openclaw/appcast.xml

<?xml version="1.0" standalone="yes"?>
<rss xmlns:sparkle="http://www.andymatuschak.org/xml-namespaces/sparkle" version="2.0">
    <channel>
        <title>OpenClaw</title>
        <item>
            <title>2026.4.25</title>
            <pubDate>Mon, 27 Apr 2026 13:34:25 +0000</pubDate>
            <link>https://raw.githubusercontent.com/openclaw/openclaw/main/appcast.xml</link>
            <sparkle:version>2026042590</sparkle:version>
            <sparkle:shortVersionString>2026.4.25</sparkle:shortVersionString>
            <sparkle:minimumSystemVersion>15.0</sparkle:minimumSystemVersion>
            <description><![CDATA[<h2>OpenClaw 2026.4.25</h2>
<h3>Highlights</h3>
<ul>
<li>Voice replies get a full TTS upgrade: <code>/tts latest</code>, chat-scoped auto-TTS controls, personas, per-agent/per-account overrides, and new Azure Speech, Xiaomi, Local CLI, Inworld, Volcengine, and ElevenLabs v3 provider coverage. Thanks @leonchui, @zoujiejun, @solar2ain, @cshape, @xuruiray, @itsuzef, and @barronlroth.</li>
<li>Plugin startup and install paths move to the cold persisted registry, cutting broad manifest scans while making plugin update, repair, provider discovery, and install metadata more deterministic. Thanks @vincentkoc and @shakkernerd.</li>
<li>OpenTelemetry coverage expands across model calls, token usage, tool loops, harness runs, exec processes, outbound delivery, context assembly, and memory pressure with bounded low-cardinality attributes. Thanks @vincentkoc, @jlapenna, @Lidang-Jiang, and @oc-factus.</li>
<li>Browser automation gets safer tab URLs, iframe-aware role snapshots, CDP readiness tuning, headless one-shot launch, and deeper browser doctor probes for slow hosts. Thanks @beat843796 and @BenediktSchackenberg.</li>
<li>Control UI and setup flows add PWA/Web Push support, Crestodian first-run repair, TUI setup, context mode selection, and a shorter startup greeting. Thanks @eduardocruz, @SebTardif, and @kevinlin-openai.</li>
<li>Install/update hardening covers Windows, macOS, Linux, Docker, bundled plugin runtime deps, Node service restarts, LaunchAgent token rotation, and mixed-version gateway verification. Thanks @Kobevictor, @igormf, @abhinas90, @jsompis, @Solvely-Colin, and @gucasbrg.</li>
</ul>
<h3>Changes</h3>
<ul>
<li>TTS/WhatsApp: add <code>/tts latest</code> read-aloud support with duplicate suppression and <code>/tts chat on|off|default</code> session-scoped auto-TTS overrides, completing the on-demand voice-note UX for current-chat replies. Fixes #66032.</li>
<li>TTS/channels: resolve channel and account TTS overrides generically, enabling Feishu and QQBot accounts to deep-merge <code>channels.<channel>.accounts.<id>.tts</code> over global and per-agent TTS config. Thanks @sahilsatralkar.</li>
<li>TTS/agents: allow <code>agents.list[].tts</code> to override global <code>messages.tts</code> for per-agent voices, and make <code>/tts audio</code>, <code>/tts status</code>, and the <code>tts</code> agent tool honor the active voice/provider override while keeping shared provider credentials and preferences in the existing TTS config surface.</li>
<li>Providers/Azure Speech: add Azure Speech as a bundled TTS provider with Speech-resource auth, voice listing, SSML escaping, native Ogg/Opus voice-note output, and telephony output. (#51776) Thanks @leonchui.</li>
<li>Google Meet: add calendar-backed attendance export workflows, export manifests, dry-run previews, and tool parity for meeting records.</li>
<li>Control UI: add PWA install support and Web Push notifications for Gateway chat. (#44590) Thanks @eduardocruz.</li>
<li>Browser automation: add safe tab URLs in agent responses plus a CDP-native role snapshot fallback with iframe-aware refs, cursor-clickable detection, target attach preparation, and <code>openclaw browser doctor --deep</code> live snapshot probing.</li>
<li>CLI/image generation: expose generic <code>--background</code> on <code>openclaw infer image generate</code> and <code>openclaw infer image edit</code>, keep <code>--openai-background</code> as an OpenAI alias, and let fal image generation honor <code>--output-format png|jpeg</code>.</li>
<li>Browser/config: allow local managed Chrome launch discovery and post-launch CDP readiness timeouts to be raised for slower hosts such as Raspberry Pi. Fixes #66803. Thanks @beat843796.</li>
<li>Discord: allow <code>channels.discord.voice.model</code> to override the LLM used for voice channel responses while keeping STT and TTS on their existing media settings. (#64368) Thanks @mrdavey.</li>
<li>Browser/CLI: add <code>openclaw browser start --headless</code> as a one-shot local managed browser launch override without rewriting persisted browser config. Thanks @BenediktSchackenberg.</li>
<li>CLI/Crestodian/TUI: add the first-run setup helper, local planner fallback, full-TUI interactive Crestodian, startup progress indicators, context mode selector, and a shorter startup greeting. (#71720, #71760) Thanks @SebTardif and @kevinlin-openai.</li>
<li>Plugins: migrate the local plugin registry automatically during package install/update, keeping install metadata in the plugin index while indexing existing plugin manifests for the new cold registry path. Thanks @vincentkoc and @shakkernerd.</li>
<li>Plugins/doctor: make <code>openclaw doctor --fix</code> refresh the plugin index and cold registry index when needed without treating plugin install records as authored config. Thanks @vincentkoc and @shakkernerd.</li>
<li>Plugins/hooks: add before-agent-finalize hooks, cron <code>jobId</code> hook context, bounded native permission fingerprints, and Codex MCP hook relay support. (#71765, #71758, #71707) Thanks @vincentkoc and @pashpashpash.</li>
<li>Plugins/tokenjuice: bump the bundled tokenjuice runtime to 0.6.3. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: align model-call GenAI span attributes with OpenTelemetry stability opt-in semantics, keeping legacy <code>gen_ai.system</code> by default while emitting <code>gen_ai.provider.name</code> under <code>OTEL_SEMCONV_STABILITY_OPT_IN=gen_ai_latest_experimental</code>. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: support signal-specific OTLP endpoint overrides for traces, metrics, and logs via config or standard OTEL environment variables. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: emit bounded telemetry exporter health diagnostics for startup and log-export failures without exporting raw error text. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: export agent harness lifecycle telemetry as bounded <code>openclaw.harness.run</code> spans and <code>openclaw.harness.duration_ms</code> metrics so QA-lab, Codex, and future harnesses share one trace shape. Thanks @vincentkoc.</li>
<li>Diagnostics/trace: propagate W3C <code>traceparent</code> headers from trusted model-call trace context to provider transports while replacing caller-supplied traceparent values. Thanks @vincentkoc.</li>
<li>Diagnostics/Prometheus: add a bundled <code>diagnostics-prometheus</code> plugin with a protected gateway scrape route for low-cardinality diagnostics metrics. Thanks @vincentkoc.</li>
<li>Plugins/CLI: add <code>openclaw plugins registry</code> for explicit persisted-registry inspection and <code>--refresh</code> repair without making normal startup rescan plugin locations. Thanks @vincentkoc.</li>
<li>Plugins/CLI: make <code>openclaw plugins list</code> read the cold persisted registry snapshot by default, leaving module-aware diagnostics to <code>plugins doctor</code> and <code>plugins inspect</code>. Thanks @vincentkoc.</li>
<li>Plugins/startup: move gateway startup plugin planning onto the versioned cold registry index, with postinstall repair for older registry files that predate startup metadata. Thanks @vincentkoc.</li>
<li>Plugins/startup: normalize startup and provider plugin enablement through registry aliases so boot paths do not need the legacy manifest alias scan. Thanks @vincentkoc.</li>
<li>Providers/plugins: resolve provider ownership, provider discovery scopes, and catalog-hook provider ids from the cold plugin registry instead of rescanning manifests on those paths. Thanks @vincentkoc.</li>
<li>Plugins/registry: keep installed plugin index records focused on install/state/load paths and resolve plugin capabilities from manifests scoped to indexed plugins. Thanks @shakkernerd.</li>
<li>Plugins/registry: route cold manifest and capability lookups through the installed plugin index so setup, channels, config, secrets, doctor, and provider metadata paths avoid broad plugin-root scans before runtime execution. Thanks @shakkernerd.</li>
<li>CLI/models: speed up <code>models list --all --provider <id></code> for static manifest-backed providers by loading catalog rows through the installed plugin index instead of broad manifest scans or runtime suppression hooks. Thanks @shakkernerd.</li>
<li>CLI/models: use OpenClaw Provider Index preview rows as the final cold fallback for installable providers, while keeping user config, installed manifests, and refreshed cache rows above provider-index metadata. Thanks @vincentkoc.</li>
<li>Providers/plugins: keep onboarding and auth-choice setup lists on cold manifest/install metadata and add Provider Index install metadata for not-yet-installed provider plugins. Thanks @vincentkoc.</li>
<li>Providers/plugins: keep provider setup guidance and configure auth imports on cold manifest metadata, with a regression guard against static provider-runtime imports on setup/configure list paths. Thanks @vincentkoc.</li>
<li>CLI/capabilities: keep capability command registration from importing the models auth runtime until <code>model auth login</code> actually runs. Thanks @vincentkoc.</li>
<li>CLI/configure: keep web-search configure prompts on cold plugin registry metadata until the user chooses managed search setup. Thanks @vincentkoc.</li>
<li>Plugins/chat commands: refresh the persisted plugin registry after <code>/plugins enable</code> and <code>/plugins disable</code>, matching the CLI mutation path. Thanks @vincentkoc.</li>
<li>Plugins/compat: mark <code>OPENCLAW_DISABLE_PERSISTED_PLUGIN_REGISTRY</code> as a deprecated break-glass switch and point operators at registry repair instead. Thanks @vincentkoc.</li>
<li>Plugins/compat: expand the central compatibility registry with dated owners, replacements, and maximum three-month removal targets for legacy SDK, manifest, setup, registry-migration, and agent-runtime surfaces. Thanks @vincentkoc.</li>
<li>Plugins/registry: ignore stale persisted registry reads when plugin policy no longer matches current config, and stamp generated registry files with a do-not-edit warning. Thanks @vincentkoc.</li>
<li>Config/plugins: keep plugin command-alias validation on cold manifest metadata instead of importing the runtime alias resolver. Thanks @vincentkoc.</li>
<li>Security/plugins: keep web-search credential presence checks on cold config, env, and manifest metadata instead of importing web-search provider runtime. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: surface provider request identifiers as bounded hashes on model-call diagnostics and span events, without exporting raw request IDs or metric labels. Thanks @Lidang-Jiang and @vincentkoc.</li>
<li>Plugins/diagnostics: add metadata-only <code>model_call_started</code> and <code>model_call_ended</code> hooks for provider/model call telemetry without exposing prompts, responses, headers, request bodies, or raw provider request IDs. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: emit bounded context assembly diagnostics and export <code>openclaw.context.assembled</code> spans with prompt/history sizes but no prompt, history, response, or session-key content. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: export existing tool-loop diagnostics as <code>openclaw.tool.loop</code> counters and spans without loop messages, session identifiers, params, or tool output. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: export diagnostic memory samples and pressure as bounded memory histograms, counters, and pressure spans to help spot leak regressions without session or payload data. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: add the GenAI <code>gen_ai.client.token.usage</code> histogram for input/output model usage while keeping session identifiers and aggregate cache counters out of the semantic metric. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: add a bounded <code>openclaw.agent</code> label to OpenClaw token metrics so per-agent Grafana dashboards can group usage without exporting session identifiers. Thanks @oc-factus.</li>
