v0.57.5

## What's Changed (since v0.56.0)

### ✨ Features

**Agent & Orchestration**
* feat(agent): multi-agent personalities with scoped memory and delegation by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2895
* feat(agent): add orchestration control plane by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2996
* feat(agent): live streaming subagent view with reopenable sub-threads by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3007
* feat(agent): apply iteration turn cap only to short-running agents by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3056
* feat(agent-registry): add configurable agent registry by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2993
* feat(subagent): persist sub-agent runs and let orchestrator relay user messages by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3055
* feat(model-council): multi-model deliberation — parallel members + chair synthesis by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2890
* feat(workflows): agent workflows — phase-keyed task lifecycle playbooks by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3008
* feat: proactive agent task pipeline — dispatcher, approval, executors, write-back by @sanil-23 in https://github.com/tinyhumansai/openhuman/pull/2974
* feat: Agent Tasks board + Settings > Agents registry panel by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2998
* Enrich agent task board workflow by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2891
* feat(task-sources): proactive task ingestion from GitHub/Notion/Linear/ClickUp by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2894

**Memory & Intelligence**
* feat(memory): redesign sync flow with ingest_summary, graph improvements, audit log by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3113
* feat(memory): smart multi-strategy retrieval walk by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3077
* feat(memory): repo-grouped GitHub raw archive + entities & priority by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3047
* feat: Memory subtabs + WebGL memory graph; drop Knowledge Vaults by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3040
* feat(memory_sources): add Memory Sources domain — user-configurable data connectors by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2893
* feat(memory-conversations): trigram inverted index for cross-thread search by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/2756
* feat(migration): add Hermes Agent memory importer (#1440) by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2799
* feat(intelligence): add Knowledge Graph Centrality by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2931
* feat(intelligence): add Knowledge Freshness by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2932
* feat(intelligence): add Entity Associations by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2942
* feat(intelligence): add Connection Path finder by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2946
* feat(intelligence): add Memory Timeline by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2950
* feat(intelligence): add Graph Cohesion by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2978
* feat(intelligence): add Namespace Overview by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2964
* feat(intelligence): add architecture diagram viewer by @sunilkumarvalmiki in https://github.com/tinyhumansai/openhuman/pull/2687

**MCP**
* feat(mcp): ship MCP registry — unhide UI, transport-aware install, reconfigure, registry creds (#3039) by @senamakel-droid in https://github.com/tinyhumansai/openhuman/pull/3058
* feat(mcp): interactive Tool Execution Playground (schema viewer + JSON args editor + run + history) by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2644
* feat(mcp): Sharable Inventory — portable secret-free manifest with export, parse, validate, and explicit per-entry import by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2655
* feat(mcp): connection health toolbar with bulk Retry All / Disconnect All actions by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2641
* feat(mcp): search + keyboard navigation for installed MCP servers by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2639
* feat(mcp): expose bundled prompts as static MCP resources by @M3gA-Mind in https://github.com/tinyhumansai/openhuman/pull/2745
* feat(mcp): implement resources/templates/list as static empty catalog by @obchain in https://github.com/tinyhumansai/openhuman/pull/2828
* feat(mcp): i18n + a11y McpStatusBadge status labels by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2635
* feat(mcp-registry): InstalledServer HTTP-remote transport by @justinhsu1477 in https://github.com/tinyhumansai/openhuman/pull/2603

**Tools**
* feat(tools): expose ~160 agent tools across 23 core domains (overextending tools default-OFF) by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3050
* feat(tool-registry): add trusted capability providers by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2551
* feat(tool_policy): diagnostics RPC and Developer Options panel by @MrMrVlad in https://github.com/tinyhumansai/openhuman/pull/2715
* Add generated tool provenance and admission checks by @vaddisrinivas in https://github.com/tinyhumansai/openhuman/pull/2549
* feat(image): add image tool contracts by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2997
* feat(search): add registry-backed search module by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2999
* feat(search): add Querit provider by @MurphyZzzzz in https://github.com/tinyhumansai/openhuman/pull/2753

