OpenClaw Newsletter - 2026-07-16
OpenClaw Newsletter
Thursday, July 16, 2026Top Stories
- openclaw 2026.7.2-beta.1 — 2026.7.2 Highlights - Remote coding sessions: run Control UI sessions on cloud workers, open Codex and Claude catalog sessions in terminals on their owning hosts, and resume OpenCode and Pi sessions...
- [Bug] Infinite DFS Recursion Crash on Circular Plugin Peer Linkage — Summary / Problem Statement In src/plugins/plugin-peer-link.ts, resolving plugin dependency peer links utilizes a Depth-First Search (DFS) traversal. If a circular linkage exists (e.g. Plugin A peers...
- perf: avoid event-loop stall during embedded_run bootstrap-context — Summary Fix event-loop stalls during embedded_run bootstrap-context that block the event loop for 14-22 seconds, causing message loss. What Problem This Solves Event-loop stalls during embedded_run...
- feat(memory-lancedb): scope memory_recall and memory_forget + configurable thresholds — Supersedes #102966 (closed: re-opened cleanly on current main; the bot review state there was stale and could not be refreshed). What Problem This Solves In a shared LanceDB memory store the memory...
- fix(agents): treat read-only shell inspection as replay-safe — What Problem This Solves Fixes #101890 — strict-agentic non-visible retry was blocked after turns that only ran proven read-only shell inspection (exec/bash with commands such as rg, ls, cat). Those...
- [Bug] Unhandled Promise Rejection in channel-server.ts on Connection Reset — Summary / Problem Statement In src/mcp/channel-server.ts, writing frame payloads to active client connections does not wrap write promises in local try-catch blocks. If a client abruptly resets the...
- feat: let plugins control inbound debounce policy — Related: #54565 What Problem This Solves Plugins cannot currently decide whether rich inbound WhatsApp messages should bypass debounce or choose a debounce window for a specific account, DM, group,...
- fix(hooks): wire message:sending internal hook to user-defined hooks — Summary The message:sending internal hook was registered but never invoked during outbound message delivery. Wires the hook into deliver.ts so user-defined hooks receive outgoing message events for...
- fix(codex): prevent corrupted UTF-8 in sandbox exec server output — What Problem This Solves Fixes an issue where the sandbox HTTP request handler (extensions/codex/src/app-server/sandbox-exec-server/http.ts) could receive corrupted stdout or stderr when a multibyte...
- fix(ui): explain replace-mode filtering in model picker — Closes #104545 What Problem This Solves When models.mode: "replace" filters the model catalog, the Control UI showed the resulting models without explaining why other bundled or discovered models...
- fix(agents): preserve ANSI sanitizer state across bash chunks — Closes #103668 What Problem This Solves Fixes an issue where users running bash commands through agents could see ANSI/OSC control-sequence fragments leak into streamed tool output, final bash...
- feat(mac): mascot accessory pack — nightcap, grad cap, hat-tip exit — What Problem This Solves The mascot's new working mood proved the pose-channel system extends well (#108735), but headwear was hardwired to one hat and one mood. Sleepy Clawd sleeps bare-headed,...
- fix(config/sessions): avoid splitting surrogate pairs when shortening group ids — What Problem This Solves buildGroupDisplayName shortens very long group ids via shortenGroupId, which previously used raw String.prototype.slice to take the first 6 and last 4 UTF-16 code units. When...
- [Bug]: VPN NEEDED users — Bug type Crash (process/app exits or hangs) Beta release blocker No Summary there is some users some HUMANS that are banned from using simple world features like openclaw, for ex. people of iran...
- fix(reef): bound inbox WebSocket frames — Follow-up to #106456 What Problem This Solves Fixes an issue where Reef users could have an oversized relay WebSocket frame materialized in the gateway process because the inbox connection had no...
- feat(android): add staged Wear status bridge — Related: #47604 What Problem This Solves OpenClaw does not currently have a small, reviewable Wear OS vertical slice that proves the intended phone-first product and security boundary. The existing...
- fix(installer): bound PowerShell download stalls — What Problem This Solves The Windows installer can wait indefinitely when a portable Node or Git metadata/download response connects but stops delivering body data. Its four Invoke-RestMethod /...
- refactor(ci): share Android toolchain setup — What Problem This Solves Android CI and Android release workflows duplicated the same pinned Java, SDK cache, command-line tools, and package setup. The copies had already drifted in shell style and...
- [Bug] SQLite Connection and File Descriptor Leak due to Missing isOpen in Node DatabaseSync — Summary / Problem Statement In src/state/openclaw-state-db.ts, openOpenClawStateDatabase checks if a cached database connection is open using cached?.db.isOpen. However, Node's node:sqlite driver...
- [Bug] Synchronous Throw Lease Leak in Telegram Send Operations — Summary / Problem Statement In extensions/telegram/src/send.ts, the sendMessageTelegram function resolves the API context and leases client options. However, because the inner function argument is...
- fix(agents): load bootstrap files from agentDir alongside workspace — Summary Per-agent agentDir bootstrap files (SOUL.md, AGENTS.md, TOOLS.md, etc.) were silently ignored when agentDir differed from workspaceDir. Loads bootstrap files from agentDir when explicitly...
- feat(gateway): add deterministic dummy model (AI-assisted) — Summary - Add dummy/dummy as an explicit “No AI - deterministic gateway only” model in the interactive default-model picker and support the same value when written directly to openclaw.json. - Return...
- fix(gateway): regenerate native hook descriptors when relay unavailable after restart — Summary Native hook descriptors were not regenerated when the relay was unavailable during gateway restart, causing hooks to silently fail until manual intervention. Falls through to normal OpenClaw...
- fix(ci): bound actionlint, OCM, and plugin-release downloads — What Problem This Solves Fixes an issue where the workflow-sanity actionlint download, the openclaw-performance OCM download, and the plugin-npm-release GitHub API call could remain stuck until their...
- fix(scripts): explain retained pr-operation locks and malformed review artifacts — What Problem This Solves Fixes an issue where operators (and agents) running scripts/pr on machines where a helper process outlives the command saw every run end with "Retaining the operation lock …...
- fix(discord): keep voice allowlists from granting owner authority — 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 where a Discord...
- feat(canvas): interactive widgets can send follow-up prompts in web chat — What Problem This Solves Inline show_widget widgets in web chat were display-only: a widget could render buttons, but clicking them did nothing. There was no supported way for a widget to drive the...
- fix(dispatch-wrapper): fix flock separator and script command detection — Related: #98964 What Problem This Solves 1. flock --: flock /tmp/lock -- echo hello not unwrapped (-- treated as option-like) 2. script: script -- echo hello mislabeled echo as transcript file Why...
- perf(usage): incremental per-session usage rollups in per-agent SQLite — What Problem This Solves sessions.usage re-parses every session transcript in full on each call. On large installs with an actively-appended multi-hundred-MB transcript this takes 15–28 s and...
- fix(ci): harden release validation gates — What Problem This Solves Full release validation exposed deterministic harness failures that obscured the real release signal: - Compatibility planning mixed diagnostics into its stdout TSV contract...
- OpenClaw repo stats for 2026-07-16 — 383100 stars, 80451 forks, 2900 contributors
- OpenClaw sponsors (103 total) — 103 sponsors supporting OpenClaw
- revert: remove Test prepublication repair (#3101)
- fix: repair stale Test prepublication data (#3100)
- fix(security): require ClawScan artifact inspection (#3099)
- feat: make Featured the default ClawHub catalog (#3096) — feat: make home catalog featured-first fix: order plugins before skills on home feat: refine featured catalog landing page fix: seed featured catalog previews * fix: reduce official creator shelf
- fix(security): classify SkillSpector findings exits correctly (#3098)
- feat(security): publish cutover health summaries (#3097) — feat(security): publish scan worker health summaries test(security): cover scan health diagnostics
- feat: add security scan comparison modes (#3095)
- feat(security): route all scan targets through ClawScan (#3094)
- feat(security): run skill-version scans through OSS ClawScan — Implements CLAW-541: an artifact-only OSS ClawScan path at the canonical worker seam while preserving the legacy production default. Includes strict artifact validation, complete secret-safe...
- perf: bound publisher profile reads (#3073)
- Add skill: ohad6k/ditto-profile
- Revise trentclaw section in README — Updated trentclaw description and image in README.
- Add skill: Piccolo123/ai-footprints — AI-powered cross-platform bookmark manager
- 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...
- Google restricting Google AI Pro/Ultra subscribers for using OpenClaw
- I Run 50+ OpenClaw Tasks Every Day on Autopilot — Here's the Cron Architecture That Makes It Possible — Last Tuesday I woke up to 47 completed tasks my OpenClaw agent had run overnight. Newsletter drafted....
- OpenClaw Just Shipped Native Session Scroll-Back and a Settings-Search Hardening Sweep — 10 commits landed across OpenClaw in the last 3 hours. The headline is a real chat feature — native...
- OpenClaw Just Shipped a Batch of Messaging and Request-Policy Fixes — 10 Commits Land — 10 commits landed across OpenClaw in the last 3 hours. This is a correctness-heavy batch — no flashy...
- OpenClaw Just Shipped NUIA Phase 3b and SQLite State Hardeners — 10 commits landed across OpenClaw in the last 3 hours. The batch is heavy on core hardening: NUIA...
- Private AI: Where Your Data Actually Goes (and How to Keep It Home)
- Securely Connecting Your OpenClaw Gateway — A self-hosted AI assistant is only as useful as its accessibility. If you plan to use OpenClaw outside your home or office network, you’ll…Continue reading on Medium »
- I Used MyClaw to Build a Research Assistant That Sends Me Updates on Telegram — Most AI tools wait for prompts. I tested whether MyClaw could run a recurring research workflow and send useful updates before I asked.Continue reading on Level Up Coding »
- ChatSonic — Combining Google's knowledge graph, ChatSonic delivers real-time data, voice interaction, AI-produced visual content, personalized avatars, and API for services. It revisits past dialogs for informed...
- openclaw flagged $WURK early. price is moving exactly as predicted $WURK ┃ mc $136.1K ┃ vol $1.9K ┃ liq $41.0K ┃ tgt $
- openclaw flagged $𝕏Money as a strong hold. not surprised $𝕏Money MC $114.8K → $328.2K (+186%) 24h vol $4.9K Was Ma
- openclaw on $Maputo —momentum strong, risk low. easy $Maputo current → $2.0K target → $5.6K vol 24h → $0
- openclaw on $26 —momentum strong, risk low. easy $26 $2.0K ▰▰▱▱▱▱▱▱▱▱ $10.5K vol $231 · liq $3.5K · +421% Was Made
- $Glove momentum score just came in hot on openclaw 🔥 $Glove current → $2.5K target → $7.3K vol 24h → $238
- openclaw called $BELGIUM bullish. i'm listening $BELGIUM $2.1K ▰▰▰▱▱▱▱▱▱▱ $6.5K vol $5 · liq $3.4K · +209% Was Mad
- openclaw been right on $CANADA twice already. trusting the scan $CANADA ├ mc $1.6K ├ vol $1 ├ liq $3.0K └ tgt $6.2
- openclaw on $ENGLAND —signals all pointing up [SCAN] $ENGLAND [MC] $6.1K [VOL] $77.2K [LIQ] $6.4K [TGT] $91.5
- openclaw on $FRANCE —momentum strong, risk low. easy $FRANCE current → $8.4K target → $43.9K vol 24h → $2.9K
- openclaw flagged $WIZARD early. price is moving exactly as predicted $WIZARD MC $201.9K → $716.8K (+255%) 24h vol $2
- just checked $DoubleT —openclaw showing strong upside signals $DoubleT — quick check • Current MC: $124.6K • 24h Vol
- just ran $ACAT through openclaw. risk is low, upside is real $ACAT $49.5K ▰▰▰▱▱▱▱▱▱▱ $163.5K vol $3.5K · liq $23.7K
- $PIGEON looking like a clean entry according to openclaw $PIGEON ├ mc $24.7K ├ vol $723 ├ liq $24.2K └ tgt $86.7K
- been tracking $VIBE on openclaw for a week. forecast holding strong $VIBE current → $106.3K target → $351.7K
- openclaw on $PEPE —momentum strong, risk low. easy > PEPE MKT CAP $20.8K 24H VOL $20.7K LIQUIDITY $10.
- openclaw on $SOLdiers —signals all pointing up $SOLdiers ┃ mc $13.1K ┃ vol $313.9K ┃ liq $8.7K ┃ tgt $137.6K ┃ +949%
- been tracking $CASHCAT on openclaw for a week. forecast holding strong $CASHCAT ├ mc $81.8K ├ vol $43.6K ├ liq $25.
- openclaw called $ASHCAT perfectly last time. running the scan again $ASHCAT ▁▂▃▅▇ → $79.2K mc $7.7K · vol $60.3K · +9
- just checked $Dinky —openclaw showing strong upside signals $Dinky ▁▂▃▅▇ → $53.7K mc $3.8K · vol $70.4K · +1328% Wa
- my openclaw scan on $juhn just confirmed what i was thinking $juhn — quick check • Current MC: $30.3K • 24h Volume:
Trending on X
- openclaw flagged $WURK early. price is moving exactly as predicted $WURK ┃ mc $136.1K ┃ vol $1.9K ┃ liq $41.0K ┃ tgt $
- openclaw flagged $𝕏Money as a strong hold. not surprised $𝕏Money MC $114.8K → $328.2K (+186%) 24h vol $4.9K Was Ma
- openclaw on $Maputo —momentum strong, risk low. easy $Maputo current → $2.0K target → $5.6K vol 24h → $0
- openclaw on $26 —momentum strong, risk low. easy $26 $2.0K ▰▰▱▱▱▱▱▱▱▱ $10.5K vol $231 · liq $3.5K · +421% Was Made
- $Glove momentum score just came in hot on openclaw 🔥 $Glove current → $2.5K target → $7.3K vol 24h → $238
- openclaw called $BELGIUM bullish. i'm listening $BELGIUM $2.1K ▰▰▰▱▱▱▱▱▱▱ $6.5K vol $5 · liq $3.4K · +209% Was Mad
- openclaw been right on $CANADA twice already. trusting the scan $CANADA ├ mc $1.6K ├ vol $1 ├ liq $3.0K └ tgt $6.2
- openclaw on $ENGLAND —signals all pointing up [SCAN] $ENGLAND [MC] $6.1K [VOL] $77.2K [LIQ] $6.4K [TGT] $91.5
- openclaw on $FRANCE —momentum strong, risk low. easy $FRANCE current → $8.4K target → $43.9K vol 24h → $2.9K
- openclaw flagged $WIZARD early. price is moving exactly as predicted $WIZARD MC $201.9K → $716.8K (+255%) 24h vol $2
- just checked $DoubleT —openclaw showing strong upside signals $DoubleT — quick check • Current MC: $124.6K • 24h Vol
- just ran $ACAT through openclaw. risk is low, upside is real $ACAT $49.5K ▰▰▰▱▱▱▱▱▱▱ $163.5K vol $3.5K · liq $23.7K
- $PIGEON looking like a clean entry according to openclaw $PIGEON ├ mc $24.7K ├ vol $723 ├ liq $24.2K └ tgt $86.7K
- been tracking $VIBE on openclaw for a week. forecast holding strong $VIBE current → $106.3K target → $351.7K
- openclaw on $PEPE —momentum strong, risk low. easy > PEPE MKT CAP $20.8K 24H VOL $20.7K LIQUIDITY $10.
- openclaw on $SOLdiers —signals all pointing up $SOLdiers ┃ mc $13.1K ┃ vol $313.9K ┃ liq $8.7K ┃ tgt $137.6K ┃ +949%
- been tracking $CASHCAT on openclaw for a week. forecast holding strong $CASHCAT ├ mc $81.8K ├ vol $43.6K ├ liq $25.
- openclaw called $ASHCAT perfectly last time. running the scan again $ASHCAT ▁▂▃▅▇ → $79.2K mc $7.7K · vol $60.3K · +9
- just checked $Dinky —openclaw showing strong upside signals $Dinky ▁▂▃▅▇ → $53.7K mc $3.8K · vol $70.4K · +1328% Wa
- my openclaw scan on $juhn just confirmed what i was thinking $juhn — quick check • Current MC: $30.3K • 24h Volume:
Releases
- openclaw 2026.7.2-beta.1 — 2026.7.2 Highlights - Remote coding sessions: run Control UI sessions on cloud workers, open Codex and Claude catalog sessions in terminals on their owning hosts, and resume OpenCode and Pi sessions...
- OpenClaw repo stats for 2026-07-16 — 383100 stars, 80451 forks, 2900 contributors
Community
- [Bug] Infinite DFS Recursion Crash on Circular Plugin Peer Linkage — Summary / Problem Statement In src/plugins/plugin-peer-link.ts, resolving plugin dependency peer links utilizes a Depth-First Search (DFS) traversal. If a circular linkage exists (e.g. Plugin A peers...
- perf: avoid event-loop stall during embedded_run bootstrap-context — Summary Fix event-loop stalls during embedded_run bootstrap-context that block the event loop for 14-22 seconds, causing message loss. What Problem This Solves Event-loop stalls during embedded_run...
- feat(memory-lancedb): scope memory_recall and memory_forget + configurable thresholds — Supersedes #102966 (closed: re-opened cleanly on current main; the bot review state there was stale and could not be refreshed). What Problem This Solves In a shared LanceDB memory store the memory...
- fix(agents): treat read-only shell inspection as replay-safe — What Problem This Solves Fixes #101890 — strict-agentic non-visible retry was blocked after turns that only ran proven read-only shell inspection (exec/bash with commands such as rg, ls, cat). Those...
- [Bug] Unhandled Promise Rejection in channel-server.ts on Connection Reset — Summary / Problem Statement In src/mcp/channel-server.ts, writing frame payloads to active client connections does not wrap write promises in local try-catch blocks. If a client abruptly resets the...
- feat: let plugins control inbound debounce policy — Related: #54565 What Problem This Solves Plugins cannot currently decide whether rich inbound WhatsApp messages should bypass debounce or choose a debounce window for a specific account, DM, group,...
- fix(hooks): wire message:sending internal hook to user-defined hooks — Summary The message:sending internal hook was registered but never invoked during outbound message delivery. Wires the hook into deliver.ts so user-defined hooks receive outgoing message events for...
- fix(codex): prevent corrupted UTF-8 in sandbox exec server output — What Problem This Solves Fixes an issue where the sandbox HTTP request handler (extensions/codex/src/app-server/sandbox-exec-server/http.ts) could receive corrupted stdout or stderr when a multibyte...
- fix(ui): explain replace-mode filtering in model picker — Closes #104545 What Problem This Solves When models.mode: "replace" filters the model catalog, the Control UI showed the resulting models without explaining why other bundled or discovered models...
- fix(agents): preserve ANSI sanitizer state across bash chunks — Closes #103668 What Problem This Solves Fixes an issue where users running bash commands through agents could see ANSI/OSC control-sequence fragments leak into streamed tool output, final bash...
- feat(mac): mascot accessory pack — nightcap, grad cap, hat-tip exit — What Problem This Solves The mascot's new working mood proved the pose-channel system extends well (#108735), but headwear was hardwired to one hat and one mood. Sleepy Clawd sleeps bare-headed,...
- fix(config/sessions): avoid splitting surrogate pairs when shortening group ids — What Problem This Solves buildGroupDisplayName shortens very long group ids via shortenGroupId, which previously used raw String.prototype.slice to take the first 6 and last 4 UTF-16 code units. When...
- [Bug]: VPN NEEDED users — Bug type Crash (process/app exits or hangs) Beta release blocker No Summary there is some users some HUMANS that are banned from using simple world features like openclaw, for ex. people of iran...
- fix(reef): bound inbox WebSocket frames — Follow-up to #106456 What Problem This Solves Fixes an issue where Reef users could have an oversized relay WebSocket frame materialized in the gateway process because the inbox connection had no...
- feat(android): add staged Wear status bridge — Related: #47604 What Problem This Solves OpenClaw does not currently have a small, reviewable Wear OS vertical slice that proves the intended phone-first product and security boundary. The existing...
- fix(installer): bound PowerShell download stalls — What Problem This Solves The Windows installer can wait indefinitely when a portable Node or Git metadata/download response connects but stops delivering body data. Its four Invoke-RestMethod /...
- refactor(ci): share Android toolchain setup — What Problem This Solves Android CI and Android release workflows duplicated the same pinned Java, SDK cache, command-line tools, and package setup. The copies had already drifted in shell style and...
- [Bug] SQLite Connection and File Descriptor Leak due to Missing isOpen in Node DatabaseSync — Summary / Problem Statement In src/state/openclaw-state-db.ts, openOpenClawStateDatabase checks if a cached database connection is open using cached?.db.isOpen. However, Node's node:sqlite driver...
- [Bug] Synchronous Throw Lease Leak in Telegram Send Operations — Summary / Problem Statement In extensions/telegram/src/send.ts, the sendMessageTelegram function resolves the API context and leases client options. However, because the inner function argument is...
- fix(agents): load bootstrap files from agentDir alongside workspace — Summary Per-agent agentDir bootstrap files (SOUL.md, AGENTS.md, TOOLS.md, etc.) were silently ignored when agentDir differed from workspaceDir. Loads bootstrap files from agentDir when explicitly...
- feat(gateway): add deterministic dummy model (AI-assisted) — Summary - Add dummy/dummy as an explicit “No AI - deterministic gateway only” model in the interactive default-model picker and support the same value when written directly to openclaw.json. - Return...
- fix(gateway): regenerate native hook descriptors when relay unavailable after restart — Summary Native hook descriptors were not regenerated when the relay was unavailable during gateway restart, causing hooks to silently fail until manual intervention. Falls through to normal OpenClaw...
- fix(ci): bound actionlint, OCM, and plugin-release downloads — What Problem This Solves Fixes an issue where the workflow-sanity actionlint download, the openclaw-performance OCM download, and the plugin-npm-release GitHub API call could remain stuck until their...
- fix(scripts): explain retained pr-operation locks and malformed review artifacts — What Problem This Solves Fixes an issue where operators (and agents) running scripts/pr on machines where a helper process outlives the command saw every run end with "Retaining the operation lock …...
- fix(discord): keep voice allowlists from granting owner authority — 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 where a Discord...
- feat(canvas): interactive widgets can send follow-up prompts in web chat — What Problem This Solves Inline show_widget widgets in web chat were display-only: a widget could render buttons, but clicking them did nothing. There was no supported way for a widget to drive the...
- fix(dispatch-wrapper): fix flock separator and script command detection — Related: #98964 What Problem This Solves 1. flock --: flock /tmp/lock -- echo hello not unwrapped (-- treated as option-like) 2. script: script -- echo hello mislabeled echo as transcript file Why...
- perf(usage): incremental per-session usage rollups in per-agent SQLite — What Problem This Solves sessions.usage re-parses every session transcript in full on each call. On large installs with an actively-appended multi-hundred-MB transcript this takes 15–28 s and...
- fix(ci): harden release validation gates — What Problem This Solves Full release validation exposed deterministic harness failures that obscured the real release signal: - Compatibility planning mixed diagnostics into its stdout TSV contract...
- OpenClaw sponsors (103 total) — 103 sponsors supporting OpenClaw
- revert: remove Test prepublication repair (#3101)
- fix: repair stale Test prepublication data (#3100)
- fix(security): require ClawScan artifact inspection (#3099)
- feat: make Featured the default ClawHub catalog (#3096) — feat: make home catalog featured-first fix: order plugins before skills on home feat: refine featured catalog landing page fix: seed featured catalog previews * fix: reduce official creator shelf
- fix(security): classify SkillSpector findings exits correctly (#3098)
- feat(security): publish cutover health summaries (#3097) — feat(security): publish scan worker health summaries test(security): cover scan health diagnostics
- feat: add security scan comparison modes (#3095)
- feat(security): route all scan targets through ClawScan (#3094)
- feat(security): run skill-version scans through OSS ClawScan — Implements CLAW-541: an artifact-only OSS ClawScan path at the canonical worker seam while preserving the legacy production default. Includes strict artifact validation, complete secret-safe...
- perf: bound publisher profile reads (#3073)
- Add skill: ohad6k/ditto-profile
- Revise trentclaw section in README — Updated trentclaw description and image in README.
- Add skill: Piccolo123/ai-footprints — AI-powered cross-platform bookmark manager
News
- 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...
- Google restricting Google AI Pro/Ultra subscribers for using OpenClaw
- I Run 50+ OpenClaw Tasks Every Day on Autopilot — Here's the Cron Architecture That Makes It Possible — Last Tuesday I woke up to 47 completed tasks my OpenClaw agent had run overnight. Newsletter drafted....
- OpenClaw Just Shipped Native Session Scroll-Back and a Settings-Search Hardening Sweep — 10 commits landed across OpenClaw in the last 3 hours. The headline is a real chat feature — native...
- OpenClaw Just Shipped a Batch of Messaging and Request-Policy Fixes — 10 Commits Land — 10 commits landed across OpenClaw in the last 3 hours. This is a correctness-heavy batch — no flashy...
- OpenClaw Just Shipped NUIA Phase 3b and SQLite State Hardeners — 10 commits landed across OpenClaw in the last 3 hours. The batch is heavy on core hardening: NUIA...
- Private AI: Where Your Data Actually Goes (and How to Keep It Home)
- Securely Connecting Your OpenClaw Gateway — A self-hosted AI assistant is only as useful as its accessibility. If you plan to use OpenClaw outside your home or office network, you’ll…Continue reading on Medium »
- I Used MyClaw to Build a Research Assistant That Sends Me Updates on Telegram — Most AI tools wait for prompts. I tested whether MyClaw could run a recurring research workflow and send useful updates before I asked.Continue reading on Level Up Coding »
- ChatSonic — Combining Google's knowledge graph, ChatSonic delivers real-time data, voice interaction, AI-produced visual content, personalized avatars, and API for services. It revisits past dialogs for informed...
Don't miss what's next. Subscribe to Openclaw Newsletter: