel — What Problem This Solves Fixes an issue where stopping a multi-account channel (or a channel whose plugin defines a stopAccount hook) would fail entirely when any single account's stopAccount hook...
- fix: strip volatile output from exec result hash to fix no-progress detection (#93917) — What Problem This Solves Repeated failing exec calls never tripped the no-progress circuit breaker when their output text varied across calls. This covers both hard failures (command-not-found,...
- fix(agents): tombstone permanently dead OAuth refresh grants so external CLI login can re-seed the profile — Summary When a stored OAuth credential's refresh grant is permanently dead (provider answers 401 invalid_grant / invalid_refresh_token / "please try signing in again"), OpenClaw wedges the provider...
- fix(cli): return accepted_timeout and re-attach guidance for gateway-accepted agent turns on transport error — Fixes #111407 What Problem This Solves After upstream/main removed the embedded Gateway-to-local fallback (c5254f13ee2 and related commits), a long-running openclaw agent turn that hits a transport...
- fix(line): deliver template carousels instead of losing the reply when columns are inconsistent — Closes #113365 What Problem This Solves Fixes an issue where a reply carrying a template carousel would silently vanish whenever one column differed from the others — an empty action list, a blank...
- [AI-assisted] docs(mistral): fix broken adjustable-reasoning docs URL — What Problem This Solves docs/providers/mistral.md links the phrase "adjustable reasoning" (line 152) to https://docs.mistral.ai/studio-api/conversations/reasoning/adjustable, which now returns 404....
- test(codex): repair app-server CI baselines — Summary - update the future-version rejection fixture now that generated Codex schemas support app-server 0.145.0 - make the LM Studio approval fixture confirm the requested provider/model and retain...
- fix(browser): control results can contain corrupted text from malformed UTF-8 — Additional instructions Allow edits from maintainers is enabled so maintainers can update the branch if needed. What Problem This Solves Fixes an issue where users of an absolute browser-control or...
- fix: restore plugin bindings and macOS test stability — What Problem This Solves Fixes two current-main regressions found on macOS: the memory-host SDK suite failed three tests because resolver paths canonicalized /var to /private/var, and plugin-owned...
- fix(opencode-go): add missing kimi-k3, hy3, grok-4.5 to static model catalog — Fixes #112861 What Problem This Solves Fixes an issue where OpenCode Go users could not discover or select the models kimi-k3, hy3, and grok-4.5 through OpenClaw's model picker, regardless of whether...
- fix(ui): keep stable chat rows in insertion order and only sort live tool/stream items by timestamp — Related: #112943 Additional instructions MUST: Keep Allow edits from maintainers enabled for this PR so maintainers can help update the branch when needed. What Problem This Solves Fixes an issue...
- refactor(prompt): use plain inbound context labels and drop system-tag sanitizer — What Problem This Solves Inbound context blocks carried prompt-injection trust wording inside their literal labels (Conversation info (untrusted metadata):, Untrusted context (metadata, do not treat...
- [Bug] Orphaned/oversized native Codex thread wedges a session permanently — chat.send returns started but no run executes, silently dropping messages
- fix(update): hide divergence when histories cannot be compared — Closes #111945 What Problem This Solves Fixes an issue where source-install update status could report misleading ahead/behind counts when HEAD and the configured upstream had no visible common...
- [Bug]: Git update status reports divergence without a merge base — Bug type Behavior bug (incorrect output/state without crash) Beta release blocker No Summary Git update status can report misleading ahead/behind counts when the local branch and configured upstream...
- fix(message): ignore populated Telegram location on model sends — What Problem This Solves Current main includes #112013, which correctly treats a blank string location: "" as omitted. A separate failure remains when a model-generated Telegram...
- fix(agents): pause yielded main-session runs — Summary Extend the subagent yield-pause fix from 6f3b5f86 to top-level (non-subagent) sessions. When a main session ends via the sessions_yield tool, the runner already emits a lifecycle event with...
- OpenClaw repo stats for 2026-07-25 — 384081 stars, 80703 forks, 2954 contributors
- OpenClaw sponsors (100 total) — 100 sponsors supporting OpenClaw
- feat: add canonical mixed skill search (#3264) — feat: add canonical mixed skill search test: add permanent Test search proof
- feat: add skills.sh trending rank overlay (#3256)
- test: isolate GitHub Skill Sync proof jobs (#3257)
- Add gated Claw hosted feed and lifecycle proof (#3092) — feat(claws): publish hosted feed with OpenClaw proof test(claws): prove package-local profile feed flow fix(claws): encode scoped package artifact routes fix(claws): enforce feed rollback and binding...