<li>Plugins/install: consolidate managed plugin install metadata into the state-managed plugin index at <code>plugins/installs.json</code>, replacing the temporary <code>plugins/installed-index.json</code> path and removing <code>plugins.installs</code> as an authored config surface. Thanks @vincentkoc and @shakkernerd.</li>
<li>Diagnostics/OTEL: add the GenAI <code>gen_ai.client.operation.duration</code> histogram for model-call latency in seconds with bounded provider/model/API and error attributes. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: add GenAI usage token attributes to model-usage spans, including cache read/write input token counts without session identifiers or prompt/response content. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: include bounded GenAI operation, provider, and request-model attributes on model-usage spans so token usage remains self-describing without diagnostic identifiers. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: keep model-usage span GenAI provider attributes aligned with the existing semantic-convention opt-in policy, using legacy <code>gen_ai.system</code> unless latest experimental GenAI conventions are enabled. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: keep <code>gen_ai.request.model</code> present on GenAI token usage metrics with a bounded <code>unknown</code> fallback when model usage events do not include a model. Thanks @vincentkoc.</li>
<li>Docs/OTEL: document the GenAI token and model-call duration metrics, model-usage span attributes, and <code>OTEL_SEMCONV_STABILITY_OPT_IN=gen_ai_latest_experimental</code> provider-attribute behavior. Thanks @vincentkoc.</li>
<li>Docs: refresh the MCP, model provider, doctor, troubleshooting, BlueBubbles, media generation, TTS, subagents, skills, cron/tasks, exec approvals, and voice-call guides with structured Steps, Tabs, and Accordion content.</li>
<li>Diagnostics/trace: add an internal traceparent propagation helper that only formats trusted dispatcher metadata, keeping plugin-emitted diagnostic traces out of outbound propagation by default. Thanks @vincentkoc.</li>
<li>Diagnostics/OTEL: add bounded outbound message delivery lifecycle diagnostics and export them as low-cardinality delivery spans/metrics without message body, recipient, room, or media-path data. (#71471) Thanks @vincentkoc and @jlapenna.</li>
<li>Diagnostics/OTEL: emit bounded exec-process diagnostics and export them as <code>openclaw.exec</code> spans without exposing command text, working directories, or container identifiers. (#71451) Thanks @vincentkoc and @jlapenna.</li>
<li>Diagnostics/OTEL: support <code>OPENCLAW_OTEL_PRELOADED=1</code> so the plugin can reuse an already-registered OpenTelemetry SDK while keeping OpenClaw diagnostic listeners wired. (#71450) Thanks @vincentkoc and @jlapenna.</li>
<li>Providers/Xiaomi: add MiMo TTS as a bundled speech provider with MP3/WAV output and voice-note Opus transcoding. Fixes #52376. (#55614) Thanks @zoujiejun.</li>
<li>Providers/ElevenLabs: include <code>eleven_v3</code> in the bundled TTS model catalog so model selection surfaces can offer ElevenLabs v3. (#68321) Thanks @itsuzef.</li>
<li>Providers/Local CLI TTS: add a bundled local command speech provider with file/stdout input, voice-note Opus conversion, and telephony PCM output. (#56239) Thanks @solar2ain.</li>
<li>Providers/Inworld: add Inworld as a bundled speech provider with streaming TTS synthesis, voice listing, voice-note output, and PCM telephony output. (#55972) Thanks @cshape.</li>
<li>Providers/Volcengine: add Volcengine/BytePlus Seed Speech as a bundled TTS provider with API-key auth, native Ogg/Opus voice-note output, and MP3 audio-file output. (#55641) Thanks @xuruiray.</li>
<li>Android/Talk Mode: expose Talk Mode in the Voice tab with runtime-owned voice capture modes and microphone foreground-service escalation. Thanks @alex-latitude.</li>
<li>Providers/LiteLLM: register <code>litellm</code> as an image-generation provider so <code>image_generate model=litellm/...</code> calls and <code>agents.defaults.imageGenerationModel.fallbacks</code> entries resolve through the LiteLLM proxy. Thanks @zqchris.</li>
<li>Providers/fal: add Seedance 2.0 reference-to-video models with multi-image, video, and audio reference input mapping plus model-specific capability limits for <code>video_generate</code>. Thanks @shivanker.</li>
<li>Codex harness: require Codex app-server <code>0.125.0</code> or newer and cover native MCP <code>PreToolUse</code>, <code>PostToolUse</code>, and <code>PermissionRequest</code> payloads through the OpenClaw hook relay.</li>
<li>Agents/Codex: teach prompts and <code>agents_list</code> to surface native Codex app-server availability so agents prefer <code>/codex ...</code> over Codex ACP unless ACP/acpx is explicit. Thanks @vincentkoc.</li>
<li>ACPX/Droid: add Factory Droid to the live ACP bind Docker matrix, including <code>.factory</code> settings staging, <code>FACTORY_API_KEY</code> forwarding, and the single-agent <code>test:docker:live-acp-bind:droid</code> recipe.</li>
<li>TTS/personas: add provider-aware TTS personas with deterministic provider binding merges, <code>/tts persona</code> controls, gateway/CLI persona state, Google Gemini <code>audio-profile-v1</code> prompt wrapping, and OpenAI instruction mapping. (#70748) Thanks @barronlroth.</li>
<li>Voice Wake: add trigger-based routing so macOS voice wake phrases can select a configured agent or session target, with Gateway routing APIs and node update events. (#30354) Thanks @longbiaochen.</li>
</ul>
<h3>Fixes</h3>
<ul>
<li>Auto-reply: poison inbound message dedupe after replay-unsafe provider/runtime failures so retries stay safe before visible progress but cannot duplicate messages after block output, tool side effects, or session progress. Fixes #69303; keeps #58549 and #64606 as duplicate validation. Thanks @martingarramon, @NikolaFC, and @zeroth-blip.</li>
<li>Logging/sessions: apply configured redaction patterns to persisted session transcript text and accept escaped character classes in safe custom redaction regexes, so transcript JSONL no longer keeps matching sensitive text in the clear. Fixes #42982. Thanks @panpan0000.</li>
<li>Agents/OpenAI: keep Responses web search compatible with minimal thinking by raising <code>web_search</code> requests to the lowest supported reasoning effort instead of sending a rejected minimal payload.</li>
<li>Agents/tools: honor the <code>bundle-mcp</code> allowlist token when deciding whether bundled MCP tools are available, so restricted tool policies can still enable bundled MCP without exposing unrelated tools.</li>
<li>Agents/model fallback: jump directly to a known later live-session model redirect instead of walking unrelated fallback candidates, while preserving the already-landed live-session/fallback loop guard. Fixes #57471; related loop family already closed via #58496. Thanks @yuxiaoyang2007-prog.</li>
<li>Skills/memory: restore Chokidar v5 hot reloads by watching concrete skill and memory roots with filters, including SKILL.md removals and deleted skill folders without broad workspace recursion. Fixes #27404, #33585, and #41606. Thanks @shelvenzhou, @08820048, and @rocke2020.</li>
<li>Discord/gateway: count failed health-monitor restart attempts toward cooldown and hourly caps, and evict stale account lifecycle state during channel reloads so repeated Discord gateway recovery cannot loop on old status. Fixes #38596. (#40413) Thanks @jellyAI-dev and @vashquez.</li>
<li>Plugins/CLI: let flag-driven <code>openclaw channels add</code> install the selected channel plugin from its default source without opening an interactive prompt, fixing published npm Telegram setup in stdin-closed automation.</li>
<li>Plugins/startup: load the default <code>memory-core</code> slot during Gateway startup when permitted so active-memory recall can call <code>memory_search</code> and <code>memory_get</code> without requiring an explicit <code>plugins.slots.memory</code> entry, while preserving <code>plugins.slots.memory: "none"</code>.</li>
<li>Plugins/install: materialize plugin-owned root chunks in external bundled-runtime mirrors so staged plugin dependencies resolve under native ESM in packaged installs. Fixes #72058; supersedes #72084. Thanks @amnesia106 and @drvoss.</li>
<li>Plugins/CLI: prefer native require for compiled bundled plugin JavaScript before jiti so read-only config, status, device, and node commands avoid unnecessary transform overhead on slow hosts. Fixes #62842. Thanks @Effet.</li>
<li>Plugins/compat: inventory doctor-side deprecation migrations separately from runtime plugin compatibility so release sweeps preserve needed repairs while enforcing dated removal windows. Thanks @vincentkoc.</li>
<li>Plugins/compat: add missing dated compatibility records for legacy extension-api, memory registration, provider hook/type aliases, runtime aliases, channel SDK helpers, and approval/test utility shims. Thanks @vincentkoc.</li>
<li>Plugins/CLI: refresh the persisted registry after managed plugin files are removed so ClawHub uninstall cannot leave stale <code>plugins list</code> entries.</li>
<li>Plugins/CLI: make plugin install and uninstall config writes conflict-aware, clear stale denylist entries on explicit reinstall/removal, and delete managed plugin files only after config/index commit succeeds.</li>
<li>Plugins: fail <code>plugins update</code> when tracked plugin or hook updates error, keep bundled runtime-dependency repair behind restrictive allowlists, and reject package installs with unloadable extension entries.</li>
<li>Gateway/chat: keep duplicate attachment-backed <code>chat.send</code> retries with the same idempotency key on the documented in-flight path so aborts still target the real active run. Fixes #70139. Thanks @Feelw00.</li>
<li>Plugins: share package entrypoint resolution between install and discovery, reject mismatched <code>runtimeExtensions</code>, and cache bundled runtime-dependency manifest reads during scans.</li>
<li>WhatsApp/Web: keep quiet but healthy linked-device sessions connected by basing the watchdog on WhatsApp Web transport activity, while retaining a longer app-silence cap so frame activity cannot mask a stuck session forever. Fixes #70678; carries forward the focused #71466 approach and keeps #63939 as related configurable-timeout follow-up. Thanks @vincentkoc and @oromeis.</li>
<li>Onboarding/setup: keep first-run config reads, plugin compatibility notices, and post-model sanity checks on cold metadata paths unless the user chooses to browse all models, avoiding full plugin/runtime catalog work between prompts. Thanks @shakkernerd.</li>
<li>Onboarding/auth: run manifest-owned provider auth choices through scoped setup providers so selecting OpenAI Codex browser/device auth no longer loads every provider runtime before OAuth starts. Thanks @shakkernerd.</li>
<li>Onboarding/auth: keep the post-auth default-model policy lookup on manifest/setup metadata so the next prompt appears without loading broad provider runtime. Thanks @shakkernerd.</li>
<li>Onboarding/models: keep skip-auth and provider-scoped model picker prompts off the full global model catalog path, and cache provider catalog hook resolution so setup no longer stalls after auth on large plugin registries. Thanks @shakkernerd.</li>
<li>Gateway/Bonjour: suppress known @homebridge/ciao cancellation and network assertion failures through scoped process handlers so malformed mDNS packets or restricted VPS networking disable/restart Bonjour instead of crashing the gateway. Fixes #67578. Thanks @zenassist26-create.</li>
<li>Discord: keep late clicks on already-resolved exec approval buttons quiet when elevated mode auto-resolved the request, while still surfacing real approval submission failures. Fixes #66906. Thanks @rlerikse.</li>
<li>Agents/subagents: deliver completed yielded-subagent results back to no-thread requester routes via direct fallback when the dormant parent announce turn produces no visible reply, and add QA-lab coverage for the regression. Thanks @vincentkoc.</li>
<li>Gateway/Tailscale: let Tailscale-authenticated Control UI operator sessions with browser device identity skip the device-pairing round trip while still rejecting device-less and node-role connections. Refs #71986. Thanks @jokedul.</li>
<li>Doctor: honor <code>OPENCLAW_SERVICE_REPAIR_POLICY=external</code> by reporting gateway service health while skipping service install/start/restart/bootstrap, supervisor rewrites, and legacy service cleanup for externally managed environments. Thanks @shakkernerd.</li>
<li>CLI/update: run package post-update doctor with <code>--fix</code> so package updates repair config migrations before restart. Thanks @shakkernerd.</li>
<li>CLI/update: retry failed npm global updates with <code>--omit=optional</code> and ignore the superseded first failure when the fallback succeeds. Thanks @shakkernerd.</li>
<li>Plugins/uninstall: migrate and reset <code>plugins.slots.contextEngine</code> alongside memory slots when plugin ids change or selected plugins are removed. Thanks @shakkernerd.</li>
<li>Agents/Discord: keep raw <code>Agent failed before reply</code> runner failures out of Discord group/channel chats and show detailed runner errors in direct chats only when <code>/verbose</code> is enabled.</li>
<li>UI/Windows: quote resolved pnpm <code>.cmd</code> launcher paths before spawning UI install/build/test commands so Node installs under <code>C:\Program Files</code> no longer fail as <code>C:\Program</code>. Fixes #45275. Thanks @Kobevictor, @stoppieboy, and @iubns.</li>
<li>Codex/agent: translate <code>--thinking minimal</code> to <code>low</code> for modern Codex models (gpt-5.5, gpt-5.4, gpt-5.4-mini, gpt-5.2) at request build time so the first turn is accepted instead of paying a wasted call + retry-with-low fallback. Older Codex models still receive <code>minimal</code> directly. Fixes #71946. Thanks @hclsys.</li>
<li>Plugins/uninstall: remove tracked plugin files from their recorded managed extensions root even when the current state directory points somewhere else, so <code>openclaw plugins uninstall --force</code> does not leave the plugin discoverable. Thanks @shakkernerd.</li>
<li>Agents/runtime: add <code>agentRuntime.id</code> as the canonical config key, migrate legacy runtime-policy configs with <code>openclaw doctor --fix</code>, route canonical Anthropic models through <code>claude-cli</code> without passing CLI backend aliases to embedded harness selection, and load CLI backend owner plugins before channel startup. Fixes #71957. Thanks @WolvenRA.</li>
<li>CLI/update: guard Windows scheduled-task stops by state and timeout so auto-update restart cannot hang indefinitely on <code>schtasks /End</code> before stale-listener cleanup. Fixes #69970. Thanks @yangswld and @sherlock-huang.</li>
<li>Windows install/Lobster: execute <code>pnpm.exe</code> directly when <code>npm_execpath</code> points at the native pnpm binary, add an installed-package fallback for the Lobster embedded runtime, and include the Lobster runner regression test in Windows CI. Fixes #69456. Thanks @igormf.</li>
<li>Gateway/install: refresh loaded gateway service installs when the current service embeds stale gateway auth instead of returning already-installed, avoiding LaunchAgent token-mismatch loops after token rotation. Fixes #70752. Thanks @hyspacex.</li>
<li>Update: ignore bundled plugin <code>.openclaw-install-stage</code> directories during global install verification and packaged dist pruning so leftover runtime-dep staging files do not turn successful updates into <code>unexpected packaged dist file</code> failures. Fixes #71752. Thanks @waynegault.</li>
<li>CLI/update: fail package updates when post-update plugin sync fails and refresh legacy npm plugin install records before trusting unchanged artifacts, preventing successful updates from restarting with stale or failed plugin state. Thanks @vincentkoc and @shakkernerd.</li>
<li>Release/update: reject pre-populated bundled plugin <code>.openclaw-install-stage</code> directories, including mixed-case path variants, before package inventory generation so release tarballs cannot ship poisoned runtime-dependency staging debris. Fixes #71752. Thanks @hclsys.</li>
<li>Node runtime: keep node-host retry timers alive across Gateway restarts and exit on terminal credential pauses so supervised nodes do not become silent zombies. Fixes #69800. Thanks @meroli28.</li>
<li>Gateway/plugins: stop persisted WhatsApp auth state from activating bundled channel runtime-dependency repair during startup when <code>channels.whatsapp</code> is absent, avoiding npm/git stalls on packaged Linux installs. Fixes #71994. Thanks @xiao398008.</li>
<li>Gateway/device tokens: enforce caller-scope containment inside token rotation and revocation so pairing-only sessions cannot mutate higher-scope operator tokens. Fixes #71990. Thanks @coygeek.</li>
<li>Plugins/channels: keep security checks, thread-binding placement, provider summaries, health formatting, and message action labels on read-only or already-loaded channel metadata instead of importing full channel runtime. Thanks @shakkernerd.</li>
<li>Plugins/status: keep config-only channel labels and status security summaries from importing plugin runtime modules just to render metadata. Thanks @shakkernerd.</li>
<li>Sessions/channels: stop group-session metadata from loading bundled channel runtime just to classify <code>#channel</code> subjects, using only already-loaded channel capabilities on that path. Thanks @shakkernerd.</li>
<li>Plugins/channels: keep native command and native skill <code>auto</code> defaults on static channel metadata so config, audit, and command-list checks do not load channel runtime just to read those defaults. Thanks @shakkernerd.</li>
<li>CLI/channels: keep channel remove selection and all-channel capabilities summaries on read-only plugin metadata, loading channel runtime only for the selected mutation path. Thanks @shakkernerd.</li>
<li>CLI/models: keep Provider Index preview rows out of <code>models list --all --provider <id></code> when the owning provider plugin is disabled, preserving config authority for cold catalog fallbacks. Thanks @shakkernerd.</li>
<li>CLI/model runs: keep <code>openclaw infer model run</code> on explicit OpenRouter models from loading the full provider catalog or inheriting chat-agent silent-reply policy, restoring non-empty one-shot probe output. Fixes #68791. Thanks @limpredator.</li>
<li>Installer/macOS: rerun Homebrew install steps without the gum spinner when raw-mode ioctl failures occur, and avoid claiming <code>node@24</code> was installed when the Homebrew keg binary is missing. Fixes #70411. Thanks @1fanwang and @dad-io.</li>
<li>Installer: load nvm before Node.js detection so <code>curl | bash</code> installs respect nvm-managed Node instead of stale system Node. Fixes #49556. Thanks @heavenlxj.</li>
<li>Installer/Windows: route PowerShell install failures through a top-level handler so <code>iwr ... | iex</code> returns control to the current shell while direct script-file runs still exit non-zero. Fixes #38054. Thanks @PwrSrg.</li>
<li>CLI/Volta: respawn raw <code>openclaw</code> CLI runs through the named <code>node</code> shim when the current Node executable resolves to <code>volta-shim</code>, avoiding direct shim execution failures in non-interactive shells. Fixes #68672. Thanks @sanchezm86.</li>
<li>Installer: warn when multiple npm global roots contain OpenClaw installs, showing active Node/npm/openclaw plus each install path and version so stale version-manager installs are visible. Fixes #40839. Thanks @zhixianio.</li>
<li>Cron/tasks: recover completed cron task ledger records from durable run logs and job state before marking them <code>lost</code>, reducing false <code>backing session missing</code> audit errors for isolated cron runs and keeping offline CLI audit from treating its empty local cron active-job set as authoritative. Fixes #71963.</li>
<li>Docker: copy patched dependency files into runtime images so downstream <code>pnpm install</code> layers keep working. Fixes #69224. Thanks @gucasbrg.</li>
<li>Package: include patched dependency files in the published npm package so downstream installs can resolve <code>patchedDependencies</code>. (#69224) Thanks @gucasbrg and @vincentkoc.</li>
<li>Plugins/channels: treat malformed bundled channel plugin loaders that return <code>undefined</code> as unavailable instead of crashing config and help paths. Fixes #69044. Thanks @frankhli843 and @vincentkoc.</li>
<li>Scripts/watch: show corrupted dependency package-config recovery guidance when <code>gateway:watch</code> fails during watcher startup, without double-logging unrelated import failures. (#58780) Thanks @roytong9 and @vincentkoc.</li>
<li>Signal: read signal-cli RPC, health checks, and SSE events through Node's HTTP client so Node 24/25 fetch regressions do not break Signal sends or inbound events. Fixes #51716 and #53040. Thanks @Barukimang, @minupla, and @vincentkoc.</li>
<li>Skills/Docker: run npm-backed skill dependency installs with an OpenClaw-managed user prefix so non-root Docker images do not write to <code>/usr/local</code>. Fixes #59601. Thanks @chanjarster and @vincentkoc.</li>
<li>Agents/runtime: submit heartbeat, cron, and exec wakeups as transient runtime context instead of visible user prompts, keeping synthetic system work out of chat transcripts. Fixes #66496 and #66814. Thanks @jeades and @mandomaker.</li>
<li>Telegram: include native quote excerpts automatically for threaded replies and reply tags when the original Telegram text is available, without adding another config knob. Fixes #6975. Thanks @rex05ai.</li>
<li>Node/Linux: make <code>openclaw node install</code> enable and restart the <code>openclaw-node</code> systemd unit instead of the gateway unit on node-only VMs. Fixes #68287. Thanks @dlebee-agent.</li>
<li>Browser/CDP: retry transient raw-CDP WebSocket handshake failures before any browser command is sent, and reconnect stale persistent Playwright CDP sessions for safe tab-list reads without replaying mutating browser actions. Fixes #67728.</li>
<li>Gateway/Linux: retry <code>systemctl --user enable</code> after a second daemon reload when the freshly written gateway unit is not visible yet on migrated systemd installs. Fixes #65184. Thanks @liushuaiiu.</li>
<li>Telegram: preserve exact selected quote text when sending native quote replies, and retry with legacy replies if Telegram rejects quote parameters. (#71952) Thanks @rubencu.</li>
<li>Plugins/CLI: preserve manifest name, description, format, and source metadata in cold <code>openclaw plugins list</code> output without importing plugin runtime. Thanks @shakkernerd.</li>
<li>Security/audit: read channel exposure and plugin allowlist ownership from read-only plugin index metadata so cold audits do not depend on loaded channel runtime. Thanks @shakkernerd.</li>
<li>Plugins/chat: keep <code>/plugins list</code>, <code>/plugins enable</code>, and <code>/plugins disable</code> on the persisted plugin index path so chat plugin management does not load diagnostic/runtime plugin registries before execution. Thanks @shakkernerd.</li>
<li>Plugins/doctor: read workspace plugin status and legacy web-search ownership through installed-index manifest metadata instead of broad manifest registry scans. Thanks @shakkernerd.</li>
<li>CLI/agents: read channel provider status from read-only plugin index metadata for text <code>agents list</code> output instead of the loaded channel registry. Thanks @shakkernerd.</li>
<li>Logging: redact configured secret patterns at console and file-log sink exits so credentials that reach the logger are masked before terminal display or JSONL persistence. Fixes #67953. Thanks @Ziy1-Tan.</li>
<li>Gateway/services: refuse process and service mutations from an older OpenClaw binary when the config was last written by a newer version, preventing split-brain installs from stopping or rewriting newer gateway services. Fixes #57079.</li>
<li>Gateway: reserve <code>/healthz</code> and <code>/readyz</code> ahead of plugin, canvas, and Control UI HTTP stages so liveness/readiness probes still answer when a later route handler stalls. Fixes #69674. Thanks @Xike-Creek.</li>
<li>Logging: load <code>logging.file</code> and redaction settings directly from the active OpenClaw config path in bundled runtimes, so packaged gateways stop falling back to <code>/tmp/openclaw</code>. Fixes #59370, #67168, and #61295. Thanks @KeaneYan, @Pan9hu, and @zsjlovelike.</li>
<li>Logging: rotate file logs at <code>logging.maxFileBytes</code>, keep bounded numbered archives, and make long-lived rolling loggers follow the current-day file instead of suppressing diagnostics or writing stale dated files. Fixes #58583 and #62381. Thanks @jpeghead and @zhaoleink.</li>
<li>Agents/groups: treat clean empty assistant stops as silent <code>NO_REPLY</code> only for always-on groups where silent replies are allowed, while keeping direct and mention-gated sessions on the incomplete-turn retry path. Thanks @MagnaAI.</li>
<li>macOS/Node: keep native remote app nodes from advertising <code>browser.proxy</code>, start browser-capable CLI node services through the restored <code>openclaw node start</code> command, and show an actionable browser-control error when the local control service is missing. Fixes #66637.</li>
<li>Gateway/update: fail package updates when the restarted managed gateway reports the wrong version, including fallback restarts and JSON mode, avoiding false-success mixed-version restarts after macOS LaunchAgent updates. Fixes #71835. Thanks @abhinas90 and @jsompis.</li>
<li>Gateway/update: warn before package updates and bundled plugin runtime-dependency repairs when the target volume appears low on disk space, without blocking installs on best-effort filesystem checks. Fixes #71835. Thanks @abhinas90 and @jsompis.</li>
<li>Plugins/runtime deps: surface activated plugin load failures in health and fail package-update restart verification or doctor repair when bundled runtime deps still cannot load, avoiding false-success repairs. (#71883) Thanks @Solvely-Colin.</li>
<li>Gateway/Linux: include fnm <code>aliases/default/bin</code> in generated service PATHs and let doctor accept either modern fnm aliases or the legacy <code>current/bin</code> symlink, avoiding false PATH repair prompts. Fixes #68169. Thanks @richard-scott.</li>
<li>Installer/Linux: run apt installs with noninteractive dpkg and needrestart settings so fresh Ubuntu 24.04 <code>curl | bash</code> installs do not hang while installing Node.js, Git, or build tools. Fixes #41146. Thanks @iht76, @alexcarv318, @cs3gallery, @firofame, and @cgdusek.</li>
<li>Providers/Bedrock: defer the AWS SDK import until Bedrock discovery actually runs so plugin registration and setup stay lightweight on cold start. Fixes #71690. Thanks @jarvis-ai-gregmoser.</li>
<li>Installer/macOS: stop immediately when Homebrew <code>node@24</code> installation fails and avoid printing PATH advice for missing Homebrew Node installs. Fixes #70411. Thanks @1fanwang.</li>
<li>WhatsApp: remove ack reactions after a visible reply when <code>messages.removeAckAfterReply</code> is enabled, matching other reaction-capable channels. Fixes #26183. Thanks @MrUnforsaken.</li>
<li>Providers/Z.AI: map OpenClaw thinking controls to Z.AI's <code>thinking</code> payload and add opt-in preserved thinking replay via <code>params.preserveThinking</code>, so GLM 5.x can keep prior <code>reasoning_content</code> when requested. Fixes #58680. Thanks @xuanmingguo.</li>
<li>Channels/status: keep read-only channel lists on manifest and package metadata by default, loading setup runtime only for explicit fallback callers. Thanks @shakkernerd.</li>
<li>Plugins: scope setup and web-provider metadata manifest reads to explicit plugin ids when callers already know the owning plugin set. Thanks @vincentkoc.</li>
<li>Plugins/onboarding: defer onboarding install-record index writes until the guarded config commit so setup failures cannot leave the plugin index ahead of <code>openclaw.json</code>. Thanks @shakkernerd.</li>
<li>Plugins/registry: resolve web provider ownership from the installed plugin index instead of broad manifest scans on secret, tool, and pricing paths. Thanks @shakkernerd.</li>
<li>Config/providers: accept <code>video</code> and <code>audio</code> in configured model <code>input</code> values and preserve them in provider catalog entries. Fixes #20721. Thanks @alvinttang.</li>
<li>Models/auth: honor the parent <code>--agent</code> flag for auth write commands (<code>add</code>, <code>login</code>, <code>setup-token</code>, <code>paste-token</code>, and the GitHub Copilot shortcut) so OAuth/API-key/token results are written to the requested agent store instead of the default agent. Fixes #71864. (#71933) Thanks @balric-seo.</li>
<li>TTS: strip model-emitted TTS directives from streamed block text before channel delivery, including directives split across adjacent blocks, while preserving the accumulated raw reply for final-mode synthesis. Fixes #38937.</li>
<li>TTS: keep explicit <code>provider=...</code> directive keys scoped to that provider and warn on unsupported keys instead of letting another speech provider consume overlapping keys. Fixes #60131.</li>
<li>TTS/Feishu: normalize final-mode streamed TTS-only audio before delivery so generated voice-note files use the same safe media path and native voice routing as normal final replies. Fixes #71920.</li>
<li>Feishu: transcribe inbound voice-note audio with the shared media audio path before agent dispatch and keep raw Feishu <code>file_key</code> payloads out of message text. Fixes #67120 and #61876.</li>
<li>Tasks: terminalize async Gateway agent task records from the Gateway run result while preserving aborted, failed, and cancelled outcomes instead of leaving completed runs stuck as active or lost. (#71905) Thanks @likewen-tech.</li>
<li>WhatsApp: let authorized group voice-note transcripts satisfy mention gating before reply dispatch, while keeping unmentioned transcripts in pending group history. Fixes #44908.</li>
<li>Media understanding: carry channel voice-note preflight state into attachment selection so WhatsApp, Feishu, Telegram, and Discord do not transcribe the same inbound audio twice. Fixes #70580.</li>
<li>TTS/BlueBubbles: deliver compatible auto-TTS audio as iMessage voice memo bubbles instead of plain MP3/CAF file attachments. Fixes #16848.</li>
<li>TTS: resolve voice-note and voice-memo routing from channel plugin capabilities instead of speech-core-owned channel id lists.</li>
<li>ACP: send subagent and async-task completion wakes to external ACP harnesses as plain prompts instead of OpenClaw internal runtime-context envelopes, while keeping those envelopes out of ACP transcripts.</li>
<li>TTS/status: show configured TTS model, voice, and sanitized custom endpoint in <code>/status</code>, preserve OpenAI-compatible TTS instructions on custom endpoints, and retry empty Microsoft/Edge TTS output once. Addresses #46602, #47232, and #43936. Thanks @leekuangtao, @Huntterxx, and @rex993.</li>
<li>Agents/Gateway: steer agent-driven config edits and restarts through the owner-only <code>gateway</code> tool, document <code>config.schema.lookup</code> as the field-doc source, and warn against using <code>gateway stop && gateway start</code> as a restart substitute on macOS. Fixes #71929. Thanks @ygc3817922006-sketch.</li>
<li>Media understanding/audio: inject a deterministic transcript placeholder for too-small voice notes so agents do not hallucinate transcription or provider failures. Fixes #48944. Thanks @eulicesl.</li>
<li>Providers/vLLM: send Nemotron 3 chat-template kwargs when thinking is off and honor configured <code>params.chat_template_kwargs</code> for OpenAI-compatible completions, so vLLM/Nemotron replies stay visible instead of becoming thinking-only. Fixes #71891. Thanks @jmystaki-create and @dennis-lynch.</li>
<li>Channels/replies: strip copied inbound metadata blocks from user-facing assistant replies and model replay history, so Discord/vLLM sessions do not leak <code>Conversation info</code> / <code>UNTRUSTED ... message body</code> envelopes after a model echoes them. Fixes #71847. Thanks @jmystaki-create.</li>
<li>Subagents/memory: keep inter-session completion wakes out of memory and dreaming session exports, and strip internal runtime-context blocks from realtime Control UI chat events.</li>
<li>Agents/Claude: treat zero-token empty <code>stop</code> turns as failed provider output, retry once, repair replay, and allow configured model fallback instead of preserving them as successful silent replies. Fixes #71880. Thanks @MagnaAI.</li>
<li>Tasks: normalize task lifecycle timestamps at create, update, and restore time, and report retained lost tasks as audit warnings until their cleanup window expires. (#71871) Thanks @likewen-tech.</li>
<li>Diagnostics/OTEL: treat normal early model stream cleanup as a completed model call instead of exporting a misleading <code>StreamAbandoned</code> error span. Thanks @vincentkoc.</li>
<li>Gateway/pairing: stop corrupt or unreadable device/node pairing stores from being treated as empty state, preserving <code>paired.json</code> for repair instead of overwriting approved pairings. Fixes #71873. Thanks @iret77.</li>
<li>ACP: keep <code>/acp</code> management commands, plus local <code>/status</code> and <code>/unfocus</code>, on the Gateway path inside ACP-bound threads so they are not consumed as ACP prompt text. Fixes #66298. Thanks @kindomLee.</li>
<li>ACPX: stop probing ACP agents during normal Gateway startup; the embedded backend now registers without spawning Codex/ACP child processes unless <code>OPENCLAW_ACPX_RUNTIME_STARTUP_PROBE=1</code> is explicitly set.</li>
<li>CLI/image edit: accept <code>--size</code>, <code>--aspect-ratio</code>, and <code>--resolution</code> on <code>openclaw infer image edit</code> and report all supported edit flags from <code>capability inspect image.edit</code>. Thanks @Pinghuachiu.</li>
<li>ACP: wait for the configured runtime backend to become healthy before startup identity reconciliation, avoiding transient acpx warnings during Gateway boot. Fixes #40566.</li>
<li>Channels/ACP bindings: time out configured binding readiness checks instead of letting Discord preflight hang forever when an ACP target never settles. Fixes #68776.</li>
<li>Control UI: hide the chat loading skeleton during background history reloads when existing messages or active stream content are already visible, avoiding reload flashes on high-latency local gateways. Fixes #71844. Thanks @WolvenRA.</li>
<li>Control UI: keep locally optimistic chat messages visible when a history reload temporarily returns empty, avoiding lost first-turn messages on high-latency gateways. Fixes #71878. Thanks @WolvenRA.</li>
<li>Control UI: keep chat history limits based on visible messages after filtering heartbeat and control-only transcript rows, so recent hidden entries no longer make older visible replies disappear. Thanks @WolvenRA.</li>
<li>Agents/images: scrub old <code>[media attached: ...]</code>, <code>[Image: source: ...]</code>, and <code>media://inbound/...</code> markers from pruned model replay context so stale media refs are not rehydrated as fresh prompt images. Fixes #71868. Thanks @jmeadlock.</li>
<li>Docker/Bonjour: disable Bonjour/mDNS advertising by default for bundled Compose gateways on bridge networking, while keeping host/macvlan opt-in with <code>OPENCLAW_DISABLE_BONJOUR=0</code>. Fixes #71879. Thanks @gbballpack.</li>
<li>CLI/status: label the OpenClaw Serve/Funnel setting as <code>Tailscale exposure</code> and show daemon state separately when available, so <code>gateway.tailscale.mode: "off"</code> no longer reads like the Tailscale daemon is stopped. Fixes #71790. Thanks @pesvobodak.</li>
<li>Plugins/Bonjour: stop ciao mDNS watchdog failures from looping forever when the advertiser stays stuck in <code>probing</code> or <code>announcing</code>; Bonjour now disables itself for the current Gateway process after repeated failed restarts while the Gateway keeps running. Fixes #69011. Thanks @siddharthaagarwalofficial-ux, @FiredMosquito831, and @spikefcz.</li>
<li>Gateway/Fly.io: seed Control UI allowed origins from the actual runtime bind and port so CLI-driven non-loopback starts do not crash before config exists. Fixes #71823.</li>