**Inference & Providers**
* Expand inference provider catalog (#3057) by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3062
* feat(inference): add claude_agent_sdk provider (Claude CLI subprocess) by @M3gA-Mind in https://github.com/tinyhumansai/openhuman/pull/2746
* Add Codex OAuth provider support by @junjunjunbong in https://github.com/tinyhumansai/openhuman/pull/2748
* feat(dashboard): per-model health comparison table by @Sathvik-1007 in https://github.com/tinyhumansai/openhuman/pull/2823

**UI & Chat**
* feat(chat): auto-resize composer textarea to fit content by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/3139
* feat(chat): add image attachment support to composer by @rsd-darshan in https://github.com/tinyhumansai/openhuman/pull/2676
* feat(chat): render agent-bubble LaTeX with KaTeX and safe math detection by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2697
* feat(ui): hide dev-only features in production, clean up homepage by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3157
* feat(ui): user-friendly tool and agent labels in the chat tool timeline by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3054
* feat(ui): add Routines page — user-friendly scheduled tasks view in https://github.com/tinyhumansai/openhuman/pull/2858
* feat(events): live domain event stream log panel by @Sathvik-1007 in https://github.com/tinyhumansai/openhuman/pull/2653

**Settings**
* feat(settings): Agents settings hub + dedicated agent editor by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3005
* feat(settings): Crypto + Notifications hubs, wallet send/receive & multi-network balances by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3027
* feat(settings): surface approval history audit trail by @obchain in https://github.com/tinyhumansai/openhuman/pull/2949
* feat(settings): make autocomplete tuning fields editable by @obchain in https://github.com/tinyhumansai/openhuman/pull/2973
* feat(settings): add Dev Workflow config panel by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2703
* feat(config): add UI control for agent action timeout (#3100) by @senamakel-droid in https://github.com/tinyhumansai/openhuman/pull/3112
* feat(cost): add settings cost dashboard with global tracker, dashboard RPCs, and charts by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2762

**Wallet & Web3**
* feat(web3): add swap/bridge/dapp module; slim wallet to primitives by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3006
* feat(wallet): surface multi-chain balances in Settings by @obchain in https://github.com/tinyhumansai/openhuman/pull/2686
* feat(wallet): bind prepared transaction quotes to originating chat session in https://github.com/tinyhumansai/openhuman/pull/2688

**Cron & Skills**
* feat(cron): complete smart scheduling flow from chat + polish cron management UI by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2943
* feat(skills): scheduled dashboard + run/new pages + [github] preflight gate + composio-only GitHub I/O by @M3gA-Mind in https://github.com/tinyhumansai/openhuman/pull/2882
* feat(skills,tools): schema↔serde roundtrip tests, triggers field, bus subscriber, registry integrity by @zahica1234 in https://github.com/tinyhumansai/openhuman/pull/2718

**Voice & Mascot**
* feat(voice): polish mascot TTS/STT session states and lifecycle by @staimoorulhassan in https://github.com/tinyhumansai/openhuman/pull/2916
* feat(mascot): add emotional reaction layer for conversation state by @staimoorulhassan in https://github.com/tinyhumansai/openhuman/pull/2900

**Other Features**
* feat(approval): per-launch UUID session_id + scrub legacy bearer leaks by @oxoxDev in https://github.com/tinyhumansai/openhuman/pull/2952
* feat(core): pass in-process RPC bearer via internal handle, not process env by @oxoxDev in https://github.com/tinyhumansai/openhuman/pull/2709
* feat(artifacts): define artifact storage types, config, and persistence layer by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2801
* feat(vault): expose approved markdown writes by @Zavianx in https://github.com/tinyhumansai/openhuman/pull/2936
* feat(meet-agent): Flow A second brain — orchestrator + tools + pre-roll + cache + barge-in in https://github.com/tinyhumansai/openhuman/pull/2861
* feat(tauri): OPENHUMAN_FORCE_GPU env override for Linux CEF disable by @zahica1234 in https://github.com/tinyhumansai/openhuman/pull/2862
* feat(review): default agent shortcuts to skip-permissions mode by @sanil-23 in https://github.com/tinyhumansai/openhuman/pull/2847
* feat(about_app): catalog entries for embedding provider selection (#2583 follow-up) by @justinhsu1477 in https://github.com/tinyhumansai/openhuman/pull/2656
* feat(tokenjuice): add Rust toolchain reduction rules by @ARYAN007H in https://github.com/tinyhumansai/openhuman/pull/2963
* feat: prompt user consent when OS keyring is unavailable by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3075
* feat(composio): migrate GitHub provider to memory_tree pipeline (#2885) by @justinhsu1477 in https://github.com/tinyhumansai/openhuman/pull/2889
* feat(composio): migrate Notion provider to memory_tree pipeline (#2885) by @justinhsu1477 in https://github.com/tinyhumansai/openhuman/pull/2887
* feat(memory-sync): pause Composio periodic tick when Memory Tree is toggled off (#2719 follow-up) by @justinhsu1477 in https://github.com/tinyhumansai/openhuman/pull/2825
* feat(gameplay): AI gameplay review and highlight workflows for streamers (#1674) by @sanil-23 in https://github.com/tinyhumansai/openhuman/pull/2856

### 🐛 Bug Fixes

**Auth & Login**
* fix(auth): resolve login redirect loop on first login by @senamakel-droid in https://github.com/tinyhumansai/openhuman/pull/3109
* fix(auth): gracefully drop OAuth profiles with missing access_token by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3125
* fix(oauth): extend loopback listener lifetime 60s -> 300s by @sanil-23 in https://github.com/tinyhumansai/openhuman/pull/3145
* fix(socket): refresh token before reconnect, fast-fail on Invalid token (#2892) by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2905
* fix(socket): recover from stale disconnected socket to prevent permanent "Connecting..." by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2487
* fix(provider): fallback to config.api_key when auth-profiles.json has no key by @ccpty in https://github.com/tinyhumansai/openhuman/pull/2724

**Inference**
* fix(inference): accept `reasoning` field alias for thinking-mode providers by @senamakel-droid in https://github.com/tinyhumansai/openhuman/pull/3124
* fix(inference): preserve reasoning_content in multi-turn thinking model conversations by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2818
* fix(inference): replay deepseek reasoning_content on tool-call turns (Sentry TAURI-RUST-4KB) by @M3gA-Mind in https://github.com/tinyhumansai/openhuman/pull/2876
* fix(agent): replay reasoning_content across native tool-call turns to prevent DeepSeek 400s by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2918
* fix(inference): disable Responses API fallback for ollama/lmstudio — stops 404 Sentry noise by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2902
* fix(inference): default empty model to reasoning-v1 on OpenHuman backend by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2837
* fix(inference): include model field in nvidia-nim API requests by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2791
* fix(inference): fail-fast on empty model for cloud providers, demote nvidia-nim Sentry noise by @staimoorulhassan in https://github.com/tinyhumansai/openhuman/pull/2811
* fix(inference): return empty model list on 404 from /models endpoint by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2939
* fix(inference): distinguish missing-data-field from wrong-type in /models parser by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2794
* fix(providers): dedup tool specs at wire boundary to prevent 400 "Tool names must be unique" by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2846
* fix(providers): include body snippet on list_models JSON parse failure by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2838
* fix(agent): drop bisected tool_call/tool_result pairs before sending to provider by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2840

**Tools & RPC**
* fix(tools): additive-safe tool filter prevents stale snapshots from stripping cron tools by @senamakel-droid in https://github.com/tinyhumansai/openhuman/pull/3118
* fix(rpc): register missing MCP and tool registry RPC handlers by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2803
* fix(rpc): register openhuman.system_info and add legacy alias by @M3gA-Mind in https://github.com/tinyhumansai/openhuman/pull/2877
* fix(rpc): add health_snapshot legacy alias by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2853
* fix(legacy_aliases): tolerate unquoted keys + skip comments in alias-table parser by @sanil-23 in https://github.com/tinyhumansai/openhuman/pull/2865

**Memory & Composio**
* fix(memory): stamp memory freshness so stale context is never served as current by @sanil-23 in https://github.com/tinyhumansai/openhuman/pull/2979
* fix(memory): scope Gmail cleanup to disconnected connection by @Felyx-Fu in https://github.com/tinyhumansai/openhuman/pull/2698
* fix(memory): dedupe Composio sources to the latest authorization per identity by @sanil-23 in https://github.com/tinyhumansai/openhuman/pull/2843
* fix(memory-conversations): release STORE_LOCK during cold index rebuild (#2849) by @staimoorulhassan in https://github.com/tinyhumansai/openhuman/pull/2881
* fix(conversations): release store lock during cold index rebuild (#2849) by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2886
* fix(memory-sync): migrate Linear provider to memory_tree (#2885) by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/3018
* fix(vault): sync writes to memory_tree, not legacy UnifiedMemory by @justinhsu1477 in https://github.com/tinyhumansai/openhuman/pull/2720
* fix(composio): update GitHub action slugs to current values by @staimoorulhassan in https://github.com/tinyhumansai/openhuman/pull/2835
* fix(composio): forward tags filter in direct (BYO-key) tool listing by @sanil-23 in https://github.com/tinyhumansai/openhuman/pull/2772
* fix(composio): map trigger-enable 403 to actionable error by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/3009
* fix(composio): clarify Zoho Mail readiness by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/3013
* fix(composio): log daily budget reset by @sunilkumarvalmiki in https://github.com/tinyhumansai/openhuman/pull/2672

**Observability (Sentry noise reduction)**
* fix(observability): demote expected-error Sentry buckets across embeddings, provider, memory-store, FS, and thinking-mode wire shapes by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2830
* fix(observability): suppress 429/rate-limit Sentry noise — TAURI-RUST-3 by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2899
* fix(observability): suppress context-window-exceeded provider error from Sentry by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2820
* fix(observability): demote backend Cloudflare anti-bot wrap by @oxoxDev in https://github.com/tinyhumansai/openhuman/pull/2692
* fix(observability): classify 'operation timed out' transport phrase as expected by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2782
* fix(observability): demote unknown-provider list_models error by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2783
* fix(observability): classify list_models 404 as ProviderUserState by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2793
* fix(observability): classify 'does not support tools' provider 400 as user-state by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2796
* fix(observability): demote reliable_chat all_exhausted aggregate as ProviderConfigRejection by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2797
* fix(observability): classify OpenHuman/Embedding/streaming backend 'Invalid token' 401 as SessionExpired by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2786
* fix(observability): classify web-channel empty-provider-response re-report as expected by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2808
* fix(observability): classify DeepSeek 'Insufficient Balance' 402 as user-state by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2809
* fix(observability): classify RPC filesystem path validation errors as expected by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2795
* fix(observability): classify WS protocol HTTP-version handshake error as expected by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2792
* fix(observability): classify list_models 404 as ProviderUserState by @M3gA-Mind in https://github.com/tinyhumansai/openhuman/pull/2873
* fix(observability): classify embedding API 401 'Invalid token' as SessionExpired by @sanil-23 in https://github.com/tinyhumansai/openhuman/pull/2869
* fix(observability): suppress VOYAGE_API_KEY not configured Sentry noise by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2915
* fix(observability): demote WhatsApp ingest SQLite lock noise by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2917
* fix(observability): demote "Config loading timed out" out of Sentry by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2920
* fix(observability): demote ollama daemon-unreachable to expected error by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2923
* fix(observability): suppress billing 401 Sentry noise by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2924
* fix(observability): demote expected chat failures out of Sentry by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2928
* fix(observability): suppress composio-direct HTTP 403 Sentry noise by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2930
* fix(observability): suppress embeddings API-key 401 Sentry noise by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2934
* fix(observability): extend is_budget_event to text-only tier-2 match by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2937
* fix(observability): demote memory-store PII rejection errors from Sentry to warn by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2850
* fix(observability): demote channel supervisor restart noise by @M3gA-Mind in https://github.com/tinyhumansai/openhuman/pull/2879
* fix(agent): suppress empty-provider-response from Sentry by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2790
* fix(backend_api): suppress Sentry on 401 via typed BackendApiError::Unauthorized by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2781
* fix(composio/triage): demote provider-config-rejection rollups by @oxoxDev in https://github.com/tinyhumansai/openhuman/pull/2689
* fix(subconscious): suppress repeated Sentry noise when DB init fails by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2906
* fix(inference): suppress spurious Sentry events for Ollama tool-schema rejection 400s by @staimoorulhassan in https://github.com/tinyhumansai/openhuman/pull/2812
* fix(inference): suppress Sentry noise when ollama model doesn't support tools by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2813
* fix(inference): publish SessionExpired for backend 401 on chat_completions by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2814
* fix: resolve Sentry issue 526 by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2815
* fix: resolve Sentry issue 5677 by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2810
* fix(embeddings): add 429 backoff to stop 31K Sentry event flood by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2904
* fix(subconscious): preserve anyhow chain + retry DDL with busy_timeout on SQLITE_BUSY by @oxoxDev in https://github.com/tinyhumansai/openhuman/pull/2851

**Platform & Core**
* fix(core): fall back to other ports when preferred is OS-excluded (Windows WSAEACCES) by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2816
* fix(config): retry transient config-read race on Windows by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2807
* fix(tauri): verify openhuman:// registry registration on Windows by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2757
* fix(cwd_jail/windows): wire AppContainer network capabilities for jail.allow_net by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2637
* fix(appimage): cd "$APPDIR" in AppRun before exec to fix sharun preload CWD resolution by @staimoorulhassan in https://github.com/tinyhumansai/openhuman/pull/2829
* fix(reset): release log file handle + queue locked entries on Windows reboot by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2668
* fix(bootcheck): warn instead of block on public HTTP core URLs by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3031

**UI & Chat**
* fix(chat): surface real inference error detail instead of generic apology by @senamakel-droid in https://github.com/tinyhumansai/openhuman/pull/3033
* fix(chat): rearm safety timer on sub-agent tool / task-board updates by @zahica1234 in https://github.com/tinyhumansai/openhuman/pull/2857
* fix(channels): render Lark/DingTalk/iMessage logos + add connect forms by @EvanCarson in https://github.com/tinyhumansai/openhuman/pull/2859
* fix(settings): Background Loops page layout overflow and duplicate label by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2693
* fix: clear stale backend errors while reconnecting by @shriii257 in https://github.com/tinyhumansai/openhuman/pull/2616
* fix: add Open button to knowledge vault rows with Obsidian deep link fallback by @HariNayan in https://github.com/tinyhumansai/openhuman/pull/3014
* fix(voice): interrupt mascot speech when listening starts by @sunilkumarvalmiki in https://github.com/tinyhumansai/openhuman/pull/2678
* fix(voice): make Whisper/Piper chips reachable from Voice settings by @obchain in https://github.com/tinyhumansai/openhuman/pull/2821
* fix(artifacts): harden artifact ID validation and absolute_path by @M3gA-Mind in https://github.com/tinyhumansai/openhuman/pull/2860
* fix(local-ai): unblock summary diagnostics by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2940
* fix(yuanbao): correct AuthBind plugin_version / bot_version mapping by @lrt4836 in https://github.com/tinyhumansai/openhuman/pull/2804
* fix(cron): accept bare cron-expression string in Schedule deserializer by @M3gA-Mind in https://github.com/tinyhumansai/openhuman/pull/2874
* fix(embeddings): recover from Ollama NaN-encoding 500 instead of failing whole batch by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2834
* mergeTooShort() never merges a short first segment by @shriii257 in https://github.com/tinyhumansai/openhuman/pull/2615

**i18n**
* fix(i18n): localize ChatGPT OAuth strings with real translations by @staimoorulhassan in https://github.com/tinyhumansai/openhuman/pull/3021
* fix(i18n): localize MicComposer getUserMedia error messages by @obchain in https://github.com/tinyhumansai/openhuman/pull/2966
* fix(i18n): correct key/value semantic mismatch in chunk-5 locale files by @graycyrus in https://github.com/tinyhumansai/openhuman/pull/2765
* fix(i18n): translate locale fallback strings by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3004
* i18n: translate chat.attachment.* keys across all locales by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2888

### ⚡ Performance

* perf(memory_tree): batch drill_down BFS walker per depth by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/3032
* perf(memory_tree): batch embedding fetch in topic rerank by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/2927
* perf(memory_tree): batch chunk + score fetch in fetch_leaves by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/2941
* perf(memory_tree): batch summary fetch in hydrate_summary_inputs by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/2948
* perf(memory_tree): batch tree fetch in flush_stale_buffers by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/2976
* perf(prompt_injection): batch detection rules via RegexSet, drop dead compact scan by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/2842
* perf(routing/quality): batch refusal detection via Aho-Corasick DFA by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/2870
* perf(learning/user_profile): Aho-Corasick DFA for preference extraction + pattern coverage by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/2878
* perf(security): cache canonical workspace_dir to skip per-call canonicalize syscall by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/2798
* perf(tokenjuice): hoist gh table split regex to a Lazy<Regex> static by @mysma-9403 in https://github.com/tinyhumansai/openhuman/pull/2770

### 🔧 Refactors

* refactor(agent): unify the three agent-turn loops into one TurnEngine by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3012
* refactor(memory): remove global+topic trees in favor of source trees by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3059
* refactor(tools): move domain-owned tools into modules by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2977
* refactor(openhuman): codify canonical module shape, tidy offenders by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2986
* refactor(rust): split oversized modules by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2991
* refactor(rust): move large inline tests to files by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2987
* refactor(react): extract repeated UI primitives by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3003
* refactor(tools): update GitHub tool slugs for clarity and remove deprecated entry by @YellowSnnowmann in https://github.com/tinyhumansai/openhuman/pull/2766
* refactor(memory): route Obsidian deep link through workspace-link layer by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2702
* Revert gameplay-specific review workflows by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2962

### 🧪 Tests

* test: fix flaky/stale tests blocking Rust E2E + coverage CI on main by @sanil-23 in https://github.com/tinyhumansai/openhuman/pull/3147
* test: expand e2e raw coverage across agent, channels, inference, memory, tools by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3023
* test(security): cover env-overlay path in load_and_get_security_policy_info by @obchain in https://github.com/tinyhumansai/openhuman/pull/2695
* test(memory): avoid PII-like tool rule names by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2680
* fix(test): correct round21 summarize assertion to expect real newlines by @senamakel-droid in https://github.com/tinyhumansai/openhuman/pull/3038
* fix(memory_tree): code quality improvements on batch tree fetch (PR #2976 follow-up) by @staimoorulhassan in https://github.com/tinyhumansai/openhuman/pull/3020

### 📦 CI & Infra

* ci(release): add linux arm64 desktop artifacts by @sunilkumarvalmiki in https://github.com/tinyhumansai/openhuman/pull/2675
* ci(release): gate main_head push to main behind manual approval by @senamakel-droid in https://github.com/tinyhumansai/openhuman/pull/3045
* ci: gate PRs on i18n key parity by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3011
* ci: consolidate pull request checks by @senamakel in https://github.com/tinyhumansai/openhuman/pull/3001
* ci: extend Windows secrets ACL timeout by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2654

### 📖 Docs

* docs: truth-up architecture.md + 3 new domain pages + Linux/Arch caveats in localized READMEs by @aashir-athar in https://github.com/tinyhumansai/openhuman/pull/2634
* docs(openhuman): add per-module READMEs for domain modules by @senamakel in https://github.com/tinyhumansai/openhuman/pull/2992
* docs(spec): telegram remote-control phase 2 — inline approvals by @CodeGhost21 in https://github.com/tinyhumansai/openhuman/pull/2502
* docs(prompts): add zero-match search guardrail to SOUL.md/Personality by @zahica1234 in https://github.com/tinyhumansai/openhuman/pull/2855
* docs(memory-tree): update stale comment on translate_query_global_args by @zahica1234 in https://github.com/tinyhumansai/openhuman/pull/2713
* docs: follow up on #2636 — smoke line + DANGEROUS_ENV_PREFIXES cross-ref by @YOMXXX in https://github.com/tinyhumansai/openhuman/pull/2701

**Full Changelog**: https://github.com/tinyhumansai/openhuman/compare/v0.56.0...v0.57.5