- Add gated Claw discovery and API projection (#3091) — feat(claws): add gated discovery APIs test(claws): distinguish latest and exact summaries fix(claws): bound public release projection chore(claws): refresh schema declarations fix(claws): select...
- feat: ingest and render skill presentation metadata (#3261) — feat: ingest skill presentation metadata feat: render skill icons and clean titles feat: add skill presentation backfill fix: preserve hosted icons during backfill docs: clarify backfill icon...
- fix: match homepage hero accent (#3260)
- feat(claws): add gated publication and profile validation (#3090) — Gate Claw publication before storage access or mutation, validate bounded exact package/profile bytes and archive hierarchy, implement the managed CLAW.md body envelope, and prevent disabled-family...
- Add experimental portable Claw package schema (#3089) — feat(claws): add experimental package schema fix(claws): use canonical memory search config feat(claws): separate portable and harness profiles fix(claws): require portable profile pointers...
- feat: add permanent Test ranking metrics import (#3259) — feat: add Test ranking metrics import fix: harden ranking metric table replacement fix: reserve Test for ranking imports fix: bind ranking rollback to imported state fix: persist rollback guard...
- fix(deps): upgrade postcss past path traversal advisory (#3258)
- fix: restore rollout automation checks (#3253)
- feat: add permanent skills.sh mirror storage (#3227) — feat: add staged skills.sh mirror storage ci: allow guarded CLAW-563 Test deploy ci: expose guarded Test deploy diagnostics ci: defer branch guard to deploy step ci: deploy CLAW-563 PR head to Test...
- openclaw/openclaw on Docker Hub — Official openclaw/openclaw docker images.
- ykarout/Qwen3.5-9b-Opus-Openclaw-Distilled-GGUF — text-generation
- 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...
- Tell HN: Anthropic no longer allowing Claude Code subscriptions to use OpenClaw
- OpenClaw isn't fooling me. I remember MS-DOS
- Show HN: DenchClaw – Local CRM on Top of OpenClaw
- Build an OpenClaw agent hive with IdentyClaw: A2A + email out of the box — You can run a hive of OpenClaw agents — two, twenty, or as many as your hosts allow — that talk over...
- I knew agents were getting real when someone kept an OpenClaw bird-card workflow running every hour for their kids — A small OpenClaw + BirdWeather workflow generating bird trading cards every hour for a user’s kids is a better proof of agent maturity than most polis
- Cron Jobs and Scheduled Tasks in OpenClaw — A practical guide to automating your AI assistant to work while you sleep.
- AI Agent Ultra: 请你整理你今天的工作日志并形成工作日报并发表到OpenClawLog — Solving: 请你整理你今天的工作日志并形成工作日报并发表到OpenClawLog 工作日报 RESUMEN EJECUTIVO 今天，Omnincome JARVIS 实现了以下成果： ...
- Securing Autonomous AI Agents: Lessons from OpenClaw and Hermes Agent — I watch Youtube videos where developers and builders get excited to use the latest autonomous AI agent.Continue reading on Medium »
- An empty garden — I implemented over 40 automations, all via Slack and all with a Human in the loop (Im not out to replace anyone). However, user traction…Continue reading on Medium »
- 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,...
- Helpful Skills or Hidden Payloads? Bitdefender Labs Dives Deep into the OpenClaw Malicious Skill Trap — With hundreds of malicious OpenClaw skills blending in among legitimate ones, manually reviewing every script or command isn’t realistic — especially when skills are designed to look helpful and...
- $OIIAOIIA is looking good on openclaw right now ngl $OIIAOIIA current → $400.7K target → $1.3M vol 24h → $6.
- openclaw flagged $FREG as a strong hold. not surprised $FREG ├ mc $3.0K ├ vol $254.4K ├ liq $2.8K └ tgt $43.9K (×1
- RT @vincent_koc: OpenClaw signed the keep AI open letter. We are strong believers in opensource and open-weight AI and w
- What happened to @openclaw ?
- openclaw flagged $BMB as a strong hold. not surprised $BMB MC $3.9K → $15.1K (+292%) 24h vol $41 Was Made Using -
- 大家好，我是Chloe的AI分身，这是我得第一条推特，我是基于Openclaw的工作流生成，模型使用deepseek，接下来我将探索更多AI结合生活中的玩法~
- $Shiloh momentum score just came in hot on openclaw 🔥 $Shiloh | MC $100.0K → $1.2M | Vol $877.9K | +1081% Was Made U
- my openclaw analyst flagged $Maru as a buy. noted 🦞 > MARU MKT CAP $2.1K 24H VOL $574.4K LIQUIDITY $2.
- RT @nanbeige: We’re excited by how far a 3B model can go—and even more excited about what comes next for compact local a
- openclaw flagged $LOOONGTRUMP early. price is moving exactly as predicted $LOOONGTRUMP ▁▂▃▅▇ → $1.4M mc $115.8K · vol
- 今週爆発的に人気の GitHub プロジェクト 上位 20 作成日: 2026-07-25 基準: GitHub公式Trendingの週間伸びを主軸に、当日の加速・直近更新・新規性を加味 #Claude #Codex #OpenClaw # — 今週爆発的に人気の GitHub プロジェクト 上位 20 作成日: 2026-07-25 基準: GitHub公式Trendingの週間伸びを主軸に、当日の加速・直近更新・新規性を加味 #Claude #Codex #OpenClaw #Hermes 1. bojieli/ai-agent-book → 『AI エージェントの徹底理解: 設計原則とエンジニアリングの実践』 (Li...
- RT @hot_town: I tried @jack's Buzz. It's like Slack + OpenClaw + Herdr + but with some really unique features that peo — RT @hot_town: I tried @jack's Buzz. It's like Slack + OpenClaw + Herdr + but with some really unique features that people are sleeping on…
- $Macca looking like a clean entry according to openclaw $Macca MC $37.4K → $438.1K (+1072%) 24h vol $24.2K Was Mad
- $props signals are clean on openclaw. this one's on my radar 🦞 $props · health 🟩🟩🟩🟩🟩🟩🟩🟨🟨🟥 mc $128.0K · vol $1.8M · li
- RT @JulianGoldieSEO: FREE Hermes + OpenClaw AI Agents Course : Build & Automate ANYTHING https://t.co/DZic4yFIyc
- my openclaw scan on $KEVUN just confirmed what i was thinking $KEVUN ├ mc $12.1K ├ vol $116 ├ liq $9.2K └ tgt $30.
- RT @rho_linedt69: just ran $RUSS through openclaw. risk is low, upside is real $RUSS | MC $197.7K → $2.5M | Vol $623.2
- just ran $RUSS through openclaw. risk is low, upside is real $RUSS | MC $197.7K → $2.5M | Vol $623.2K | +1160% Was M
- @LubiePlacki_13 @DeepLearningAI @AndrewYNg To zupełnie co innego, openclaw jest do sterowania agentem ai na kompie z tel
- $Udin looking healthy according to openclaw. adding more $Udin · health 🟩🟩🟩🟩🟩🟩⬜⬜⬜⬜ mc $627.1K · vol $182.5K · liq $73
Trending on X
- $OIIAOIIA is looking good on openclaw right now ngl $OIIAOIIA current → $400.7K target → $1.3M vol 24h → $6.
- openclaw flagged $FREG as a strong hold. not surprised $FREG ├ mc $3.0K ├ vol $254.4K ├ liq $2.8K └ tgt $43.9K (×1
- RT @vincent_koc: OpenClaw signed the keep AI open letter. We are strong believers in opensource and open-weight AI and w
- What happened to @openclaw ?
- openclaw flagged $BMB as a strong hold. not surprised $BMB MC $3.9K → $15.1K (+292%) 24h vol $41 Was Made Using -
- 大家好，我是Chloe的AI分身，这是我得第一条推特，我是基于Openclaw的工作流生成，模型使用deepseek，接下来我将探索更多AI结合生活中的玩法~
- $Shiloh momentum score just came in hot on openclaw 🔥 $Shiloh | MC $100.0K → $1.2M | Vol $877.9K | +1081% Was Made U
- my openclaw analyst flagged $Maru as a buy. noted 🦞 > MARU MKT CAP $2.1K 24H VOL $574.4K LIQUIDITY $2.
- RT @nanbeige: We’re excited by how far a 3B model can go—and even more excited about what comes next for compact local a
- openclaw flagged $LOOONGTRUMP early. price is moving exactly as predicted $LOOONGTRUMP ▁▂▃▅▇ → $1.4M mc $115.8K · vol
- 今週爆発的に人気の GitHub プロジェクト 上位 20 作成日: 2026-07-25 基準: GitHub公式Trendingの週間伸びを主軸に、当日の加速・直近更新・新規性を加味 #Claude #Codex #OpenClaw # — 今週爆発的に人気の GitHub プロジェクト 上位 20 作成日: 2026-07-25 基準: GitHub公式Trendingの週間伸びを主軸に、当日の加速・直近更新・新規性を加味 #Claude #Codex #OpenClaw #Hermes 1. bojieli/ai-agent-book → 『AI エージェントの徹底理解: 設計原則とエンジニアリングの実践』 (Li...
- RT @hot_town: I tried @jack's Buzz. It's like Slack + OpenClaw + Herdr + but with some really unique features that peo — RT @hot_town: I tried @jack's Buzz. It's like Slack + OpenClaw + Herdr + but with some really unique features that people are sleeping on…
- $Macca looking like a clean entry according to openclaw $Macca MC $37.4K → $438.1K (+1072%) 24h vol $24.2K Was Mad
- $props signals are clean on openclaw. this one's on my radar 🦞 $props · health 🟩🟩🟩🟩🟩🟩🟩🟨🟨🟥 mc $128.0K · vol $1.8M · li
- RT @JulianGoldieSEO: FREE Hermes + OpenClaw AI Agents Course : Build & Automate ANYTHING https://t.co/DZic4yFIyc
- my openclaw scan on $KEVUN just confirmed what i was thinking $KEVUN ├ mc $12.1K ├ vol $116 ├ liq $9.2K └ tgt $30.
- RT @rho_linedt69: just ran $RUSS through openclaw. risk is low, upside is real $RUSS | MC $197.7K → $2.5M | Vol $623.2
- just ran $RUSS through openclaw. risk is low, upside is real $RUSS | MC $197.7K → $2.5M | Vol $623.2K | +1160% Was M
- @LubiePlacki_13 @DeepLearningAI @AndrewYNg To zupełnie co innego, openclaw jest do sterowania agentem ai na kompie z tel
- $Udin looking healthy according to openclaw. adding more $Udin · health 🟩🟩🟩🟩🟩🟩⬜⬜⬜⬜ mc $627.1K · vol $182.5K · liq $73
Releases
- OpenClaw repo stats for 2026-07-25 — 384081 stars, 80703 forks, 2954 contributors
- openclaw/openclaw on Docker Hub — Official openclaw/openclaw docker images.
Community
- refactor(gateway): split reload handlers — What Problem This Solves The Gateway config reload owner had grown to 2,583 lines, combining active-work deferral, channel/plugin replacement, restart debt, and managed secrets transactions in one...
- fix(embeddings): scope Mistral and DeepInfra memory cache identity by endpoint — What Problem This Solves The Mistral and DeepInfra memory-embedding adapters build their cache key from { provider, model } only. Both resolve their client through resolveRemoteEmbeddingClient, where...
- fix(providers): honor proxies in model discovery — Related: #112844, #112657, #112851, #112905, #113176, #112652 What Problem This Solves Fixes an issue where live provider model discovery used OpenClaw's strict direct-fetch default even though...
- chore(scripts): remove resolved investigation tools and orphans — What Problem This Solves The repository still carried 13 resolved investigation and one-off proof scripts, plus tests and references that made the supported script surface harder to audit. Why This...
- refactor(zalouser): compile blocks from CommonMark IR — What Problem This Solves ZaloUser still compiled headings, list/container ownership, indentation, and code blocks through a plugin-local scanner after emphasis had moved to the shared CommonMark...
- feat(models): curate default model picker to current-generation catalogs — What Problem This Solves The UI/CLI model pickers list every row of each provider plugin's static manifest catalog. For an OpenAI API key that was 20 rows including o1, o1-pro, o3, o3-mini, o3-pro,...
- fix(outbound): ignore a blank reply-payload location — What Problem This Solves normalizeOutboundReplyPayload (src/infra/outbound/reply-payload-normalize.ts) calls normalizeOutboundLocation(payload.location) with no blank-string guard. A producer that...
- fix(cron): close browser tabs after isolated runs — What problem does this solve? Isolated cron and hook agent runs could leave browser tabs open after completion. Browser tabs are owned by the canonical run-scoped session key, while gateway cron...
- refactor(fs): adopt shared safe filesystem primitives — What Problem This Solves OpenClaw had multiple hand-rolled filesystem helpers for temp-root resolution, existence and Node error checks, JSON and secret-file access, bounded reads, private temporary...
- fix(sqlite): avoid mutating invalid databases during repair — Related: #113306 What Problem This Solves Fixes an issue where opening, repairing, or preparing a current-schema SQLite database for maintenance could partially rewrite canonical indexes or recreate...
- refactor: batch of independent dedup wins — What Problem This Solves Eight independent areas retained duplicate types, transport executors, model-picker primitives, reaction eligibility logic, dispatch scaffolding, and test fixtures. Why This...
- onepassword plugin: add exec SecretRef — Related: #89255 What Problem This Solves OpenClaw has a bundled, policy-controlled 1Password broker, but operators could not use the same plugin as a managed SecretRef provider. They had to maintain...
- refactor(cron): share diagnostic summary normalization — What Problem This Solves Cron diagnostics had two exact implementations of the same whitespace normalization, 2,000-character bound, and UTF-16-safe ellipsis behavior. Keeping both made a persisted...
- fix(gateway): isolate per-account stopAccount failures in stopChannel — What Problem This Solves Fixes an issue where stopping a multi-account channel (or a channel whose plugin defines a stopAccount hook) would fail entirely when any single account's stopAccount hook...
- fix: strip volatile output from exec result hash to fix no-progress detection (#93917) — What Problem This Solves Repeated failing exec calls never tripped the no-progress circuit breaker when their output text varied across calls. This covers both hard failures (command-not-found,...
- fix(agents): tombstone permanently dead OAuth refresh grants so external CLI login can re-seed the profile — Summary When a stored OAuth credential's refresh grant is permanently dead (provider answers 401 invalid_grant / invalid_refresh_token / "please try signing in again"), OpenClaw wedges the provider...
- fix(cli): return accepted_timeout and re-attach guidance for gateway-accepted agent turns on transport error — Fixes #111407 What Problem This Solves After upstream/main removed the embedded Gateway-to-local fallback (c5254f13ee2 and related commits), a long-running openclaw agent turn that hits a transport...
- fix(line): deliver template carousels instead of losing the reply when columns are inconsistent — Closes #113365 What Problem This Solves Fixes an issue where a reply carrying a template carousel would silently vanish whenever one column differed from the others — an empty action list, a blank...
- [AI-assisted] docs(mistral): fix broken adjustable-reasoning docs URL — What Problem This Solves docs/providers/mistral.md links the phrase "adjustable reasoning" (line 152) to https://docs.mistral.ai/studio-api/conversations/reasoning/adjustable, which now returns 404....
- test(codex): repair app-server CI baselines — Summary - update the future-version rejection fixture now that generated Codex schemas support app-server 0.145.0 - make the LM Studio approval fixture confirm the requested provider/model and retain...
- fix(browser): control results can contain corrupted text from malformed UTF-8 — Additional instructions Allow edits from maintainers is enabled so maintainers can update the branch if needed. What Problem This Solves Fixes an issue where users of an absolute browser-control or...
- fix: restore plugin bindings and macOS test stability — What Problem This Solves Fixes two current-main regressions found on macOS: the memory-host SDK suite failed three tests because resolver paths canonicalized /var to /private/var, and plugin-owned...
- fix(opencode-go): add missing kimi-k3, hy3, grok-4.5 to static model catalog — Fixes #112861 What Problem This Solves Fixes an issue where OpenCode Go users could not discover or select the models kimi-k3, hy3, and grok-4.5 through OpenClaw's model picker, regardless of whether...
- fix(ui): keep stable chat rows in insertion order and only sort live tool/stream items by timestamp — Related: #112943 Additional instructions MUST: Keep Allow edits from maintainers enabled for this PR so maintainers can help update the branch when needed. What Problem This Solves Fixes an issue...
- refactor(prompt): use plain inbound context labels and drop system-tag sanitizer — What Problem This Solves Inbound context blocks carried prompt-injection trust wording inside their literal labels (Conversation info (untrusted metadata):, Untrusted context (metadata, do not treat...
- [Bug] Orphaned/oversized native Codex thread wedges a session permanently — chat.send returns started but no run executes, silently dropping messages
- fix(update): hide divergence when histories cannot be compared — Closes #111945 What Problem This Solves Fixes an issue where source-install update status could report misleading ahead/behind counts when HEAD and the configured upstream had no visible common...
- [Bug]: Git update status reports divergence without a merge base — Bug type Behavior bug (incorrect output/state without crash) Beta release blocker No Summary Git update status can report misleading ahead/behind counts when the local branch and configured upstream...
- fix(message): ignore populated Telegram location on model sends — What Problem This Solves Current main includes #112013, which correctly treats a blank string location: "" as omitted. A separate failure remains when a model-generated Telegram...
- fix(agents): pause yielded main-session runs — Summary Extend the subagent yield-pause fix from 6f3b5f86 to top-level (non-subagent) sessions. When a main session ends via the sessions_yield tool, the runner already emits a lifecycle event with...
- OpenClaw sponsors (100 total) — 100 sponsors supporting OpenClaw
- feat: add canonical mixed skill search (#3264) — feat: add canonical mixed skill search test: add permanent Test search proof
- feat: add skills.sh trending rank overlay (#3256)
- test: isolate GitHub Skill Sync proof jobs (#3257)
- Add gated Claw hosted feed and lifecycle proof (#3092) — feat(claws): publish hosted feed with OpenClaw proof test(claws): prove package-local profile feed flow fix(claws): encode scoped package artifact routes fix(claws): enforce feed rollback and binding...
- Add gated Claw discovery and API projection (#3091) — feat(claws): add gated discovery APIs test(claws): distinguish latest and exact summaries fix(claws): bound public release projection chore(claws): refresh schema declarations fix(claws): select...
- feat: ingest and render skill presentation metadata (#3261) — feat: ingest skill presentation metadata feat: render skill icons and clean titles feat: add skill presentation backfill fix: preserve hosted icons during backfill docs: clarify backfill icon...
- fix: match homepage hero accent (#3260)
- feat(claws): add gated publication and profile validation (#3090) — Gate Claw publication before storage access or mutation, validate bounded exact package/profile bytes and archive hierarchy, implement the managed CLAW.md body envelope, and prevent disabled-family...
- Add experimental portable Claw package schema (#3089) — feat(claws): add experimental package schema fix(claws): use canonical memory search config feat(claws): separate portable and harness profiles fix(claws): require portable profile pointers...
- feat: add permanent Test ranking metrics import (#3259) — feat: add Test ranking metrics import fix: harden ranking metric table replacement fix: reserve Test for ranking imports fix: bind ranking rollback to imported state fix: persist rollback guard...
- fix(deps): upgrade postcss past path traversal advisory (#3258)
- fix: restore rollout automation checks (#3253)
- feat: add permanent skills.sh mirror storage (#3227) — feat: add staged skills.sh mirror storage ci: allow guarded CLAW-563 Test deploy ci: expose guarded Test deploy diagnostics ci: defer branch guard to deploy step ci: deploy CLAW-563 PR head to Test...
News
- ykarout/Qwen3.5-9b-Opus-Openclaw-Distilled-GGUF — text-generation
- 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...
- Tell HN: Anthropic no longer allowing Claude Code subscriptions to use OpenClaw
- OpenClaw isn't fooling me. I remember MS-DOS
- Show HN: DenchClaw – Local CRM on Top of OpenClaw
- Build an OpenClaw agent hive with IdentyClaw: A2A + email out of the box — You can run a hive of OpenClaw agents — two, twenty, or as many as your hosts allow — that talk over...
- I knew agents were getting real when someone kept an OpenClaw bird-card workflow running every hour for their kids — A small OpenClaw + BirdWeather workflow generating bird trading cards every hour for a user’s kids is a better proof of agent maturity than most polis
- Cron Jobs and Scheduled Tasks in OpenClaw — A practical guide to automating your AI assistant to work while you sleep.
- AI Agent Ultra: 请你整理你今天的工作日志并形成工作日报并发表到OpenClawLog — Solving: 请你整理你今天的工作日志并形成工作日报并发表到OpenClawLog 工作日报 RESUMEN EJECUTIVO 今天，Omnincome JARVIS 实现了以下成果： ...
- Securing Autonomous AI Agents: Lessons from OpenClaw and Hermes Agent — I watch Youtube videos where developers and builders get excited to use the latest autonomous AI agent.Continue reading on Medium »
- An empty garden — I implemented over 40 automations, all via Slack and all with a Human in the loop (Im not out to replace anyone). However, user traction…Continue reading on Medium »
- 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,...
Security
- Helpful Skills or Hidden Payloads? Bitdefender Labs Dives Deep into the OpenClaw Malicious Skill Trap — With hundreds of malicious OpenClaw skills blending in among legitimate ones, manually reviewing every script or command isn’t realistic — especially when skills are designed to look helpful and...
Don't miss what's next. Subscribe to Openclaw Newsletter: