# openclaw changelog

> Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

- Vendor: openclaw
- Category: AI
- Official site: https://openclaw.ai
- Tracked by: What's New (https://whatsnew.fyi/product/openclaw)
- Harvested from: GitHub (openclaw/openclaw)
- Entries below: 11 (newest first)

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

## Releases

### v2026.7.2-beta.5 — openclaw 2026.7.2-beta.5

- Date: 2026-07-28
- Version: v2026.7.2-beta.5
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.2-beta.5
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.7.2-beta.5
- Labels: Pre-release

- **added** — Protect persisted data with a quarantine store that survives primary-database damage, crash-recoverable SQLite snapshots, crash-durable filesystem publication, and rollback-writer snapshot recovery
- **added** — Keep accepted messages recoverable across gateway restarts and local crashes through shared ingress drain and dead-letter recovery for Telegram, Signal, Slack, QQBot, Twitch, Synology Chat, Tlon, IRC, and Zalo User
- **added** — Rewind or fork conversations from individual messages and switch transcript branches across web and native apps
- **added** — Host ticketed MCP Apps with bound tools, resources, and bounded context updates, open them from channel replies, and pin them to durable dashboards
- **added** — Let agents ask structured questions with option cards across web, channels, macOS, and native apps
- **added** — Add approvals with push notifications, history, fair queuing, headless resolution, Claude tool-request relay, and reviewer detail
- **added** — Join Teams, Zoom, and Google Meet calls with default-enabled meeting plugins and durable transcript collection
- **added** — Add realtime Talk with OpenAI and Gemini video plus GPT Live through Codex OAuth
- **added** — Add phone-proxied Wear OS companion with home-screen agent/session/model selection, realtime Talk controls, and audio-reactive playback
- **added** — Guide setup across browser, Linux, and macOS with local-provider detection, strongest-model selection, and downloadable models
- **added** — Add in-process RAM-gated llama.cpp/Gemma inference path for local inference
- **added** — Add Claude Opus 5 across catalog and runtime
- **added** — Add Kimi K3 model support
- **added** — Detect local inference providers during onboarding and discover models from live provider catalogs
- **added** — Add in-process llama.cpp GGUF inference support
- **added** — Add Baseten Model API support
- **added** — Add session boards, dashboards, and archived/visibility/draft/incognito session states
- **added** — Bring Quick Chat to macOS and Linux with streaming, routing, context capture, dictation, and model controls
- **added** — Add multi-gateway apps and mobile dashboards
- **added** — Add the Buzz plugin for channels
- **added** — Add fast active-memory recall and default cross-conversation recall for personal installs

##### 2026.7.2

###### Highlights

- **State safety and recovery:** protect persisted data with a quarantine store that survives primary-database damage, crash-recoverable SQLite snapshots, crash-durable filesystem publication, schema-upgrade data-loss rejection, and rollback-writer snapshot recovery. (#110453, #113367, #113453, #113473, #113580) Thanks @vincentkoc.
- **Durable channel delivery:** keep accepted messages recoverable across gateway restarts and local crashes through the shared ingress drain and dead-letter recovery, covering Telegram, Signal, Slack, QQBot, Twitch, Synology Chat, Tlon, IRC, and Zalo User. #108656, #107246, #109911 (#108924, #107288, #109907, #109910, #110844, #110852, #110899, #110910, #110914, #110916, #111029) Thanks @obviyus and @edenfunf.
- **Session rewind and branching:** rewind or fork conversations from individual messages, switch transcript branches across web and native apps, fork upstream Codex sessions, preserve branch-safe queued sends, reject stale-pane writes, and restore prompt images after a fork. (#110660, #110857, #110886, #111149, #112056, #112284, #113073, #113945)
- **Interactive MCP Apps and dashboards:** host ticketed MCP Apps with bound tools, resources, and bounded context updates; open them from channel replies, pin them to durable dashboards, harden their shared sandbox, and let native plugins declare them directly. #109851, #110451, #113218 (#109861, #109807, #110515, #111211, #111212, #111524, #111687, #111748, #113224) Thanks @fuller-stack-dev.
- **Questions and approvals everywhere:** let agents ask structured questions with option cards across web, channels, macOS, and native apps, while approvals gain push notifications, history, fair queuing, headless resolution, Claude tool-request relay, reviewer detail, and clearer formatted prompts. #85954 (#108505, #108709, #108776, #109922, #110242, #110372, #110584, #110681, #110989, #111060, #112918, #113027, #113193) Thanks @omarshahine.
- **Meetings and realtime Talk:** join Teams, Zoom, and Google Meet calls with default-enabled meeting plugins and durable transcript collection, while realtime Talk adds OpenAI and Gemini video plus GPT Live through Codex OAuth. #86425, #113353 (#109579, #109719, #109964, #111048, #113022, #113053, #113122, #113354) Thanks @shushushv and @Solvely-Colin.
- **Wear OS companion:** add the phone-proxied Wear companion with home-screen agent/session/model selection, realtime Talk controls, audio-reactive playback, and an instant-talk tile. #108781 (#108835, #109341, #109433, #109483, #110661, #111516, #112721) Thanks @sibbl, @IWhatsskill, and @Solvely-Colin.
- **Guided setup and local inference:** guide setup across browser, Linux, and macOS with local-provider detection, strongest-model selection, downloadable models, lean mode, memory imports, and an in-process RAM-gated llama.cpp/Gemma path. #108604 (#108605, #108868, #108977, #109250, #109444, #109585, #110054, #110141, #110596, #113476)

###### Changes

- **Models and providers:** add Claude Opus 5 across catalog and runtime, Kimi K3, and GPT Live through Codex OAuth. (#113391, #113392, #113633; related #113412; #113909, #113354; related #113353) Thanks @fuller-stack-dev, @vincentkoc, and @Solvely-Colin.
- **Local inference and setup:** detect local inference providers during onboarding, add in-process llama.cpp GGUF inference and Baseten Model API support, discover models from live provider catalogs, and offer model downloads from web and macOS setup. (#108605; related #108604; #109444, #108708; related #108665; #112412; related #112405; #113476) Thanks @fuller-stack-dev.
- **Sessions and dashboards:** add rewind/fork and branch switching across web and native chat, session boards and dashboards, archived/visibility/draft/incognito session states, and suggestion queues with typing indicators. (#110660, #110857, #110886, #111149, #110644, #110960, #112554, #112787, #113006, #113127, #113173)
- **Native apps:** bring Quick Chat to macOS

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.2-beta.5]_

### v2026.7.2-beta.3 — openclaw 2026.7.2-beta.3

- Date: 2026-07-18
- Version: v2026.7.2-beta.3
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.2-beta.3
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.7.2-beta.3
- Labels: Pre-release

- **added** — Remote coding sessions that 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 directly in a terminal
- **added** — Native automation and nodes that bring Automations parity to mobile, add foreground Voice Wake on Android, and expose camera, location, and notification capabilities from headless Linux nodes
- **added** — Linux deb and AppImage bundles with Gateway guidance, published from stable main-based releases
- **added** — External gateway supervision with OPENCLAW_SUPERVISOR_MODE=external for lifecycle owners such as OCM
- **added** — ClickClack guided setup that configures ClickClack from openclaw onboard or openclaw channels add clickclack with URL, token, and workspace prompts
- **added** — ClickClack command menus that publish each bot's native OpenClaw commands to ClickClack composer autocomplete at gateway startup
- **added** — Skill Workshop history review that performs a manual, newest-first session scan for skill ideas and stores SQLite cursor metadata
- **added** — iOS offline chat that pre-paints recent sessions and canonical transcripts from a protected, bounded per-gateway cache
- **added** — Control UI Talk controls that keep voice, model, sensitivity, and other realtime defaults in Settings → Communications → Talk
- **added** — Control UI session workspace shortcut to expand or collapse the active Chat pane's session workspace rail with ⇧⌘B
- **added** — Control UI Settings shortcut to open Settings with ⇧⌘
- **added** — Control UI session titles with truncated recent-session names revealed through a reduced-motion-safe hover animation
- **changed** — Skill Workshop approvals to run agent-initiated apply, reject, and quarantine actions without an additional approval prompt by default
- **changed** — TUI fuzzy selectors to delegate list matching to pi-tui, adding slash-token and alpha-number matching
- **changed** — macOS paired-node terminals to advertise duplex Codex and Claude terminal resume commands from the embedded node host
- **changed** — Control UI catalog terminals to open eligible Codex and Claude Code sessions in the native CLI on their Gateway or paired-node host
- **changed** — OpenAI GPT-5.6 defaults to use openai/gpt-5.6 for fresh API-key setup and exact openai/gpt-5.6-sol for fresh Codex/OAuth setup
- **changed** — Slack progress indicators to use Slack's native assistant thread status and rotating loading messages by default
- **changed** — Control UI chat layout to center the transcript on the composer axis with assistant and tool output left and user bubbles right
- **fixed** — Prevent Telegram durable-ingress loss after restarts
- **fixed** — Keep Signal stop and approval controls responsive during active turns

##### 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 directly in a terminal. (#107670, #107086, #107200)
- **Native automation and nodes:** bring Automations parity to mobile, add foreground Voice Wake on Android, and expose camera, location, and notification capabilities from headless Linux nodes. (#106355, #107081, #107193)
- **Safer channel operation:** prevent Telegram durable-ingress loss after restarts, keep Signal stop and approval controls responsive during active turns, and stop channel allowlists from granting owner access. (#107288, #107422, #107403) Thanks @obviyus, @arduano, and @yetval.
- **Guided Control UI setup:** configure model providers from Settings, onboard channels through a guided setup page, and choose images and models while creating sessions. (#106490, #106469, #107358) Thanks @alexandre-leng and @fuller-stack-dev.
- **Gateway and session recovery:** prevent restart admission from wedging the Gateway, recover reply sessions after finalization stalls, and keep one-shot cron jobs enabled through lifecycle claim races. (#107339, #106792, #107236) Thanks @obviyus, @joshavant, @charliemeyer2000, and @SL4N.
- **Install and packaging:** add Linux deb and AppImage bundles with Gateway guidance, publish them from stable main-based releases, and let Windows installs continue immediately after winget adds Node.js. (#106533, #106891, #106862)

###### Changes

- **External gateway supervision:** add `OPENCLAW_SUPERVISOR_MODE=external` for lifecycle owners such as OCM, preserving verified restart and deferral behavior without exposing native service authority, blocking native service mutation and self-update, and providing a versioned atomic restart-handoff consume contract. Thanks @shakkernerd.
- **ClickClack guided setup:** configure ClickClack from `openclaw onboard` or `openclaw channels add clickclack` with URL, token, and workspace prompts, default-account env fallback, nonfatal live connection validation, and gateway-aware next steps that connect automatically when OpenClaw is already running. Thanks @shakkernerd.
- **ClickClack command menus:** publish each bot's native OpenClaw commands to ClickClack composer autocomplete at gateway startup, with per-account opt-out and nonfatal compatibility handling for older tokens and servers. Thanks @shakkernerd and @vincentkoc.
- **Skill Workshop approvals:** run agent-initiated apply, reject, and quarantine actions without an additional approval prompt by default while preserving `skills.workshop.approvalPolicy: "pending"` as an opt-in approval gate. Thanks @shakkernerd.
- **TUI fuzzy selectors:** delegate list matching to pi-tui, adding slash-token and alpha-number matching while removing the local matcher fork.
- **macOS paired-node terminals:** advertise duplex Codex and Claude terminal resume commands from the embedded node host and forward interactive input and cancellation through the native app bridge. (#107335)
- **Control UI catalog terminals:** open eligible Codex and Claude Code sessions in the native CLI on their Gateway or paired-node host, with viewer-versus-terminal preferences, validated resume commands, and an interactive PTY relay. (#107086) Thanks @vincentkoc.
- **Skill Workshop history review:** add a manual, newest-first session scan that progressively searches older substantial work for conservative skill ideas, stores only SQLite cursor metadata, and leaves up to three results as pending proposals even when autonomous self-learning is disabled. (#106182)
- **OpenAI GPT-5.6 defaults:** use `openai/gpt-5.6` (Sol alias) for fresh API-key setup and exact `openai/gpt-5.6-sol` for fresh Codex/OAuth setup, default Sol to medium reasoning across both runtimes, and preserve existing primaries, fallbacks, aliases, and explicit GPT-5.5 selections. (#103234)
- **iOS offline chat:** pre-paint

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.2-beta.3]_

### v2026.7.2-beta.2 — openclaw 2026.7.2-beta.2

- Date: 2026-07-17
- Version: v2026.7.2-beta.2
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.2-beta.2
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.7.2-beta.2
- Labels: Pre-release

- **added** — Add remote coding sessions to 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 directly in a terminal
- **added** — Add native automation and nodes to bring Automations parity to mobile, add foreground Voice Wake on Android, and expose camera, location, and notification capabilities from headless Linux nodes
- **added** — Add OPENCLAW_SUPERVISOR_MODE=external for lifecycle owners such as OCM, preserving verified restart and deferral behavior without exposing native service authority
- **added** — Add ClickClack guided setup to configure ClickClack from openclaw onboard or openclaw channels add clickclack with URL, token, and workspace prompts
- **added** — Add ClickClack command menus to publish each bot's native OpenClaw commands to ClickClack composer autocomplete at gateway startup
- **added** — Add Skill Workshop history review with a manual, newest-first session scan that progressively searches older substantial work for conservative skill ideas
- **added** — Add iOS offline chat to pre-paint recent sessions and canonical transcripts from a protected, bounded per-gateway cache
- **added** — Add Control UI Talk controls to keep voice, model, sensitivity, and other realtime defaults in Settings → Communications → Talk
- **added** — Add Control UI session workspace shortcut to expand or collapse the active Chat pane's session workspace rail with ⇧⌘B
- **added** — Add Control UI Settings shortcut to open Settings with ⇧⌘
- **fixed** — Prevent Telegram durable-ingress loss after restarts
- **fixed** — Keep Signal stop and approval controls responsive during active turns
- **fixed** — Stop channel allowlists from granting owner access
- **fixed** — Prevent restart admission from wedging the Gateway
- **fixed** — Recover reply sessions after finalization stalls
- **fixed** — Keep one-shot cron jobs enabled through lifecycle claim races
- **changed** — Delegate TUI fuzzy selector list matching to pi-tui, adding slash-token and alpha-number matching while removing the local matcher fork
- **changed** — Advertise duplex Codex and Claude terminal resume commands from the embedded node host and forward interactive input and cancellation through the native app bridge on macOS paired-node terminals
- **changed** — Use openai/gpt-5.6 for fresh API-key setup and exact openai/gpt-5.6-sol for fresh Codex/OAuth setup, with default Sol to medium reasoning across both runtimes

##### 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 directly in a terminal. (#107670, #107086, #107200)
- **Native automation and nodes:** bring Automations parity to mobile, add foreground Voice Wake on Android, and expose camera, location, and notification capabilities from headless Linux nodes. (#106355, #107081, #107193)
- **Safer channel operation:** prevent Telegram durable-ingress loss after restarts, keep Signal stop and approval controls responsive during active turns, and stop channel allowlists from granting owner access. (#107288, #107422, #107403) Thanks @obviyus, @arduano, and @yetval.
- **Guided Control UI setup:** configure model providers from Settings, onboard channels through a guided setup page, and choose images and models while creating sessions. (#106490, #106469, #107358) Thanks @alexandre-leng and @fuller-stack-dev.
- **Gateway and session recovery:** prevent restart admission from wedging the Gateway, recover reply sessions after finalization stalls, and keep one-shot cron jobs enabled through lifecycle claim races. (#107339, #106792, #107236) Thanks @obviyus, @joshavant, @charliemeyer2000, and @SL4N.
- **Install and packaging:** add Linux deb and AppImage bundles with Gateway guidance, publish them from stable main-based releases, and let Windows installs continue immediately after winget adds Node.js. (#106533, #106891, #106862)

###### Changes

- **External gateway supervision:** add `OPENCLAW_SUPERVISOR_MODE=external` for lifecycle owners such as OCM, preserving verified restart and deferral behavior without exposing native service authority, blocking native service mutation and self-update, and providing a versioned atomic restart-handoff consume contract. Thanks @shakkernerd.
- **ClickClack guided setup:** configure ClickClack from `openclaw onboard` or `openclaw channels add clickclack` with URL, token, and workspace prompts, default-account env fallback, nonfatal live connection validation, and gateway-aware next steps that connect automatically when OpenClaw is already running. Thanks @shakkernerd.
- **ClickClack command menus:** publish each bot's native OpenClaw commands to ClickClack composer autocomplete at gateway startup, with per-account opt-out and nonfatal compatibility handling for older tokens and servers. Thanks @shakkernerd and @vincentkoc.
- **Skill Workshop approvals:** run agent-initiated apply, reject, and quarantine actions without an additional approval prompt by default while preserving `skills.workshop.approvalPolicy: "pending"` as an opt-in approval gate. Thanks @shakkernerd.
- **TUI fuzzy selectors:** delegate list matching to pi-tui, adding slash-token and alpha-number matching while removing the local matcher fork.
- **macOS paired-node terminals:** advertise duplex Codex and Claude terminal resume commands from the embedded node host and forward interactive input and cancellation through the native app bridge. (#107335)
- **Control UI catalog terminals:** open eligible Codex and Claude Code sessions in the native CLI on their Gateway or paired-node host, with viewer-versus-terminal preferences, validated resume commands, and an interactive PTY relay. (#107086) Thanks @vincentkoc.
- **Skill Workshop history review:** add a manual, newest-first session scan that progressively searches older substantial work for conservative skill ideas, stores only SQLite cursor metadata, and leaves up to three results as pending proposals even when autonomous self-learning is disabled. (#106182)
- **OpenAI GPT-5.6 defaults:** use `openai/gpt-5.6` (Sol alias) for fresh API-key setup and exact `openai/gpt-5.6-sol` for fresh Codex/OAuth setup, default Sol to medium reasoning across both runtimes, and preserve existing primaries, fallbacks, aliases, and explicit GPT-5.5 selections. (#103234)
- **iOS offline chat:** pre-paint

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.2-beta.2]_

### v2026.7.2-beta.1 — openclaw 2026.7.2-beta.1

- Date: 2026-07-15
- Version: v2026.7.2-beta.1
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.2-beta.1
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.7.2-beta.1
- Labels: Pre-release

- **added** — 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 directly in a terminal
- **added** — Bring Automations parity to mobile, add foreground Voice Wake on Android, and expose camera, location, and notification capabilities from headless Linux nodes
- **added** — Add session placement, dispatch, and worker-turn routing for remote session execution
- **added** — Discover OpenCode and Pi sessions and continue Codex and Claude catalog sessions through streaming CLI agent runs
- **added** — Create eligible catalog sessions directly from the Control UI sidebar
- **added** — Configure cleanup limits by count and total size from Settings for managed worktrees
- **added** — Serve scheduled-run history from the task ledger
- **added** — Import Codex and Claude Code memory into the Control UI
- **added** — Notify agents when voice-channel participants change in Discord voice
- **added** — Show the signed-in account email alongside app-server usage windows for Codex
- **added** — Scan prior session history for conservative, reviewable skill ideas in Skill Workshop
- **added** — Show the latest tasks from the running-tasks status row as task previews
- **added** — Show the active branch and local changed-file state in Control UI sessions
- **added** — Configure model providers from Settings, onboard channels through a guided setup page, and choose images and models while creating sessions
- **added** — Add Linux deb and AppImage bundles with Gateway guidance and publish them from stable main-based releases
- **changed** — Reserve progress drafts for long-running work and use the model's own preamble as the status headline for channel progress
- **changed** — Bump the bundled plugin to Codex CLI 0.144.4
- **fixed** — Prevent Telegram durable-ingress loss after restarts
- **fixed** — Keep Signal stop and approval controls responsive during active turns
- **fixed** — Stop channel allowlists from granting owner access
- **fixed** — Prevent restart admission from wedging the Gateway

##### 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 directly in a terminal. (#107670, #107086, #107200)
- **Native automation and nodes:** bring Automations parity to mobile, add foreground Voice Wake on Android, and expose camera, location, and notification capabilities from headless Linux nodes. (#106355, #107081, #107193)
- **Safer channel operation:** prevent Telegram durable-ingress loss after restarts, keep Signal stop and approval controls responsive during active turns, and stop channel allowlists from granting owner access. (#107288, #107422, #107403) Thanks @obviyus, @arduano, and @yetval.
- **Guided Control UI setup:** configure model providers from Settings, onboard channels through a guided setup page, and choose images and models while creating sessions. (#106490, #106469, #107358) Thanks @alexandre-leng and @fuller-stack-dev.
- **Gateway and session recovery:** prevent restart admission from wedging the Gateway, recover reply sessions after finalization stalls, and keep one-shot cron jobs enabled through lifecycle claim races. (#107339, #106792, #107236) Thanks @obviyus, @joshavant, @charliemeyer2000, and @SL4N.
- **Install and packaging:** add Linux deb and AppImage bundles with Gateway guidance, publish them from stable main-based releases, and let Windows installs continue immediately after winget adds Node.js. (#106533, #106891, #106862)

###### Changes

- **Cloud workers:** add session placement, dispatch, and worker-turn routing for remote session execution. (#106332)
- **Paired-node coding agents:** discover OpenCode and Pi sessions and continue Codex and Claude catalog sessions through streaming CLI agent runs. (#106941, #106927, #105833)
- **Catalog sessions:** create eligible catalog sessions directly from the Control UI sidebar. (#105810) Thanks @fuller-stack-dev.
- **Managed worktrees:** configure cleanup limits by count and total size from Settings. (#106224)
- **Cron history:** serve scheduled-run history from the task ledger. (#106392)
- **Coding-agent memory:** import Codex and Claude Code memory into the Control UI. (#106406)
- **MCP isolation:** scope MCP server connections to their requesting session. (#106359) Thanks @obviyus.
- **Discord voice:** notify agents when voice-channel participants change. (#107004)
- **Codex usage:** show the signed-in account email alongside app-server usage windows. (#106500)
- **Skill Workshop:** scan prior session history for conservative, reviewable skill ideas. (#106766)
- **Task previews:** show the latest tasks from the running-tasks status row. (#107297)
- **Session changes:** show the active branch and local changed-file state in Control UI sessions. (#106835)
- **Channel progress:** reserve progress drafts for long-running work and use the model's own preamble as the status headline. (#106026)
- **Codex CLI:** bump the bundled plugin to Codex CLI 0.144.4.

###### Fixes

- **Terminal sessions:** preserve early keystrokes, enable paired-node terminal actions, authorize paired-node catalog reads, recover after stale Control UI connections, and keep slow clients attached under heavy output. (#107214, #107361, #107410, #107419, #107348) Thanks @vincentkoc and @zw-xysk.
- **Control UI configuration:** auto-save validated edits, restore autosave gates, show unapplied-restart state, and keep normal content scrolling and selection behavior. (#107458, #107477, #107577, #107349)
- **Signal recovery:** reconnect after stalled container handshakes, prevent overlapping daemons during restart, and preserve username targets during phone normalization. (#107386, #107409, #107365) Thanks @hugenshen, @ZHOUKAILIAN, @UberKitten, and @ly-wang19.
- **Authorization boundaries:** require operator-admin authority for paired-node directory browsing and block prototype pollution in migration config merges. (#106004, #

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.2-beta.1]_

### v2026.7.1 — openclaw 2026.7.1

- Date: 2026-07-13
- Version: v2026.7.1
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.1
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.7.1

- **changed** — Control UI overhaul makes conversations easier to organize and work with side by side, with live Tasks, clearer chat controls, better usage and cost views, files, downloads, pairing, approvals, and Gateway health kept close to the conversation
- **changed** — Guided setup provides clearer steps toward a working first chat, checks connections before saving them, and preserves earlier choices when setup is interrupted
- **changed** — iOS and iPadOS, Android, and macOS apps received substantial updates across setup, navigation, chat, voice, permissions, localization, files, scheduled work, offline reading, queued sends, and connection recovery
- **added** — GPT-5.6 compatibility improves across supported OpenAI and Codex routes
- **added** — Tencent Hy3 gains a complete setup path
- **added** — Meta Model API adds Muse Spark 1.1
- **added** — openclaw attach gives Claude Code temporary access to a selected session
- **changed** — Codex delegation and native subagents return tracked results more reliably
- **changed** — Copilot gains broader provider choices
- **changed** — Long-running sessions and goals are easier to resume
- **changed** — Telegram receives substantial work on live progress, photos and documents, topics, commands, retries, account routing, setup, and delivery
- **changed** — Slack threads, cards, progress, identity, reactions, and duplicate prevention improve, while longer conversations avoid more unnecessary waits
- **changed** — Discord replies, attachments, voice sessions, progress, reconnects, and multi-account behavior improve with better unread cues and recovery from repeated session-resume failures
- **changed** — Apple Messages replies, typing, media, routing, setup guidance, and chat continuity improve across supported bridge setups
- **fixed** — Repeatedly failing Gateways now leave a stable repair path instead of restarting indefinitely
- **added** — Scheduled work can wake only when something changes
- **added** — Selected signed-in browser tabs can pair remotely and save completed downloads safely
- **added** — Guarded terminals are available across web, iOS, and Android
- **changed** — Session groups, titles, unread state, checkpoints, context handling, goals, and recovery paths are more consistent across web, mobile, CLI, and connected agents

##### 2026.7.1

OpenClaw v2026.7.1 brings major Control UI and onboarding overhauls, major updates to the official iOS, Android, and macOS apps, expanded model and provider support including GPT-5.6 compatibility, Tencent Hy3, and Meta Muse Spark 1.1, and stronger Codex and connected coding-agent workflows. Telegram, Slack, Discord, and Apple Messages each receive substantial updates, while Gateway crash loops, scheduled work, remote browser control, workspace terminals, sessions, and goals also improve. There are also many general fixes and refinements throughout OpenClaw.

[Read the full release notes](https://docs.openclaw.ai/releases/2026.7.1)

This release brings together 3,063 contributions from 532 contributors.

###### Highlights

- **Control UI overhaul:** conversations are easier to organize and work with side by side, with live Tasks, clearer chat controls, better usage and cost views, files, downloads, pairing, approvals, and Gateway health kept close to the conversation.
- **Easier setup from install to first chat:** guided setup provides clearer steps toward a working first chat, checks connections before saving them, and preserves earlier choices when setup is interrupted.
- **Official app updates:** iOS and iPadOS, Android, and macOS received substantial work across setup, navigation, chat, voice, permissions, localization, files, scheduled work, offline reading, queued sends, connection recovery, and native session controls.
- **Models and providers:** GPT-5.6 compatibility improves across supported OpenAI and Codex routes, Tencent Hy3 gains a complete setup path, Meta Model API adds Muse Spark 1.1, and broader Claude, Ollama, ClawRouter, LongCat, and provider work expands choice and reliability.
- **Codex and connected coding agents:** `openclaw attach` gives Claude Code temporary access to a selected session, Codex delegation and native subagents return tracked results more reliably, Copilot gains broader provider choices, and long-running sessions and goals are easier to resume.
- **Telegram:** live progress, photos and documents, topics, commands, retries, account routing, setup, and delivery all received substantial work.
- **Slack:** threads, cards, progress, identity, reactions, and duplicate prevention improve, while longer conversations avoid more unnecessary waits.
- **Discord:** replies, attachments, voice sessions, progress, reconnects, and multi-account behavior improve, with better unread cues and recovery from repeated session-resume failures.
- **Apple Messages:** replies, typing, media, routing, setup guidance, and chat continuity improve across supported bridge setups.
- **Gateway crashes stop restarting forever:** repeatedly failing Gateways now leave a stable repair path instead of restarting indefinitely.
- **Scheduled work, remote browser control, and workspace terminals:** scheduled work can wake only when something changes, selected signed-in browser tabs can pair remotely and save completed downloads safely, and guarded terminals are available across web, iOS, and Android.

###### Changes

- **Channels and messaging:** expanded native capabilities, clearer progress and delivery behavior, stronger media handling, and broader controls across Telegram, Slack, Discord, Signal, WhatsApp, Apple Messages, Teams, Matrix, Feishu, and other channels.
- **Models and providers:** more models can be discovered, selected, authenticated, and used across cloud, managed, and local routes, with clearer reasoning controls and provider-specific setup.
- **Memory, sessions, and state:** session groups, titles, unread state, checkpoints, context handling, goals, and recovery paths are more consistent across web, mobile, CLI, and connected agents.
- **Tools and automation:** browser control, downloads, cron jobs, terminals, files, search, fetch, MCP, and background task workflows gain new capabilities and clearer controls.
- **Account, device, and data protection:** passwords and tokens stay out of more logs

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.1]_

### v2026.7.1-beta.6 — openclaw 2026.7.1-beta.6

- Date: 2026-07-13
- Version: v2026.7.1-beta.6
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.1-beta.6
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.7.1-beta.6
- Labels: Pre-release

- **added** — Add Featherless, Claude Sonnet 5 and Mythos 5, Meta Muse Spark 1.1, and ClawRouter as new models and providers
- **added** — Add native macOS session browser with model and thinking pickers, slash commands, transcript export, and context usage
- **added** — Add conversational onboarding through Crestodian with AI-guided provider setup, model-judged approvals, and masked credential prompts
- **added** — Add iOS and Android offline session and transcript caches per gateway
- **added** — Add Apple Watch full voice turns
- **added** — Add iOS reply speak-through via configured Gateway TTS with on-device fallback
- **added** — Add generated session titles through utility-model routing
- **added** — Add Gateway-backed groups with rename, fork, archive, and delete controls
- **added** — Add session unread state tracking
- **added** — Add Telegram `/login`, `/steer`, and `/tell` commands for Codex continuity
- **added** — Add Meta provider bundled `muse-spark-1.1` Responses API support with streaming, tool calls, and encrypted reasoning replay
- **added** — Add Android chat agent selector on live chat screen
- **added** — Add Gateway host status display showing host, network address, OS, runtime, uptime, CPU, memory, and disk details in Control UI Settings
- **added** — Add iOS offline chat with recent sessions and canonical transcripts from protected per-gateway cache
- **added** — Add Control UI Talk controls in Settings for voice, model, and sensitivity defaults
- **added** — Add cron model selection in Control UI Quick Create
- **added** — Add Control UI GitHub previews showing issue and pull request state, title, author, activity, and statistics
- **added** — Add Logbook work journal bundled plugin that creates private timeline and daily standup from screen snapshots
- **added** — Add Control UI sidebar customization with pinned destinations and More menu
- **added** — Add Android chat code highlighting for Kotlin, Swift, TypeScript, JavaScript, Python, Bash, and JSON
- **changed** — Make GPT-5.6 the new-setup default with `/think ultra` for Sol and Terra plus `max` for Luna
- **changed** — Make sessions primary with minimal searchable sidebar in Control UI
- **changed** — Make Control UI context ring more compact
- **changed** — Make Control UI dashboard chrome cleaner
- **changed** — Refresh model availability after OAuth renewal
- **changed** — Update Slack progress indicators to use native assistant thread status and rotating loading messages by default
- **changed** — Move Control UI Settings to footer
- **changed** — Remove provider usage quota row from expanded Control UI sidebar
- **changed** — Reveal Control UI per-message token, context, and model details from timestamp hover instead of separate Context button
- **fixed** — Stop recoverable legacy state from blocking startup
- **fixed** — Run container migrations before Gateway readiness
- **fixed** — Preserve Telegram messages when transcript acknowledgement is missing
- **fixed** — Recover final sends across Telegram formatting and flood-wait failures
- **fixed** — Enter control-plane-safe mode after repeated unclean boots instead of restart flapping
- **fixed** — Exit Gateway with `EX_CONFIG` for fatal configuration errors so systemd and launchd stop restart flapping

##### 2026.7.1

###### Highlights

- **New models and providers:** add Featherless, Claude Sonnet 5 and Mythos 5, Meta Muse Spark 1.1, and ClawRouter; make GPT-5.6 the new-setup default with `/think ultra` for Sol and Terra plus `max` for Luna, honor Z.AI `max`, and refresh model availability after OAuth renewal. (#101092, #98254, #101238, #102873, #103070, #103163, #99658, #99759, #98333, #103581, #98021, #104778, #102289) Thanks @vincentkoc, @vortexopenclaw, @HamidShojanazeri, @davemorin, @Solvely-Colin, @jalehman, @steipete-oai, @bdjben, @obviyus, @sallyom, @anyech, @100yenadmin, and @fuller-stack-dev.
- **Control UI and native macOS chat:** make sessions primary with a minimal searchable sidebar, compact context ring, reasoning-effort slider, cleaner dashboard chrome, and a native macOS session browser with model and thinking pickers, slash commands, transcript export, and context usage. (#99289, #99838, #100386, #101103, #101497)
- **Conversational onboarding:** Crestodian now runs a real agent-loop setup across the CLI, web install, and macOS app, with AI-guided provider setup, model-judged approvals bound to exact operations, masked credential prompts, and deterministic fallback when no model is available. (#99935, #100029, #100656, #101887) Thanks @fuller-stack-dev.
- **Offline and spoken mobile chat:** iOS and Android now pre-paint bounded per-gateway session and transcript caches offline; Apple Watch gains full voice turns, and iOS can speak replies through configured Gateway TTS with on-device fallback. (#100219, #100227, #100283, #100770, #100771)
- **Session organization and generated titles:** generate concise session titles through utility-model routing and add Gateway-backed groups, unread state, rename, fork, archive, and delete controls, plus group management across web, iOS, and Android. (#87643, #100814, #101117, #101234) Thanks @zhangguiping-xydt and @Juliangsm.
- **Telegram and Codex continuity:** pair Codex through private `/login`, steer active runs with `/steer` and `/tell`, preserve messages when transcript acknowledgement is missing, adopt durable turns safely, and recover final sends across Telegram formatting and flood-wait failures. (#98006, #98126, #98786, #103664, #103916) Thanks @100yenadmin, @Kyzcreig, @obviyus, and @jalehman.
- **Startup and upgrade recovery:** run container migrations before Gateway readiness, stop recoverable legacy state from blocking startup, and enter control-plane-safe mode after repeated unclean boots instead of restart flapping. (#101881, #104529) Fixes #98565 Thanks @sallyom, @jacobtomlinson, @bdjben, @obviyus, and @shakkernerd.

###### Changes

- **Meta provider:** add bundled `muse-spark-1.1` Responses API support with streaming, tool calls, encrypted reasoning replay, onboarding, exact-model live validation, and standalone npm/ClawHub distribution as `@openclaw/meta-provider`. (#102873, #103070, #103163) Thanks @HamidShojanazeri, @davemorin, @Solvely-Colin, @jalehman, and @vincentkoc.
- **Android chat agent selector:** switch the active agent directly from the live chat screen while keeping chat, Talk mode, and home canvas on the same canonical session. (#80422) Thanks @bcperry and @joshavant.
- **Gateway host status:** show the connected Gateway's host, network address, OS, runtime, uptime, CPU, memory, and disk details in Control UI Settings. (#100478)
- **Gateway crash-loop recovery:** persist boot outcomes, enter control-plane-safe mode after repeated unclean starts, hold transport and provider activation until recovery, and exit with `EX_CONFIG` for fatal configuration errors so systemd and launchd stop restart flapping. Thanks @obviyus.
- **iOS offline chat:** pre-paint recent sessions and canonical transcripts from a protected, bounded per-gateway cache, keep sending disabled offline, and purge cached conversation text when pairing is reset. (#100219)
- **Slack progress indicators:** use Slack's native assistant thread status and rotating loading messages b

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.1-beta.6]_

### v2026.7.1-beta.5 — OpenClaw 2026.7.1-beta.5

- Date: 2026-07-11
- Version: v2026.7.1-beta.5
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.1-beta.5
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.7.1-beta.5
- Labels: Pre-release

- **added** — Add Featherless, Claude Sonnet 5 and Mythos 5, Meta Muse Spark 1.1, and ClawRouter models
- **added** — Add native macOS session browser with model and thinking pickers, slash commands, transcript export, and context usage
- **added** — Add conversational onboarding with AI-guided provider setup, model-judged approvals, masked credential prompts, and deterministic fallback
- **added** — Add iOS and Android pre-painted bounded per-gateway session and transcript caches for offline use
- **added** — Add Apple Watch full voice turns and iOS reply speaking through configured Gateway TTS with on-device fallback
- **added** — Add session generation with concise titles through utility-model routing and Gateway-backed groups with rename, fork, archive, and delete controls
- **added** — Add Telegram `/login` pairing for Codex, `/steer` and `/tell` commands for active runs, and durable turn adoption
- **added** — Add bundled Meta Muse Spark 1.1 Responses API support with streaming, tool calls, encrypted reasoning replay, and standalone npm/ClawHub distribution
- **added** — Add Android chat agent selector to switch active agent directly from live chat screen
- **added** — Add Gateway host status showing connected Gateway's host, network address, OS, runtime, uptime, CPU, memory, and disk details in Control UI Settings
- **added** — Add Control UI Talk controls to keep voice, model, sensitivity, and realtime defaults in Settings and use composer microphone caret to select browser audio input
- **added** — Add agent-turn model selection in Control UI Quick Create for cron jobs
- **added** — Add Control UI GitHub previews showing issue and pull request state, title, author, activity, comments, and change statistics
- **added** — Add Logbook work journal plugin that turns paired-node screen snapshots into private timeline, daily standup, and timeline-grounded Q&A
- **added** — Add Control UI sidebar navigation with customizable pinned destination set and Settings moved to footer
- **added** — Add Android chat code highlighting for Kotlin, Swift, TypeScript, JavaScript, Python, Bash, and JSON blocks with theme-aware syntax colors
- **changed** — Make sessions primary with minimal searchable sidebar and compact context ring in Control UI
- **changed** — Make GPT-5.6 the new-setup default with /think ultra for Sol and Terra plus max for Luna
- **changed** — Honor Z.AI max and refresh model availability after OAuth renewal
- **changed** — Run container migrations before Gateway readiness
- **changed** — Show Slack native assistant thread status and rotating loading messages by default with acknowledgement reactions requiring messages.statusReactions.enabled: true
- **changed** — Reveal per-message token, context, and model details from timestamp on hover or activation
- **changed** — Reveal truncated recent-session names with reduced-motion-safe hover animation
- **changed** — Remove provider usage quota row from expanded sidebar while keeping usage details in chat composer and Usage page
- **fixed** — Stop recoverable legacy state from blocking startup
- **fixed** — Enter control-plane-safe mode after repeated unclean boots instead of restart flapping
- **fixed** — Preserve messages when Telegram transcript acknowledgement is missing
- **fixed** — Recover final sends across Telegram formatting and flood-wait failures

##### 2026.7.1

###### Highlights

- **New models and providers:** add Featherless, Claude Sonnet 5 and Mythos 5, Meta Muse Spark 1.1, and ClawRouter; make GPT-5.6 the new-setup default with `/think ultra` for Sol and Terra plus `max` for Luna, honor Z.AI `max`, and refresh model availability after OAuth renewal. (#101092, #98254, #101238, #102873, #103070, #103163, #99658, #99759, #98333, #103581, #98021, #104778, #102289) Thanks @vincentkoc, @vortexopenclaw, @HamidShojanazeri, @davemorin, @Solvely-Colin, @jalehman, @steipete-oai, @bdjben, @obviyus, @sallyom, @anyech, @100yenadmin, and @fuller-stack-dev.
- **Control UI and native macOS chat:** make sessions primary with a minimal searchable sidebar, compact context ring, reasoning-effort slider, cleaner dashboard chrome, and a native macOS session browser with model and thinking pickers, slash commands, transcript export, and context usage. (#99289, #99838, #100386, #101103, #101497)
- **Conversational onboarding:** Crestodian now runs a real agent-loop setup across the CLI, web install, and macOS app, with AI-guided provider setup, model-judged approvals bound to exact operations, masked credential prompts, and deterministic fallback when no model is available. (#99935, #100029, #100656, #101887) Thanks @fuller-stack-dev.
- **Offline and spoken mobile chat:** iOS and Android now pre-paint bounded per-gateway session and transcript caches offline; Apple Watch gains full voice turns, and iOS can speak replies through configured Gateway TTS with on-device fallback. (#100219, #100227, #100283, #100770, #100771)
- **Session organization and generated titles:** generate concise session titles through utility-model routing and add Gateway-backed groups, unread state, rename, fork, archive, and delete controls, plus group management across web, iOS, and Android. (#87643, #100814, #101117, #101234) Thanks @zhangguiping-xydt and @Juliangsm.
- **Telegram and Codex continuity:** pair Codex through private `/login`, steer active runs with `/steer` and `/tell`, preserve messages when transcript acknowledgement is missing, adopt durable turns safely, and recover final sends across Telegram formatting and flood-wait failures. (#98006, #98126, #98786, #103664, #103916) Thanks @100yenadmin, @Kyzcreig, @obviyus, and @jalehman.
- **Startup and upgrade recovery:** run container migrations before Gateway readiness, stop recoverable legacy state from blocking startup, and enter control-plane-safe mode after repeated unclean boots instead of restart flapping. (#101881, #104529) Fixes #98565 Thanks @sallyom, @jacobtomlinson, @bdjben, @obviyus, and @shakkernerd.

###### Changes

- **Meta provider:** add bundled `muse-spark-1.1` Responses API support with streaming, tool calls, encrypted reasoning replay, onboarding, exact-model live validation, and standalone npm/ClawHub distribution as `@openclaw/meta-provider`. (#102873, #103070, #103163) Thanks @HamidShojanazeri, @davemorin, @Solvely-Colin, @jalehman, and @vincentkoc.
- **Android chat agent selector:** switch the active agent directly from the live chat screen while keeping chat, Talk mode, and home canvas on the same canonical session. (#80422) Thanks @bcperry and @joshavant.
- **Gateway host status:** show the connected Gateway's host, network address, OS, runtime, uptime, CPU, memory, and disk details in Control UI Settings. (#100478)
- **Gateway crash-loop recovery:** persist boot outcomes, enter control-plane-safe mode after repeated unclean starts, hold transport and provider activation until recovery, and exit with `EX_CONFIG` for fatal configuration errors so systemd and launchd stop restart flapping. Thanks @obviyus.
- **iOS offline chat:** pre-paint recent sessions and canonical transcripts from a protected, bounded per-gateway cache, keep sending disabled offline, and purge cached conversation text when pairing is reset. (#100219)
- **Slack progress indicators:** use Slack's native assistant thread status and rotating loading messages b

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.1-beta.5]_

### v2026.7.1-beta.2 — openclaw 2026.7.1-beta.2

- Date: 2026-07-05
- Version: v2026.7.1-beta.2
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.1-beta.2
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.7.1-beta.2
- Labels: Pre-release

- **added** — Add Featherless, Claude Sonnet 5 and Mythos 5, Meta Muse Spark 1.1, and ClawRouter as new models and providers
- **added** — Add native macOS session browser with model and thinking pickers, slash commands, transcript export, and context usage
- **added** — Add conversational onboarding through Crestodian with agent-loop setup across CLI, web install, and macOS app
- **added** — Add iOS and Android offline session and transcript caches per gateway
- **added** — Add Apple Watch full voice turns and iOS text-to-speech replies through configured Gateway TTS with on-device fallback
- **added** — Add session generation with concise titles through utility-model routing and Gateway-backed groups with unread state, rename, fork, archive, and delete controls
- **added** — Add Telegram `/login` pairing for Codex, `/steer` and `/tell` commands to manage active runs, and recovery for final sends across formatting and flood-wait failures
- **added** — Add bundled Meta provider muse-spark-1.1 Responses API support with streaming, tool calls, encrypted reasoning replay, and standalone npm/ClawHub distribution
- **added** — Add Android chat agent selector to switch active agent directly from live chat screen
- **added** — Add Gateway host status display showing connected Gateway's host, network address, OS, runtime, uptime, CPU, memory, and disk details in Control UI Settings
- **added** — Add Control UI Talk controls in Settings → Communications → Talk for voice, model, sensitivity, and realtime defaults
- **added** — Add agent-turn model selection in Control UI Quick Create for cron jobs
- **added** — Add Control UI GitHub previews showing issue and pull request state, title, author, activity, comments, and change statistics
- **added** — Add Logbook work journal disabled-by-default bundled plugin that turns paired-node screen snapshots into private timeline, daily standup, and timeline-grounded Q&A
- **added** — Add Android chat code highlighting with Kotlin, Swift, TypeScript, JavaScript, Python, Bash, and JSON syntax colors
- **changed** — Make sessions primary with minimal searchable sidebar, compact context ring, and reasoning-effort slider in Control UI
- **changed** — Make GPT-5.6 the default model for new setups with /think ultra for Sol and Terra plus max for Luna
- **changed** — Make Slack use native assistant thread status and rotating loading messages by default while keeping acknowledgement reactions static
- **changed** — Change Control UI message context to reveal per-message token, context, and model details from timestamp on hover or activation
- **changed** — Change Control UI session titles to reveal truncated recent-session names with reduced-motion-safe hover animation
- **changed** — Change Control UI sidebar navigation to show customizable pinned destination set with remaining pages under More and Settings in footer
- **changed** — Remove provider usage quota row from expanded Control UI sidebar while keeping usage details in chat composer and Usage page
- **changed** — Refresh model availability after OAuth renewal
- **fixed** — Fix startup by running container migrations before Gateway readiness
- **fixed** — Fix recoverable legacy state from blocking startup
- **fixed** — Fix repeated unclean boots by entering control-plane-safe mode instead of restart flapping
- **fixed** — Fix Telegram message preservation when transcript acknowledgement is missing
- **fixed** — Fix Gateway crash-loop recovery by persisting boot outcomes and entering control-plane-safe mode after repeated unclean starts
- **fixed** — Fix iOS offline chat to keep sending disabled offline and purge cached conversation text when pairing is reset

##### 2026.7.1

###### Highlights

- **New models and providers:** add Featherless, Claude Sonnet 5 and Mythos 5, Meta Muse Spark 1.1, and ClawRouter; make GPT-5.6 the new-setup default with `/think ultra` for Sol and Terra plus `max` for Luna, honor Z.AI `max`, and refresh model availability after OAuth renewal. (#101092, #98254, #101238, #102873, #103070, #103163, #99658, #99759, #98333, #103581, #98021, #104778, #102289) Thanks @vincentkoc, @vortexopenclaw, @HamidShojanazeri, @davemorin, @Solvely-Colin, @jalehman, @steipete-oai, @bdjben, @obviyus, @sallyom, @anyech, @100yenadmin, and @fuller-stack-dev.
- **Control UI and native macOS chat:** make sessions primary with a minimal searchable sidebar, compact context ring, reasoning-effort slider, cleaner dashboard chrome, and a native macOS session browser with model and thinking pickers, slash commands, transcript export, and context usage. (#99289, #99838, #100386, #101103, #101497)
- **Conversational onboarding:** Crestodian now runs a real agent-loop setup across the CLI, web install, and macOS app, with AI-guided provider setup, model-judged approvals bound to exact operations, masked credential prompts, and deterministic fallback when no model is available. (#99935, #100029, #100656, #101887) Thanks @fuller-stack-dev.
- **Offline and spoken mobile chat:** iOS and Android now pre-paint bounded per-gateway session and transcript caches offline; Apple Watch gains full voice turns, and iOS can speak replies through configured Gateway TTS with on-device fallback. (#100219, #100227, #100283, #100770, #100771)
- **Session organization and generated titles:** generate concise session titles through utility-model routing and add Gateway-backed groups, unread state, rename, fork, archive, and delete controls, plus group management across web, iOS, and Android. (#87643, #100814, #101117, #101234) Thanks @zhangguiping-xydt and @Juliangsm.
- **Telegram and Codex continuity:** pair Codex through private `/login`, steer active runs with `/steer` and `/tell`, preserve messages when transcript acknowledgement is missing, adopt durable turns safely, and recover final sends across Telegram formatting and flood-wait failures. (#98006, #98126, #98786, #103664, #103916) Thanks @100yenadmin, @Kyzcreig, @obviyus, and @jalehman.
- **Startup and upgrade recovery:** run container migrations before Gateway readiness, stop recoverable legacy state from blocking startup, and enter control-plane-safe mode after repeated unclean boots instead of restart flapping. (#101881, #104529) Fixes #98565 Thanks @sallyom, @jacobtomlinson, @bdjben, @obviyus, and @shakkernerd.

###### Changes

- **Meta provider:** add bundled `muse-spark-1.1` Responses API support with streaming, tool calls, encrypted reasoning replay, onboarding, exact-model live validation, and standalone npm/ClawHub distribution as `@openclaw/meta-provider`. (#102873, #103070, #103163) Thanks @HamidShojanazeri, @davemorin, @Solvely-Colin, @jalehman, and @vincentkoc.
- **Android chat agent selector:** switch the active agent directly from the live chat screen while keeping chat, Talk mode, and home canvas on the same canonical session. (#80422) Thanks @bcperry and @joshavant.
- **Gateway host status:** show the connected Gateway's host, network address, OS, runtime, uptime, CPU, memory, and disk details in Control UI Settings. (#100478)
- **Gateway crash-loop recovery:** persist boot outcomes, enter control-plane-safe mode after repeated unclean starts, hold transport and provider activation until recovery, and exit with `EX_CONFIG` for fatal configuration errors so systemd and launchd stop restart flapping. Thanks @obviyus.
- **iOS offline chat:** pre-paint recent sessions and canonical transcripts from a protected, bounded per-gateway cache, keep sending disabled offline, and purge cached conversation text when pairing is reset. (#100219)
- **Slack progress indicators:** use Slack's native assistant thread status and rotating loading messages b

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.1-beta.2]_

### v2026.7.1-beta.1 — openclaw 2026.7.1-beta.1

- Date: 2026-07-02
- Version: v2026.7.1-beta.1
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.1-beta.1
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.7.1-beta.1
- Labels: Pre-release

- **added** — Add Featherless, Claude Sonnet 5 and Mythos 5, Meta Muse Spark 1.1, and ClawRouter model providers
- **added** — Add native macOS session browser with model and thinking pickers, slash commands, transcript export, and context usage
- **added** — Add conversational onboarding with Crestodian agent-loop setup across CLI, web install, and macOS app including AI-guided provider setup and masked credential prompts
- **added** — Add offline session and transcript caches for iOS and Android with per-gateway boundaries
- **added** — Add Apple Watch full voice turns and iOS reply speech through configured Gateway TTS with on-device fallback
- **added** — Add generated session titles through utility-model routing and Gateway-backed groups with unread state, rename, fork, archive, and delete controls
- **added** — Add Telegram `/login`, `/steer`, and `/tell` commands for Codex pairing and active run steering
- **added** — Add bundled Meta provider `muse-spark-1.1` Responses API support with streaming, tool calls, and encrypted reasoning replay
- **added** — Add Android chat agent selector to switch active agent directly from live chat screen
- **added** — Add Gateway host status display showing connected Gateway's host, network address, OS, runtime, uptime, CPU, memory, and disk details in Control UI Settings
- **added** — Add iOS offline chat with pre-painted recent sessions and canonical transcripts from protected per-gateway cache
- **added** — Add Control UI Talk controls to keep voice, model, and sensitivity defaults in Settings and use composer microphone caret to select browser audio input
- **added** — Add cron model selection in Control UI Quick Create to choose agent-turn model
- **added** — Add Control UI GitHub previews showing issue and pull request state, title, author, activity, comments, and change statistics
- **added** — Add Logbook work journal disabled-by-default bundled plugin that converts paired-node screen snapshots into private timeline and daily standup
- **added** — Add Android chat code highlighting for Kotlin, Swift, TypeScript, JavaScript, Python, Bash, and JSON blocks with theme-aware syntax colors
- **changed** — Make GPT-5.6 the new-setup default with `/think ultra` for Sol and Terra plus `max` for Luna
- **changed** — Honor Z.AI `max` and refresh model availability after OAuth renewal
- **changed** — Make sessions primary with minimal searchable sidebar, compact context ring, and reasoning-effort slider in Control UI
- **changed** — Use Slack's native assistant thread status and rotating loading messages by default while keeping acknowledgement reactions static
- **changed** — Reveal per-message token, context, and model details from timestamp on hover or activation instead of separate Context button
- **changed** — Reveal truncated recent-session names with reduced-motion-safe hover animation in Control UI
- **changed** — Update Control UI sidebar navigation to show small customizable pinned destination set with remaining pages under More and Settings moved to footer
- **changed** — Remove provider usage quota row from expanded sidebar while keeping usage details available in chat composer and Usage page
- **fixed** — Run container migrations before Gateway readiness to prevent startup blocking
- **fixed** — Stop recoverable legacy state from blocking startup and enter control-plane-safe mode after repeated unclean boots instead of restart flapping
- **fixed** — Preserve Telegram messages when transcript acknowledgement is missing and recover final sends across Telegram formatting and flood-wait failures

##### 2026.7.1

###### Highlights

- **New models and providers:** add Featherless, Claude Sonnet 5 and Mythos 5, Meta Muse Spark 1.1, and ClawRouter; make GPT-5.6 the new-setup default with `/think ultra` for Sol and Terra plus `max` for Luna, honor Z.AI `max`, and refresh model availability after OAuth renewal. (#101092, #98254, #101238, #102873, #103070, #103163, #99658, #99759, #98333, #103581, #98021, #104778, #102289) Thanks @vincentkoc, @vortexopenclaw, @HamidShojanazeri, @davemorin, @Solvely-Colin, @jalehman, @steipete-oai, @bdjben, @obviyus, @sallyom, @anyech, @100yenadmin, and @fuller-stack-dev.
- **Control UI and native macOS chat:** make sessions primary with a minimal searchable sidebar, compact context ring, reasoning-effort slider, cleaner dashboard chrome, and a native macOS session browser with model and thinking pickers, slash commands, transcript export, and context usage. (#99289, #99838, #100386, #101103, #101497)
- **Conversational onboarding:** Crestodian now runs a real agent-loop setup across the CLI, web install, and macOS app, with AI-guided provider setup, model-judged approvals bound to exact operations, masked credential prompts, and deterministic fallback when no model is available. (#99935, #100029, #100656, #101887) Thanks @fuller-stack-dev.
- **Offline and spoken mobile chat:** iOS and Android now pre-paint bounded per-gateway session and transcript caches offline; Apple Watch gains full voice turns, and iOS can speak replies through configured Gateway TTS with on-device fallback. (#100219, #100227, #100283, #100770, #100771)
- **Session organization and generated titles:** generate concise session titles through utility-model routing and add Gateway-backed groups, unread state, rename, fork, archive, and delete controls, plus group management across web, iOS, and Android. (#87643, #100814, #101117, #101234) Thanks @zhangguiping-xydt and @Juliangsm.
- **Telegram and Codex continuity:** pair Codex through private `/login`, steer active runs with `/steer` and `/tell`, preserve messages when transcript acknowledgement is missing, adopt durable turns safely, and recover final sends across Telegram formatting and flood-wait failures. (#98006, #98126, #98786, #103664, #103916) Thanks @100yenadmin, @Kyzcreig, @obviyus, and @jalehman.
- **Startup and upgrade recovery:** run container migrations before Gateway readiness, stop recoverable legacy state from blocking startup, and enter control-plane-safe mode after repeated unclean boots instead of restart flapping. (#101881, #104529) Fixes #98565 Thanks @sallyom, @jacobtomlinson, @bdjben, @obviyus, and @shakkernerd.

###### Changes

- **Meta provider:** add bundled `muse-spark-1.1` Responses API support with streaming, tool calls, encrypted reasoning replay, onboarding, exact-model live validation, and standalone npm/ClawHub distribution as `@openclaw/meta-provider`. (#102873, #103070, #103163) Thanks @HamidShojanazeri, @davemorin, @Solvely-Colin, @jalehman, and @vincentkoc.
- **Android chat agent selector:** switch the active agent directly from the live chat screen while keeping chat, Talk mode, and home canvas on the same canonical session. (#80422) Thanks @bcperry and @joshavant.
- **Gateway host status:** show the connected Gateway's host, network address, OS, runtime, uptime, CPU, memory, and disk details in Control UI Settings. (#100478)
- **Gateway crash-loop recovery:** persist boot outcomes, enter control-plane-safe mode after repeated unclean starts, hold transport and provider activation until recovery, and exit with `EX_CONFIG` for fatal configuration errors so systemd and launchd stop restart flapping. Thanks @obviyus.
- **iOS offline chat:** pre-paint recent sessions and canonical transcripts from a protected, bounded per-gateway cache, keep sending disabled offline, and purge cached conversation text when pairing is reset. (#100219)
- **Slack progress indicators:** use Slack's native assistant thread status and rotating loading messages b

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.7.1-beta.1]_

### v2026.6.11 — openclaw 2026.6.11

- Date: 2026-06-30
- Version: v2026.6.11
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.6.11
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.6.11

- **fixed** — Fix Google Chat direct messages sometimes being treated as group conversations, ensuring they reach the correct one-to-one chat
- **added** — Feishu voice replies now show their duration in the chat bubble
- **fixed** — Discord and Telegram replies and mirrored chat history stay tied to the intended conversation more consistently across repeated replies and session changes
- **fixed** — Background image, video, and music results return to the chat that requested them instead of appearing to fail or being sent to the wrong peer
- **fixed** — Telegram answers stay attached to the user's current question when they quote an earlier bot message
- **changed** — QQBot group admins can choose how broadly slash commands are available, and private-only commands now direct users to a private chat
- **fixed** — Heartbeat checks using reasoning-capable models now show the assistant's intended reply instead of exposing internal reasoning
- **fixed** — Telegram progress-mode chats now clear an old progress bubble before newer tool output or artifacts appear
- **fixed** — iMessage command-and-link messages now stay together as one OpenClaw turn when delayed link previews arrive
- **fixed** — Discord replies sent through the message tool no longer trigger a misleading failure warning in message_tool_only source-channel turns
- **fixed** — WhatsApp group conversations now preserve the right message and group context more reliably during retries, reconnects, and group changes
- **fixed** — Fix OpenClaw sometimes replying to its own delayed iMessage echoes when stray leading characters keep the sent message from being recognized
- **fixed** — Telegram webhook users can keep receiving DMs and group messages through brief channel restarts and configuration reloads without message blackouts
- **fixed** — Matrix E2EE gateways can stay online during long-running use instead of gradually consuming memory until a crash
- **fixed** — Telegram users now see the intended native reaction instead of leaked instructions or a dropped reaction-only reply
- **fixed** — Telegram progress updates for commands, searches, updates, and API activity now stay readable with plain-text fallback when Telegram cannot parse the formatting

We heard the feedback. v2026.6.11 focuses on the rough edges that make OpenClaw feel less dependable, with fixes for misplaced replies, stuck sends, reconnects, model setup failures, and safer admin defaults.

[Full release notes](https://docs.openclaw.ai/releases/2026.6.11)

###### Highlights

###### Channel delivery reliability

Delivery and reconnect fixes span [Telegram](https://docs.openclaw.ai/channels/telegram), [WhatsApp](https://docs.openclaw.ai/channels/whatsapp), [Matrix](https://docs.openclaw.ai/channels/matrix), [Google Chat](https://docs.openclaw.ai/channels/googlechat), [iMessage](https://docs.openclaw.ai/channels/imessage), [Feishu](https://docs.openclaw.ai/channels/feishu), [Mattermost](https://docs.openclaw.ai/channels/mattermost), [WebChat](https://docs.openclaw.ai/web/webchat), the [Control UI](https://docs.openclaw.ai/web/control-ui), and the [terminal UI](https://docs.openclaw.ai/cli/tui).

- Fixes newer Google Chat direct messages sometimes being treated like group conversations, so they reach the correct one-to-one chat while Space and group-chat messages keep their existing routing. [#58993](https://github.com/openclaw/openclaw/pull/58993) Thanks @starhappysh, @vincentkoc.
- Feishu voice replies from OpenClaw now show their duration in the chat bubble, so recipients can see how long the audio is before playing it. [#89172](https://github.com/openclaw/openclaw/pull/89172) Related [#53798](https://github.com/openclaw/openclaw/issues/53798). Thanks @areslp, @fxz26284407, @kinrocw.
- Discord and Telegram replies and mirrored chat history stay tied to the intended conversation more consistently, including across repeated Telegram replies and session changes. [#89911](https://github.com/openclaw/openclaw/pull/89911) Thanks @jalehman.
- Background image, video, and music results now return to the chat that requested them when the task starts without a full conversation target, instead of appearing to fail after creation or being sent to the wrong peer as the session moves. [#89949](https://github.com/openclaw/openclaw/pull/89949) Related [#86034](https://github.com/openclaw/openclaw/issues/86034). Thanks @tianxiaochannel-oss88, @wangwllu.
- Telegram answers now stay attached to the user's current question when they quote an earlier bot message, while quotes of other people's messages still reply to the selected quote. [#90475](https://github.com/openclaw/openclaw/pull/90475) Thanks @moeedahmed.
- QQBot group admins can choose how broadly slash commands are available, and private-only commands now direct users to a private chat instead of being exposed or silently ignored in groups. [#92154](https://github.com/openclaw/openclaw/pull/92154) Thanks @sliverp.
- Heartbeat checks using reasoning-capable models now show the assistant's intended reply instead of exposing internal reasoning in Telegram, WhatsApp, and other channels, while opt-in Thinking messages still work. [#92356](https://github.com/openclaw/openclaw/pull/92356) Related [#92260](https://github.com/openclaw/openclaw/issues/92260). Thanks @jmpei, @tangtaizong666, @vincentkoc.
- Telegram progress-mode chats now clear an old progress bubble before newer tool output or artifacts appear, keeping the conversation in a clean, readable order. [#93002](https://github.com/openclaw/openclaw/pull/93002) Related [#90753](https://github.com/openclaw/openclaw/issues/90753). Thanks @shadow-enthusiast, @zhangguiping-xydt.
- iMessage command-and-link messages now stay together as one OpenClaw turn when delayed link previews arrive, while unrelated quick messages remain separate for users who enabled same-sender DM coalescing. [#93143](https://github.com/openclaw/openclaw/pull/93143) Thanks @omarshahine.
- Successful Discord replies sent through the message tool no longer trigger a misleading failure warning in affected `message_tool_only` source-channel turns. [#94072](https://github.com/openclaw/openclaw/pull/94072) Related [#93875](https://github.com/openclaw/openc

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.6.11]_

### v2026.6.11-beta.2 — openclaw 2026.6.11-beta.2

- Date: 2026-06-28
- Version: v2026.6.11-beta.2
- Original notes: https://github.com/openclaw/openclaw/releases/tag/v2026.6.11-beta.2
- Permalink: https://whatsnew.fyi/product/openclaw/releases/v2026.6.11-beta.2
- Labels: Pre-release

- **added** — Slack relay mode for incoming messages
- **added** — Native Mattermost /oc_queue command
- **added** — Per-DM model overrides for channel operations
- **added** — openclaw agent --message-file flag for file-driven workflows
- **added** — RAFT CLI wake bridge for remote wake-up paths
- **added** — Bundled plugin icon metadata available to installed clients
- **added** — Android settings detail panels for improved mobile configuration visibility and control
- **changed** — Channel identity hook context for more precise routing in gateway and plugin tooling
- **changed** — Per-agent usage-cost reporting for operators
- **changed** — Catalog parsing, reasoning controls, and provider model resolution to handle more live provider variants
- **changed** — Encrypted reasoning support for additional providers
- **fixed** — Telegram progress rendering for rich messages with line breaks
- **fixed** — Telegram webhook lifecycle to prevent clearing registered webhook on channel restart
- **fixed** — Telegram streaming progress placeholders materialization
- **fixed** — Telegram outbound reaction directives
- **fixed** — WhatsApp native quote replies preservation
- **fixed** — WhatsApp Baileys group reliability with missing retry and cache hooks
- **fixed** — Duplicate mirror writes in channel delivery
- **fixed** — Queued update draining
- **fixed** — WhatsApp durable reply targets
- **fixed** — Codex partial deltas to reduce lost progress
- **fixed** — Harness activation for more reliable agent turns
- **fixed** — Long-context prompt-cache stability for inconsistent runs
- **fixed** — Stuck release claims in gateway and session safety
- **fixed** — Draining-state reporting
- **fixed** — Remote probe timeouts
- **fixed** — Malformed paired access lists handling
- **fixed** — Non-delivery session identity in gateway operations
- **fixed** — Aborted runs to stop cleanly
- **fixed** — Provider response bodies to stay bounded
- **fixed** — Claude CLI credit failures to continue through fallback
- **fixed** — Codex usage-limit responses classification
- **fixed** — OpenRouter IDs resolution
- **fixed** — Ollama discovery and embeddings
- **fixed** — Gemini freshness handling
- **fixed** — Model-catalog prefixes resolution against runtime metadata
- **fixed** — Non-interactive configure to fail closed
- **fixed** — TLS paths to reject empty values
- **fixed** — Memory artifacts sanitization
- **fixed** — UI to use patched DOMPurify release
- **fixed** — No-config delivery checks
- **fixed** — Thread-aware dedupe in delivery
- **fixed** — Pending recurring runs to retain intended destinations

##### 2026.6.11

###### Highlights

- **More capable channel control:** Slack relay mode, native Mattermost `/oc_queue`, and per-DM model overrides make channel operations easier to automate and tune. (#94707, #95546, #95120) Thanks @sjf-oa, @amknight, @xydigit-zt, @thomaszta, and @gandalf-at-lerian.
- **Richer operator workflows:** `openclaw agent --message-file` and the RAFT CLI wake bridge add practical file-driven and remote wake-up paths. (#93351, #95497) Thanks @ooiuuii and @vincentkoc.
- **Safer plugin distribution:** additional official plugins are externalized cleanly, with bundled plugin icon metadata available to installed clients. (#95683, #95845) Thanks @vincentkoc and @Patrick-Erichsen.
- **Stronger mobile operations:** Android settings detail panels improve configuration visibility and control on mobile. (#95148) Thanks @Tosko4.
- **More reliable agent turns:** Codex partial deltas, harness activation, and long-context prompt-cache stability reduce lost progress and inconsistent runs. (#95404, #95652, #95624) Thanks @agonza1 and @vincentkoc.

###### Changes

- **Gateway and plugin tooling:** channel identity hook context and per-agent usage-cost reporting give integrations and operators more precise routing and accounting. (#91903, #94483) Thanks @lanzhi-lee, @vincentkoc, and @ly-wang19.
- **Provider and model coverage:** catalog parsing, reasoning controls, provider model resolution, and encrypted reasoning support now handle more live provider variants. (#95283, #95710, #95268, #95744, #95686, #93956) Thanks @ZengWen-DT, @vincentkoc, @Marvinthebored, @Darren2030, @daniel-alejandro-t, @parveshsaini, @geraint0923, @fuller-stack-dev, and @jason-allen-oneal.

###### Fixes

- **Channel delivery:** Telegram progress rendering, webhook lifecycle, reaction directives, duplicate mirror writes, queued update draining, and WhatsApp durable reply targets are now more reliable. (#95532, #93002, #95183, #94506, #94977, #95069, #95577, #95007, #95914) Thanks @amknight, @snowzlmbot, @zhangguiping-xydt, @shadow-enthusiast, @xialonglee, @travellingsoldier85, @obviyus, @hugenshen, @Cuttingwater, @heichaowo, @LiuwqGit, @freidrich-goldenflow, @mcaxtr, and @vincentkoc.
- **WhatsApp and message identity:** native quotes, Baileys group reliability, and approval reactions across JID drift now preserve the intended conversation context. (#95483, #94338, #95935) Thanks @mcaxtr, @xialonglee, and @octopuslabs-fl.
- **Gateway and session safety:** stuck release claims, draining-state reporting, remote probe timeouts, malformed paired access lists, and non-delivery session identity are handled without silent routing loss. (#95299, #94915, #89859, #92178, #95467) Thanks @mikasa0818, @kriegerbangerz-ship-it, @markoub, @vincentkoc, @maxschachere, @mushuiyu886, @gozzbb2, @wangmiao0668000666, @ly-wang19, @EmilioNicolas, @yetval, and @hellocli.
- **Agent and fallback behavior:** aborted runs stop cleanly, provider response bodies stay bounded, Claude CLI credit failures continue through fallback, and Codex usage-limit responses classify correctly. (#94412, #95218, #95508, #95420, #95418, #95417, #95400) Thanks @szsip239, @vincentkoc, @Alix-007, @mikasa0818, @sallyom, @riazrahaman, and @jason-allen-oneal.
- **Provider and model edge cases:** OpenRouter IDs, Ollama discovery and embeddings, Gemini freshness, and model-catalog prefixes now resolve against the right runtime metadata. (#95268, #94811, #93956, #95682, #95744) Thanks @Darren2030, @daniel-alejandro-t, @mushuiyu886, @jason-allen-oneal, @Sunjae-k, @parveshsaini, @vincentkoc, and @shakkernerd.
- **Configuration and UI guardrails:** non-interactive configure fails closed, TLS paths reject empty values, memory artifacts are sanitized, and the UI uses the patched DOMPurify release. (#94238, #94054, #95791, #95691) Thanks @ruomuxydt, @NianJiuZst, @miorbnli, @vincentkoc, @SweetSophia, and @YB0y.
- **Cron and delivery validation:** no-config delivery checks, thread-aware dedupe, and pending recur

_[Truncated at 4000 characters — full notes: https://github.com/openclaw/openclaw/releases/tag/v2026.6.11-beta.2]_