<li>macOS/remote SSH: keep discovered gateway hosts in <code>gateway.remote.sshTarget</code> while pinning SSH transport URLs to the local loopback tunnel, so browser automation does not regress into blocked non-loopback <code>ws://</code> endpoints. Fixes #67336.</li>
<li>Gateway/proxy: bootstrap env proxy dispatching from direct Gateway startup so provider and plugin network requests honor <code>HTTPS_PROXY</code>/<code>HTTP_PROXY</code> before the first embedded agent attempt runs. (#71833) Thanks @mjamiv.</li>
<li>Plugins/runtime deps: verify clean npm installs actually place requested bundled runtime packages in the managed install root, reporting exact missing specs instead of a false successful repair. (#71883) Thanks @Solvely-Colin.</li>
<li>Plugins/discovery: ignore stale <code>plugins.load.paths</code> aliases that point back at packaged bundled plugin directories and have doctor remove them, keeping bundled plugins on the runtime-deps staging path.</li>
<li>Models/LM Studio: preserve <code>@iq*</code> quant suffixes in model refs and provider matching so <code>/model lmstudio/...@iq3_xxs</code> keeps the exact LM Studio variant. Fixes #71474. (#71486) Thanks @Bartok9, @XinwuC, and @Sanjays2402.</li>
<li>Matrix/cron: preserve the live Matrix delivery target when creating implicit announce reminder jobs so mixed-case room IDs are not reconstructed from lowercased session keys. Fixes #71798.</li>
<li>Feishu: accept Schema 2.0 card action callbacks that report <code>context.open_chat_id</code> instead of legacy <code>context.chat_id</code>, so button callbacks no longer drop as malformed. Fixes #71670. Thanks @eddy1068.</li>
<li>Feishu: keep synthetic card-action and bot-menu ids out of platform reply targets, using the real card callback message id when Feishu provides one and plain-sending otherwise. Fixes #71673. Thanks @eddy1068.</li>
<li>Plugins/QQ Bot: prefer an installed QQ Bot plugin that declares it replaces the bundled <code>qqbot</code> channel, preventing duplicate <code>qqbot_channel_api</code> and <code>qqbot_remind</code> tool registration noise. Fixes #63102.</li>
<li>Browser automation: keep stable tab ids and labels attached when Chromium replaces the raw target after form submissions or other action-triggered navigations, and return the replacement <code>targetId</code> from <code>/act</code> when the match is provable. Fixes #46137.</li>
<li>QQ Bot: make <code>qqbot_remind</code> schedule, list, and remove Gateway cron jobs directly for owner-authorized senders instead of returning <code>cronParams</code> and relying on a follow-up generic <code>cron</code> tool call. Fixes #70865. (#70937) Thanks @GaosCode.</li>
<li>Agents/ACP: hide <code>sessions_spawn</code> ACP runtime options unless an ACP backend is loaded, and make <code>/acp doctor</code> call out <code>plugins.allow</code> blocking bundled <code>acpx</code>. Thanks @vincentkoc.</li>
<li>Agents/Codex: keep ACP prompt/skill routing hidden unless an ACP runtime backend is available, and warn in doctor when enabled Codex plugin configs still route <code>openai-codex/*</code> models through PI. Thanks @vincentkoc.</li>
<li>Media delivery: avoid sending generated image attachments twice when the assistant reply already includes explicit <code>MEDIA:</code> lines for the same turn, and reject unsafe remote <code>MEDIA:</code> URLs before delivery. Thanks @pashpashpash.</li>
<li>Codex harness: ignore retryable app-server error notifications after Codex recovers, and preserve the real nested error message for terminal app-server failures instead of replacing it with a generic failure. Thanks @pashpashpash.</li>
<li>Agents/Codex: prepare native Codex sub-agent session metadata without a nested Gateway session patch and add a focused Docker smoke for the app-server sub-agent path. Thanks @vincentkoc.</li>
<li>Agents/subagents: keep queued subagent announces session-only when the requester has no external channel target, avoiding ambiguous multi-channel delivery failures. Fixes #59201. Thanks @larrylhollan.</li>
<li>Image understanding: preserve configured provider-prefixed vision model metadata when callers request the model without the provider prefix, so custom image models keep their <code>input: ["text", "image"]</code> capability. Fixes #33185. Thanks @Kobe9312 and @vincentkoc.</li>
<li>Plugins/install: restore the previous plugin index records if a concurrent config write conflict interrupts install, update, or uninstall metadata commits. Thanks @shakkernerd.</li>
<li>Plugins/install: reject native plugin archives that do not include a valid <code>openclaw.plugin.json</code>, preventing manifestless archives from writing install records that later show missing-manifest diagnostics. Thanks @shakkernerd.</li>
<li>Plugins/uninstall: remove tracked managed plugin install directories even when the persisted install path differs from the default id-derived target, while still refusing deletes outside the managed extensions root. Thanks @shakkernerd.</li>
<li>Plugins/update: restore previous plugin index records if core update or channel setup hits a concurrent config write conflict after plugin metadata changes. Thanks @shakkernerd.</li>
<li>Plugins/onboarding: defer channel/provider plugin install records until the owning config write commits, keeping setup failures from advancing the plugin index ahead of <code>openclaw.json</code>. Thanks @shakkernerd.</li>
<li>Plugins/config: route configure and agent setup writes with pending plugin install records through the plugin index commit helper so provider onboarding metadata is not stripped by plain config writes. Thanks @shakkernerd.</li>
<li>Plugins/channels: merge pending channel plugin install records with the existing plugin index before config writes, preserving unrelated tracked installs during channel setup, resolve, remove, and capability repair flows. Thanks @shakkernerd.</li>
<li>Plugins/config: defer shipped <code>plugins.installs</code> index migration during config writes until the guarded config commit window and roll it back if the config write fails before commit. Thanks @shakkernerd.</li>
<li>Sessions: keep embedded runtime context out of the visible user prompt by sending it as a hidden next-turn custom message, and teach doctor to repair affected 2026.4.24 transcripts with duplicated prompt-rewrite branches. Fixes #71761.</li>
<li>Gateway/subagents: keep direct-loopback backend RPCs authenticated with the shared gateway token/password off stale CLI paired-device scope baselines, so internal calls no longer hit <code>scope-upgrade</code> pairing prompts while remote, browser, node, device-token, and explicit-device paths still require normal pairing approval. Fixes #63548.</li>
<li>Providers/Azure OpenAI: give deployment-scoped image generation requests a longer 600s default timeout so slow <code>gpt-image-2</code> generations can complete without a per-call <code>timeoutMs</code>. Fixes #71705. Thanks @voytas75.</li>
<li>Gateway/plugins: link source-checkout bundled runtime dependency caches instead of recursively copying <code>node_modules</code> on the gateway main thread, preventing local status, node, and skill probes from timing out during startup cache restores.</li>
<li>Skills/remote nodes: only expose remote macOS skill bins for connected nodes, clear stale bin matches when node probes fail, and include probe command, timeout, bin count, and connection state in timeout logs.</li>
<li>Skills/remote nodes: recognize <code>system.which</code> object-map responses when probing connected macOS nodes, so Linux gateways can expose macOS-only skills such as Apple Notes when the required binaries are installed remotely. Fixes #71877. Thanks @miguelarios.</li>
<li>CLI/gateway: keep diagnostic probes from creating first-time read-only device pairings, while still reusing cached device tokens for detailed read probes. Fixes #71766. Thanks @SunboZ.</li>
<li>CLI/plugins: keep <code>message</code> startup, <code>channels logs</code>, <code>agents delete</code>, and <code>agents set-identity</code> off broad plugin preloading; message delivery still loads plugins when the action actually runs.</li>
<li>Image understanding: resolve configured image models such as local LM Studio vision entries before reporting <code>Unknown model</code> when the discovery registry has not registered that provider. Fixes #66486. Thanks @zhanggpcsu.</li>
<li>QQ Bot: ignore self-echoed bot messages using the outbound ref-index marker, preventing mirrored replies from re-entering the agent loop while still allowing users to quote bot replies. Fixes #71912. Thanks @wangyc6003.</li>
<li>Sessions: separate reset freshness from session-store <code>updatedAt</code>, so heartbeat, cron, exec, and gateway bookkeeping no longer prevent configured daily/idle resets from rolling long-running channel sessions. Fixes #68315, #63732, #63820, and #69083. Thanks @maxatv, @longhairedsi, @bradfreels, and @akessel56.</li>
<li>Sessions: clear queued system-event notices during <code>/new</code>, <code>/reset</code>, gateway <code>sessions.reset</code>, and daily/idle rollover so stale background updates cannot leak into the first prompt of the fresh session. Fixes #66864. Thanks @opeyio, @Magicray1217, and @cedillarack.</li>
<li>CLI/agents: keep <code>agents bind</code>, <code>agents unbind</code>, and <code>agents bindings</code> on setup-safe channel metadata paths so they do not preload bundled plugin runtimes or stage runtime dependencies. Fixes #71743.</li>
<li>Plugins/registry: preserve explicit disabled plugin records during registry migration without persisting every unused bundled plugin discovered on disk. Thanks @shakkernerd.</li>
<li>Windows/native: keep CLI startup and bundled provider plugin loading off Windows ESM raw-path failure paths, fixing native onboarding/install smoke on Node 24.</li>
<li>Plugins/doctor: read bundled channel doctor capabilities through the same packaged plugin directory resolver used by plugin loading, so published installs keep Matrix DM allowlist repairs on <code>channels.matrix.dm.*</code> instead of writing invalid top-level <code>dmPolicy</code> keys. Fixes #71757.</li>
<li>Plugins/Windows: keep bundled plugin Jiti loaders off the native import path on Windows so channel plugins such as Telegram no longer crash with <code>ERR_UNSUPPORTED_ESM_URL_SCHEME</code> on <code>C:\...</code> paths. Fixes #71749. Thanks @smeyer9.</li>
<li>Providers/Ollama: use Ollama's current <code>/api/web_search</code> endpoint and honor <code>https://ollama.com</code> model-provider base URLs for Ollama Web Search. Fixes #71741. Thanks @madhvidua.</li>
<li>Memory/Ollama: serialize Ollama memory embedding batches and add an inline batch timeout override, with longer defaults for local/self-hosted embedding providers.</li>
<li>Sessions/usage: exclude compaction checkpoint transcript snapshots from usage totals and session discovery, while keeping old checkpoint files removable.</li>
<li>CLI/agents: keep <code>openclaw agents list --json</code> on the config-only path by default, avoiding bundled plugin loading unless callers request <code>--bindings</code>. Fixes #71739. Thanks @kaloster.</li>
<li>Plugins/install: force plugin dependency installs to stay project-local even when inherited npm config requests global installs, so successful installs still materialize the plugin's staged <code>node_modules</code>.</li>
<li>Providers/Google: transcode Gemini TTS PCM to Opus for voice-note targets so WhatsApp and other native voice-note replies can play as voice messages.</li>
<li>TTS/WhatsApp: mark non-Opus provider output as voice-note intent so channel delivery transcodes MP3/WebM replies to Ogg/Opus PTT audio.</li>
<li>Plugins/runtime deps: reuse existing external bundled-plugin stage roots when mirrored plugin roots are inspected again, avoiding second-generation <code>openclaw-unknown-*</code> stages and repeated first-turn restaging. Fixes #71599.</li>
<li>iOS/macOS Talk Mode: allow <code>talk.speechLocale</code> to set the speech recognition locale for non-English voice conversations. Fixes #44688.</li>
<li>Plugins/providers: honor explicit plugin candidate lists instead of reading a persisted registry snapshot from local state, keeping candidate-scoped provider discovery hermetic.</li>
<li>Plugins/doctor: keep bundled plugin runtime-dependency repairs inside the managed OpenClaw stage even when user npm prefix/global config points npm at <code>$HOME/node_modules</code>. Fixes #71730.</li>
<li>ACP/sessions_spawn: reject normal OpenClaw config agent ids when callers explicitly request <code>runtime="acp"</code>, while allowing agents configured with <code>runtime.type="acp"</code> to resolve to their ACP harness id. Fixes #63914.</li>
<li>ACP/sessions_spawn: apply <code>runTimeoutSeconds</code> to ACP child turns and dispatch those turns on the background subagent lane, so quota-stalled ACP harnesses do not occupy the main agent lane indefinitely. Fixes #68823.</li>
<li>ACP/oneshot: reconcile runtime session identity before closing completed oneshot ACP runs, so finished <code>sessions.json</code> entries do not stay stuck with <code>acp.identity.state="pending"</code>.</li>
<li>ACPX: bundle <code>acpx@0.6.1</code> so unsupported generic model overrides fail clearly instead of silently falling back to the target adapter default.</li>
<li>ACP/models: document that non-Codex ACP model overrides require adapter support for ACP <code>models</code> plus <code>session/set_model</code>, so unsupported harnesses fail clearly instead of silently falling back to their defaults.</li>
<li>Plugins/Voice Call: treat missing provider credentials as setup-incomplete during Gateway startup and log the missing keys as a warning instead of a runtime startup error, while keeping explicit command/tool errors when used.</li>
<li>Android/Talk Mode: prevent duplicate TTS playback when fast or repeated final chat events arrive while Talk Mode is waiting for its own response. Fixes #46546.</li>
<li>Tooling/check:changed: pass parent heavy-check lock markers to lint lanes so <code>pnpm check:changed</code> no longer waits on its own <code>lint:extensions</code> child.</li>
<li>CLI/completion: dedupe provider auth flags before registering <code>openclaw onboard</code> options, so completion-cache refresh during update no longer fails when stale core fallback flags overlap plugin manifest flags. Fixes #71667.</li>
<li>Diagnostics/trace: report live context usage from the current prompt snapshot instead of provider turn totals, avoiding false near-full context spikes on cached or tool-heavy runs.</li>
<li>Providers/Google: honor <code>models.providers.google.request.allowPrivateNetwork</code> for Gemini TTS and telephony TTS, matching Google image generation and media understanding. (#71723) Thanks @ro-hansolo.</li>
<li>Providers/MiniMax: register <code>minimax-portal</code> for music and video generation, preserving OAuth auth and regional MiniMax base URLs across the shared <code>music_generate</code> and <code>video_generate</code> tools. (#63241) Thanks @tars90percent.</li>
<li>Providers/onboarding: keep Runway and Alibaba Model Studio out of the text-inference setup picker by scoping their video-generation auth choices to the media setup flow. (#65856) Thanks @Jah-yee.</li>
<li>Plugins/Bonjour: stop the gateway from crash-looping on <code>CIAO PROBING CANCELLED</code> when the mDNS watchdog cancels a stuck probe. Restores the rejection-handler wiring dropped during the bonjour plugin migration and shares unhandled-rejection state across module instances so plugin-staged copies of <code>openclaw/plugin-sdk/runtime</code> register into the same handler set the host consults. Especially affects Docker on macOS, where mDNS probing reliably hits the watchdog. Thanks @troyhitch.</li>
<li>Google Meet: report pinned Chrome nodes as offline or missing capabilities in setup/join diagnostics, keep inaccessible nodes out of auto-selection, and preflight local BlackHole/SoX requirements before agents try local Chrome.</li>
<li>Providers/MiniMax: route <code>image-01</code> requests to the dedicated image generation endpoint while preserving CN endpoint selection. Fixes #61149. Thanks @mushuiyu886.</li>
<li>Plugins/startup: remove ownerless bundled runtime-dependency install locks after a short grace window and include lock owner details when startup times out waiting for a plugin runtime-deps lock.</li>
<li>Plugins/install: anchor bundled runtime-dependency npm installs with an OpenClaw-owned package manifest so Linux updates cannot accidentally write to a parent <code>$HOME/node_modules</code> tree. Fixes #71730.</li>
<li>Plugins/install: pass onboarding plugin config into plugin index writes so local plugin installs outside default discovery roots keep their install records. Thanks @shakkernerd.</li>
<li>Plugins/install: migrate shipped <code>plugins.installs</code> config records into the plugin index while stripping them from runtime config and future writes. Thanks @shakkernerd.</li>
<li>Plugins/install: durably remove shipped <code>plugins.installs</code> from <code>openclaw.json</code> after its records are copied into the plugin index, while rolling back the index write if config cleanup fails. Thanks @shakkernerd.</li>
<li>Plugins/install: keep migrated plugin install records in the plugin index even when the plugin manifest is missing or invalid, so update, uninstall, inspect, and audit can still recover broken installs. Thanks @shakkernerd.</li>
<li>Plugins/security: keep plugin audit JSON check ids stable while reporting plugin index install-record findings with updated wording. Thanks @shakkernerd.</li>
<li>CLI/config: reject direct <code>plugins.installs</code> edits with guidance to use <code>openclaw plugins install</code>, <code>openclaw plugins update</code>, or <code>openclaw plugins uninstall</code> instead. Thanks @shakkernerd.</li>
<li>Live tests/voice: accept common STT variants for OpenClaw and ElevenLabs brand names so provider smoke tests fail on real regressions rather than equivalent transcripts.</li>
<li>Agents/replies: forward sanitized underlying agent failure details on external channels instead of replacing unknown failures with a generic retry message.</li>
<li>CLI/MCP: translate OpenClaw <code>mcp.servers.*.transport</code> entries into Claude/Gemini CLI <code>type</code> fields so streamable HTTP MCP servers load in CLI backend sessions. (#71724) Thanks @Blockchain-Oracle.</li>
<li>Browser/CDP: honor configured remote and <code>attachOnly</code> CDP HTTP/WebSocket timeouts when opening tabs through raw CDP or <code>/json/new</code> fallback. (#54238) Thanks @FuncWei.</li>
<li>WhatsApp/TTS: send visible text separately from PTT voice-note audio instead of relying on hidden voice-note captions. Fixes #51081.</li>
<li>Browser/client: avoid telling agents to restart OpenClaw for dispatcher timeouts on external browser profiles such as <code>attachOnly</code>, remote CDP, and existing-session. (#40815) Thanks @0xsline.</li>
<li>Agents/TTS: preserve <code>[[audio_as_voice]]</code> directives on trusted text tool-result <code>MEDIA:</code> payloads so generated audio still delivers as a voice note. (#46535) Thanks @azade-c.</li>
<li>Agents/TTS: keep queued tool media when an assistant ends with <code>NO_REPLY</code> on non-block delivery paths, so media-only generated audio replies still send. (#60025) Thanks @bradlind1.</li>
<li>Telegram/STT: frame inbound voice-note transcripts as machine-generated, untrusted text in agent context while preserving raw transcript mention detection. Closes #33360. Thanks @smartchainark.</li>
<li>Subagents/browser: show an actionable <code>/tools</code> notice when browser automation is configured but filtered out by the active tool profile, and document that coding-profile agents should use <code>tools.alsoAllow: ["browser"]</code> rather than subagent allowlists alone.</li>
<li>Control UI/Quick Settings: persist the assistant avatar override to browser local storage (mirroring the user avatar) so uploaded image data URLs no longer fail config validation with "Too big: expected string to have <=200 characters". Also lift the gateway-side <code>ui.assistant.avatar</code> length cap to match the user avatar size budget for non-UI clients writing the field directly. Thanks @BunsDev.</li>
<li>Plugin SDK: share diagnostic event subscriptions across duplicate source/dist module graphs so legacy root SDK imports still receive runtime diagnostic events.</li>
<li>Agents/Bedrock: prevent empty assistant stream-error turns from poisoning Converse replay by persisting, repairing, and replaying a non-empty fallback block. Fixes #71572. (#71627) Thanks @openperf.</li>
<li>Agents/Anthropic/Bedrock: strip thinking blocks with missing, empty, or blank replay signatures before provider conversion, falling back to non-empty omitted-reasoning text when needed so corrupted signed-thinking history no longer poisons subsequent turns. Fixes #45010. (#70054) Thanks @castaples.</li>
<li>Agents/Anthropic/Bedrock: preserve stripped thinking-only assistant replay turns with non-empty omitted-reasoning text so provider adapters keep strict user/assistant turn shape. Thanks @wujiaming88.</li>
<li>ACP/Codex: pass <code>sessions_spawn(runtime="acp")</code> model and thinking overrides into Codex ACP startup, normalize <code>openai-codex/*</code> refs and slash reasoning suffixes, and recognize managed Codex ACP wrapper commands without blocking current <code>gpt-5.5</code> sessions. Fixes #40393. (#71643) Thanks @91wan.</li>
<li>Browser/CDP: make readiness diagnostics use the same discovery-first fallback as reachability for bare <code>ws://</code> Browserless and Browserbase CDP URLs. Fixes #69532.</li>
<li>Browser/CDP: explain that loopback Browserless or other externally managed CDP services need <code>attachOnly: true</code> and matching Browserless <code>EXTERNAL</code> endpoint when reporting local port ownership conflicts, and fall back to the configured bare WebSocket root when a discovered Browserless endpoint rejects CDP. Fixes #49815.</li>
<li>Gateway/reload: preserve indefinite <code>gateway.reload.deferralTimeoutMs: 0</code> semantics for channel hot reload deferrals so active agent runs are not interrupted by a forced channel restart. (#71637) Thanks @Poo-Squirry.</li>
<li>Agents/tool results: cap persisted Pi tool-result details and strip hidden diagnostics before provider conversion, preventing large debug payloads from bloating session transcripts. (#71637) Thanks @Poo-Squirry.</li>
<li>ACP/OpenCode: update the bundled acpx runtime to 0.6.0 and cover the OpenCode ACP bind path in Docker live tests.</li>
<li>Providers/OpenCode Go: add DeepSeek V4 Pro and DeepSeek V4 Flash to the Go catalog while the bundled Pi registry catches up. Fixes #71587.</li>
<li>Providers/OpenCode Go: route DeepSeek V4 Pro/Flash through the OpenAI-compatible Go endpoint and suppress invalid <code>reasoning_effort: "off"</code> payloads, fixing tool-enabled requests for <code>opencode-go/deepseek-v4-flash</code>. Fixes #71683.</li>
<li>Plugins/model defaults: run Skill Workshop review, Active Memory recall, and session-memory slug generation on the configured agent default model instead of the hardcoded OpenAI SDK fallback when hook context lacks model metadata. Fixes #71659.</li>
<li>Providers/Venice: fill the required DeepSeek V4 <code>reasoning_content</code> placeholder for <code>venice/deepseek-v4-pro</code> and <code>venice/deepseek-v4-flash</code> replay turns without sending native DeepSeek <code>thinking</code> controls that Venice rejects. Fixes #71628.</li>
<li>Browser/existing-session: support per-profile Chrome MCP command/args, map <code>cdpUrl</code> to <code>--browserUrl</code> or <code>--wsEndpoint</code>, and avoid combining endpoint flags with <code>--userDataDir</code>. Fixes #47879, #48037, and #62706. Thanks @puneet1409, @zhehao, and @madkow1001.</li>
<li>Media/plugins: bound MIME sniffing and ZIP archive preflight before handing untrusted files to <code>file-type</code> or <code>jszip</code>, reducing parser CPU and memory exposure for attachments and ClawHub plugin archives. Thanks @vincentkoc.</li>
<li>Memory-host SDK: use trusted env-proxy mode for remote embedding and batch HTTP calls only when Undici will proxy that target, preserving SSRF DNS pinning for <code>ALL_PROXY</code>-only and <code>NO_PROXY</code> bypass cases. Fixes #52162. (#71506) Thanks @DhtIsCoding.</li>
<li>Gateway/dashboard: render Control UI and WebSocket links with <code>https://</code>/<code>wss://</code> when <code>gateway.tls.enabled=true</code>, including <code>openclaw gateway status</code>. Fixes #71494. (#71499) Thanks @deepkilo.</li>
<li>Agents/OpenAI-compatible: default proxy/local completions tool requests to <code>tool_choice: "auto"</code> when tools are present, so providers enter native tool-calling mode instead of replying with plain-text tool directives. (#71472) Thanks @Speed-maker.</li>
<li>OpenAI image generation: use <code>gpt-5.5</code> for the Codex OAuth responses transport instead of the retired <code>gpt-5.4</code> model, fixing 500s from ChatGPT Codex image generation. Fixes #71513. Thanks @baolongl.</li>
<li>OpenAI image generation: route transparent-background default-model requests to <code>gpt-image-1.5</code>, document the expected <code>image_generate</code> call shape, and keep Azure/custom OpenAI-compatible deployment names untouched.</li>
<li>Google video generation: download direct MLDev Veo <code>video.uri</code> results instead of passing them through the Files API path, fixing 404s after successful generation/polling. Fixes #71200. Thanks @panhaishan.</li>
<li>Google video generation: fall back to the REST <code>predictLongRunning</code> Veo endpoint for text-only SDK 404s while keeping reference image/video generation on the SDK path. Fixes #62309 and #63008. (#62343) Thanks @leoleedev.</li>
<li>MiniMax music generation: switch the bundled default model from the unsupported <code>music-2.5+</code> id to the current <code>music-2.6</code> API model. Fixes #64870 and addresses the music default from #62315. Thanks @noahclanman and @edwardzheng1.</li>
<li>Cron: record jobs interrupted by a gateway restart as failed at their original <code>runningAtMs</code>, skip unsafe startup replay, and disable interrupted one-shot jobs so they show a visible failure instead of silently disappearing or duplicating work. Fixes #59056, #61343, #63657, and #59301. Thanks @ponchoooPenguin, @daemic24, @myradon, and @hikiwibot.</li>
<li>Cron tool: recover flat top-level schedule shorthand such as <code>cron</code>, <code>tz</code>, and <code>staggerMs</code> before gateway validation, so model-generated cron add/update calls preserve cron jitter settings. Thanks @tyxben.</li>
<li>Cron: hydrate flat legacy job rows with top-level <code>cron</code>, <code>tz</code>, <code>session</code>, and <code>message</code> fields into canonical schedule, target, and payload objects before startup recomputes run times. Fixes #43351.</li>
<li>Agents/replies: let pending group chat history trigger bare mentioned turns without treating metadata-only inbound context as user input. Fixes #71489. (#71520) Thanks @SymbolStar.</li>
<li>Google media generation: strip a configured trailing <code>/v1beta</code> from Google music/video provider base URLs before calling the Google GenAI SDK, preventing doubled <code>/v1beta/v1beta</code> paths. Fixes #63240. (#63258) Thanks @Hybirdss.</li>
<li>Discord: restore direct-message voice-note preflight transcription and classify URL-only Ogg/Opus voice attachments as audio while skipping partial attachments without usable URLs. Fixes #61314 and #64803.</li>
<li>Plugins/build: copy bundled plugin skill trees into <code>dist-runtime</code>, broaden Windows symlink-copy fallbacks, and fingerprint runtime dependencies from <code>lstat</code> so symlink-like directory entries cannot crash staging.</li>
<li>Google Chat: preserve reply text when a typing indicator message is deleted or can no longer be updated, so media captions and first text chunks are resent instead of silently disappearing. (#71498) Thanks @colin-lgtm.</li>
<li>Cron: tolerate malformed legacy job rows in startup, main-session system-event payloads, and human-readable <code>cron list</code> output so missing <code>state</code>, <code>payload.text</code>, or display fields no longer crash the scheduler or CLI. Fixes #66016, #65916, #64137, #57872, #59968, #63813, #52804, and #43163. (#71509) Thanks @vincentkoc.</li>
<li>CLI/models: make <code>openclaw models scan</code> fall back to public OpenRouter free-model metadata when no <code>OPENROUTER_API_KEY</code> is configured, avoid config secret resolution for explicit <code>--no-probe</code> scans, and apply the scan timeout to the OpenRouter catalog request.</li>
<li>Feishu: keep streaming cards to one live card per turn, flush throttled card edits after meaningful text boundaries, and skip exact block/partial repeats so tool-heavy replies do not duplicate card output. Thanks @allan0509.</li>
<li>Feishu: finish the streaming-card duplicate closeout by stripping leaked reasoning tags, preserving cross-block partial snapshots, enabling topic-thread streaming cards, omitting the generic <code>main</code> card header, surfacing transient tool/compaction status, and cleaning streaming state after close failures. Thanks @sesame437, @Vicky-v7, @maoku-family, @Pengxiao-Wang, and @Maple778.</li>
<li>Telegram: recover incomplete partial-stream previews by falling back to a final send when an ambiguous final edit failure would otherwise retain a strict prefix of the answer. Fixes #71525. (#71554) Thanks @sahilsatralkar.</li>
<li>Control UI/chat: collapse assistant token/model context details behind an explicit Context disclosure and show full dates in message footers, making historical transcript timing clear without noisy default metadata. (#71337) Thanks @BunsDev.</li>
<li>OpenAI/Codex OAuth: explain <code>unsupported_country_region_territory</code> token-exchange failures with a proxy/region hint instead of surfacing a generic OAuth error. Fixes #51175. (#71501) Thanks @vincentkoc and @wulala-xjj.</li>
<li>Browser/Linux: fall back to headless mode for local managed profiles on hosts without a display server, while preserving explicit per-profile headed overrides and reporting the headless source. (#60953) Thanks @rrpsantos.</li>
<li>Telegram: remove the startup persisted-offset <code>getUpdates</code> preflight so polling restarts do not self-conflict before the runner starts. Fixes #69304. (#69779) Thanks @chinar-amrutkar.</li>
<li>Telegram: keep the polling stall watchdog active even when grammY reports the runner as not running while its task is still pending, so a rebuilt transport cannot leave <code>getUpdates</code> silent until a manual gateway restart. Fixes #69064. Thanks @LDLoeb.</li>
<li>Subagents: fall back to direct completion delivery when the parent announce turn finishes without a visible payload, so child results still reach channel-backed requester sessions.</li>
<li>Subagents: tell parent agents to use <code>sessions_yield</code> while waiting for child completion events, preventing GPT-5 fast runs from ending silently after spawning workers.</li>
<li>Browser/Playwright: ignore benign already-handled route races during guarded navigation so browser-page tasks no longer fail when Playwright tears down a route mid-flight. (#68708) Thanks @Steady-ai.</li>
<li>Browser/CLI: lazy-load browser command groups and plugin runtime services so <code>openclaw browser --help</code> can render without loading the full browser automation stack. Fixes #65400. (#65460, #66640) Thanks @pandego and @Tianworld.</li>
<li>Browser/CLI: serve precomputed <code>openclaw browser --help</code> text from CLI startup metadata, avoiding the full plugin/config startup path for the common help invocation.</li>
<li>Browser/downloads: seed managed Chrome profiles with OpenClaw download prefs and capture unmanaged click-triggered downloads under the guarded downloads directory, while explicit download waiters still own their target file. (#64558) Thanks @Pearcekieser.</li>
<li>Browser/Chrome: stop passing redundant <code>--disable-setuid-sandbox</code> when <code>browser.noSandbox</code> is enabled; <code>--no-sandbox</code> remains the effective sandbox opt-out. (#67939) Thanks @sebykrueger.</li>
<li>Browser/client: stop telling agents to permanently avoid the browser after transient timeout or cancellation failures; keep the no-retry hint for persistent unavailable/rate-limit cases. (#46505) Thanks @jriff.</li>
<li>Browser/aria snapshots: bind <code>format=aria</code> <code>axN</code> refs to live DOM nodes through backend DOM ids when Playwright is available, so follow-up browser actions can use those refs without timing out. (#62434) Thanks @MrKipler.</li>
<li>Telegram: prevent duplicate in-process long pollers for the same bot token and add clearer <code>getUpdates</code> conflict diagnostics for external duplicate pollers. Fixes #56230. Thanks @Co-Messi.</li>
<li>Browser/Linux: detect Chromium-based installs under <code>/opt/google</code>, <code>/opt/brave.com</code>, <code>/usr/lib/chromium</code>, and <code>/usr/lib/chromium-browser</code> before asking users to set <code>browser.executablePath</code>. (#48563) Thanks @lupuletic.</li>
<li>Sessions/browser: close tracked browser tabs when idle, daily, <code>/new</code>, or <code>/reset</code> session rollover archives the previous transcript, preventing tabs from leaking past the old session. Thanks @jakozloski.</li>
<li>Sessions/forking: fall back to transcript-estimated parent token counts when cached totals are stale or missing, so oversized thread forks start fresh instead of cloning the full parent transcript. Thanks @jalehman.</li>
<li>OpenAI/Codex: send Codex Responses system prompts through top-level <code>instructions</code> while preserving the existing native Codex payload controls.</li>
<li>MCP/CLI: retire bundled MCP runtimes at the end of one-shot <code>openclaw agent</code> and <code>openclaw infer model run</code> gateway/local executions, so repeated scripted runs do not accumulate stdio MCP child processes. Fixes #71457. Thanks @spartoviMD.</li>
<li>OpenAI/Codex image generation: canonicalize legacy <code>openai-codex.baseUrl</code> values such as <code>https://chatgpt.com/backend-api</code> to the Codex Responses backend before calling <code>gpt-image-2</code>, matching the chat transport. Fixes #71460. Thanks @GodsBoy.</li>
<li>Control UI: make <code>/usage</code> use the fresh context snapshot for context percentage, and include cache-write tokens in the Usage overview cache-hit denominator. Fixes #47885. Thanks @imwyvern and @Ante042.</li>
<li>GitHub Copilot: preserve encrypted Responses reasoning item IDs during replay so Copilot can validate encrypted reasoning payloads across requests. (#71448) Thanks @a410979729-sys.</li>
<li>GitHub Copilot: never rewrite connection-bound reasoning item IDs regardless of whether <code>encrypted_content</code> is present, fixing a 400 "Encrypted content item_id did not match" error with <code>gpt-5.3-codex</code> and future Codex models that fall through to the forward-compat catch-all with <code>reasoning: false</code>. Also recognize Codex-named models as reasoning-capable so they inherit the correct capability flags. Refs #68735. Thanks @InvalidPandaa.</li>
<li>Agents/replies: recover final-answer text when streamed assistant chunks contain only whitespace, preventing completed turns from surfacing as empty-payload errors. Fixes #71454. (#71467) Thanks @Sanjays2402.</li>
<li>Feishu/TTS: transcode voice-intent MP3 and other audio replies to Ogg/Opus before sending native Feishu audio bubbles, while keeping ordinary MP3 attachments as files. Fixes #61249 and #37868. Thanks @sg1416-zg and @ycjlb2023-peteryi.</li>
<li>WhatsApp/TTS: transcode MP3/WebM audio, including Microsoft Edge TTS output, to Ogg/Opus before sending PTT voice notes.</li>
<li>QQBot/TTS: honor plain <code>audioAsVoice</code> replies by synthesizing TTS to native QQ voice messages, and mark inbound voice-only messages as audio media without exposing raw voice paths to generic media context.</li>
<li>Providers/SenseAudio: add bundled SenseAudio batch audio transcription through <code>tools.media.audio</code> with <code>SENSEAUDIO_API_KEY</code> auth. (#66943) Thanks @Fl0rencess720.</li>
<li>Providers/MiniMax: let TTS use MiniMax portal OAuth and Token Plan credentials before falling back to <code>MINIMAX_API_KEY</code>, and include current TTS HD model ids. Fixes #55017. Thanks @zx15210404690-hash.</li>
<li>Telegram/webhook: acknowledge validated webhook updates before running bot middleware, keeping slow agent turns from tripping Telegram delivery retries while preserving per-chat processing lanes. Fixes #71392. Thanks @joelforsberg46-source.</li>
<li>MCP/config reload: hot-apply <code>mcp.*</code> changes by disposing cached session MCP runtimes, and dispose bundled MCP runtimes during gateway shutdown so removed <code>mcp.servers</code> entries reap child processes promptly. Fixes #60656. Thanks @xieyuanqing.</li>
<li>Active Memory: keep silent recall sub-agent billing/auth failures out of shared auth-profile cooldown state, so a Claude CLI extra-usage rejection cannot disable normal Claude-backed turns. Fixes #71284. (#71539) Thanks @vishutdhar and @obviyus.</li>
<li>Auth/Claude CLI: sync refreshed Claude CLI OAuth credentials into the managed auth profile so long-running Claude CLI runs stop falling back to stale OpenClaw snapshots. (#70902) Thanks @starvex.</li>
<li>Sessions: make <code>sessions_spawn(mode="session")</code> errors name usable alternatives when the current channel cannot bind subagent threads. Fixes #67400. (#67790) Thanks @stainlu.</li>
<li>Agents/Claude CLI: pass the OpenClaw system prompt through Claude's prompt-file flag so Windows runs avoid argv length failures without changing system prompt semantics. Fixes #69158. (#69211) Thanks @skylee-01, @cassioanorte, @Syu0, and @Stache73.</li>
<li>Agents/CLI sessions: bind <code>google-gemini-cli</code> session auth-epoch to the Google account identity in <code>~/.gemini/oauth_creds.json</code>, so Gemini-backed agents resume their conversation after gateway restart instead of minting a fresh session, and stale bindings are invalidated when the authenticated Google account changes. Fixes #70973. (#71076) Thanks @openperf.</li>
<li>Slack: stop treating user mentions in assistant-authored message edit blocks as sender attribution, preventing edited bot messages from spoofing a mentioned DM user. (#71700) Thanks @vincentkoc.</li>
<li>Codex: consume unauthorized bound conversation inbound claims before they can fall through to other claim handlers or enqueue Codex turns. (#71702) Thanks @vincentkoc.</li>
<li>Codex media understanding: require approval-checked app-server image turns while explicitly declining tool, file, permission, and elicitation approval requests for the bounded image worker. (#71703) Thanks @vincentkoc.</li>
<li>Agents/Claude CLI: allow large live <code>stream-json</code> JSONL lines up to the existing per-turn raw limit, preventing large Telegram, WebChat, MCP, and image turns from aborting on the old stdout buffer cap. Fixes #71793, #71080, and #70766. (#71897) Thanks @chacher86, @shivamgrover21, and @tpjordan.</li>
<li>Agents/Claude CLI: unwrap nested Claude result envelopes in CLI JSON output so delegated agent responses surface as final text instead of raw result JSON. (#66819) Thanks @mraleko.</li>
<li>Agents/Claude CLI: apply the configured 1M context window override to eligible Claude CLI Opus and Sonnet models when <code>context1m</code> is enabled. (#70863) Thanks @bidadh.</li>
<li>Models/status: report fresh Claude CLI native auth instead of stale stored <code>anthropic:claude-cli</code> profile expiry when local credentials are current. Fixes #71256. (#71332) Thanks @matthiasjanke and @neeravmakwana.</li>
<li>CLI backends: compact OpenClaw transcripts after over-budget CLI turns and reseed fresh CLI sessions from the compacted transcript instead of stale external resume state. Fixes #68329. (#71916) Thanks @obviyus.</li>
<li>Telegram: keep default tool progress messages visible when answer preview streaming is disabled. (#71825) Thanks @VACInc.</li>
<li>Configure/models: clear deselected model fallbacks when updating the model picker allowlist, including provider-scoped setup flows. (#71596) Thanks @rubencu.</li>
<li>Agents/streaming: strip namespaced <code><antml:thinking></code> reasoning tags from streamed assistant replies before user-visible text is emitted. (#69288) Thanks @xialonglee.</li>
</ul>
<p><a href="https://github.com/openclaw/openclaw/blob/main/CHANGELOG.md">View full changelog</a></p>
]]></description>
            <enclosure url="https://github.com/openclaw/openclaw/releases/download/v2026.4.25/OpenClaw-2026.4.25.zip" length="48125363" type="application/octet-stream" sparkle:edSignature="RnQ01wCFgupauUdwOFan+XPGZhBJi/w3sgJYA5EaasbeGrduDHBGw1e9Zj2Lqb4ud8e6Q+tRcJVfxh5KKSEIDg=="/>
        </item>
        <item>
            <title>2026.4.24</title>
            <pubDate>Sat, 25 Apr 2026 19:34:45 +0000</pubDate>
            <link>https://raw.githubusercontent.com/openclaw/openclaw/main/appcast.xml</link>
            <sparkle:version>2026042490</sparkle:version>
            <sparkle:shortVersionString>2026.4.24</sparkle:shortVersionString>
            <sparkle:minimumSystemVersion>15.0</sparkle:minimumSystemVersion>
            <description><![CDATA[<h2>OpenClaw 2026.4.24</h2>
<h3>Highlights</h3>
<ul>
<li>Google Meet joins OpenClaw as a bundled participant plugin, with personal Google auth, Chrome/Twilio realtime sessions, paired-node Chrome support, artifact/attendance exports, and recovery tooling for already-open Meet tabs.</li>
<li>DeepSeek V4 Flash and V4 Pro are in the bundled catalog, V4 Flash is the onboarding default, and DeepSeek thinking/replay behavior is fixed for follow-up tool-call turns.</li>
<li>Talk, Voice Call, and Google Meet can use realtime voice loops that consult the full OpenClaw agent for deeper tool-backed answers.</li>
<li>Browser automation gets coordinate clicks, longer default action budgets, per-profile headless overrides, and steadier tab reuse/recovery.</li>
<li>Plugin and model infrastructure is lighter at startup: static model catalogs, manifest-backed model rows, lazy provider dependencies, and external runtime-dependency repair for packaged installs.</li>
</ul>
<h3>Fixes</h3>
<ul>
<li>Packaged installs: preserve package-root runtime dependencies and their exported subpaths when bundled plugin runtime mirrors fall back to copying shared chunks, fixing Windows npm updates that could fail to load copied <code>dist</code> modules.</li>
<li>Heartbeat: clamp oversized scheduler delays through the shared safe timer helper, preventing <code>every</code> values over Node's timeout cap from becoming a 1 ms crash loop. Fixes #71414. (#71478) Thanks @hclsys.</li>
<li>Telegram: remove the startup persisted-offset <code>getUpdates</code> preflight so polling restarts do not self-conflict before the runner starts. Fixes #69304. (#69779) Thanks @chinar-amrutkar.</li>
<li>Browser/Playwright: ignore benign already-handled route races during guarded navigation so browser-page tasks no longer fail when Playwright tears down a route mid-flight. (#68708) Thanks @Steady-ai.</li>
<li>Browser/aria snapshots: bind <code>format=aria</code> <code>axN</code> refs to live DOM nodes through backend DOM ids when Playwright is available, so follow-up browser actions can use those refs without timing out. (#62434) Thanks @MrKipler.</li>
<li>Telegram: prevent duplicate in-process long pollers for the same bot token and add clearer <code>getUpdates</code> conflict diagnostics for external duplicate pollers. Fixes #56230.</li>
<li>Browser/Linux: detect Chromium-based installs under <code>/opt/google</code>, <code>/opt/brave.com</code>, <code>/usr/lib/chromium</code>, and <code>/usr/lib/chromium-browser</code> before asking users to set <code>browser.executablePath</code>. (#48563) Thanks @lupuletic.</li>
<li>Sessions/browser: close tracked browser tabs when idle, daily, <code>/new</code>, or <code>/reset</code> session rollover archives the previous transcript, preventing tabs from leaking past the old session. Thanks @jakozloski.</li>
<li>Sessions/forking: fall back to transcript-estimated parent token counts when cached totals are stale or missing, so oversized thread forks start fresh instead of cloning the full parent transcript. Thanks @jalehman.</li>
<li>OpenAI/Codex: send Codex Responses system prompts through top-level</li>
</ul>
  <code>instructions</code> while preserving the existing native Codex payload controls.
<ul>
<li>MCP/CLI: retire bundled MCP runtimes at the end of one-shot <code>openclaw agent</code> and <code>openclaw infer model run</code> gateway/local executions, so repeated scripted runs do not accumulate stdio MCP child processes. Fixes #71457.</li>
<li>OpenAI/Codex image generation: canonicalize legacy <code>openai-codex.baseUrl</code> values such as <code>https://chatgpt.com/backend-api</code> to the Codex Responses backend before calling <code>gpt-image-2</code>, matching the chat transport. Fixes #71460.</li>
<li>Control UI: make <code>/usage</code> use the fresh context snapshot for context percentage, and include cache-write tokens in the Usage overview cache-hit denominator. Fixes #47885. Thanks @imwyvern and @Ante042.</li>
<li>GitHub Copilot: preserve encrypted Responses reasoning item IDs during replay so Copilot can validate encrypted reasoning payloads across requests. (#71448) Thanks @a410979729-sys.</li>
<li>Agents/replies: recover final-answer text when streamed assistant chunks contain only whitespace, preventing completed turns from surfacing as empty-payload errors. Fixes #71454. (#71467) Thanks @Sanjays2402.</li>
<li>Feishu/TTS: transcode voice-intent MP3 and other audio replies to Ogg/Opus before sending native Feishu audio bubbles, while keeping ordinary MP3 attachments as files. Fixes #61249 and #37868.</li>
<li>Telegram/webhook: acknowledge validated webhook updates before running bot middleware, keeping slow agent turns from tripping Telegram delivery retries while preserving per-chat processing lanes. Fixes #71392. Thanks @joelforsberg46-source.</li>
<li>MCP: retire one-shot embedded bundled MCP runtimes at run end, skip bundle-MCP startup when a runtime tool allowlist cannot reach bundle-MCP tools, and add <code>mcp.sessionIdleTtlMs</code> idle eviction for leaked session runtimes. Fixes #71106, #71110, #70389, and #70808.</li>
<li>MCP/config reload: hot-apply <code>mcp.*</code> changes by disposing cached session MCP runtimes, and dispose bundled MCP runtimes during gateway shutdown so removed <code>mcp.servers</code> entries reap child processes promptly. Fixes #60656.</li>
<li>Gateway/restart continuation: durably hand restart continuations to a session-delivery queue before deleting the restart sentinel, recover queued continuation work after crashy restarts, and fall back to a session-only wake when no channel route survives reboot. (#70780) Thanks @fuller-stack-dev.</li>
<li>Agents/tool-result pruning: harden the tool-result character estimator and context-pruning loops against malformed <code>{ type: "text" }</code> blocks created by void or undefined tool handler results, serializing non-string text payloads for size accounting so they cannot bypass trimming as zero-sized. Fixes #34979. (#51267) Thanks @cgdusek, @alvinttang, and @coffeexcoin.</li>
<li>Daemon/service-env: add Nix Home Manager profile bin directories to generated gateway service PATHs on macOS and Linux, honoring <code>NIX_PROFILES</code> right-to-left precedence and falling back to <code>~/.nix-profile/bin</code> when unset. Fixes #44402. (#59935) Thanks @jerome-benoit.</li>
</ul>
<p><a href="https://github.com/openclaw/openclaw/blob/main/CHANGELOG.md">View full changelog</a></p>
]]></description>
            <enclosure url="https://github.com/openclaw/openclaw/releases/download/v2026.4.24/OpenClaw-2026.4.24.zip" length="48033180" type="application/octet-stream" sparkle:edSignature="wxOfxadSZ/9iXMitaC6SA9J6YPZC3P2tkeK7HZPHzjUIlzQTvOl7EjR4aRyXzaYt1N1AK5ba+YhuCwEngrTdCQ=="/>
        </item>
        <item>
            <title>2026.4.22</title>
            <pubDate>Thu, 23 Apr 2026 15:18:00 +0000</pubDate>
            <link>https://raw.githubusercontent.com/openclaw/openclaw/main/appcast.xml</link>
            <sparkle:version>2026042290</sparkle:version>
            <sparkle:shortVersionString>2026.4.22</sparkle:shortVersionString>
            <sparkle:minimumSystemVersion>15.0</sparkle:minimumSystemVersion>
            <description><![CDATA[<h2>OpenClaw 2026.4.22</h2>
<h3>Changes</h3>
<ul>
<li>Providers/xAI: add image generation, text-to-speech, and speech-to-text support, including <code>grok-imagine-image</code> / <code>grok-imagine-image-pro</code>, reference-image edits, six live xAI voices, MP3/WAV/PCM/G.711 TTS formats, <code>grok-stt</code> audio transcription, and xAI realtime transcription for Voice Call streaming. (#68694) Thanks @KateWilkins.</li>
<li>Providers/STT: add Voice Call streaming transcription for Deepgram, ElevenLabs, and Mistral, alongside the existing OpenAI and xAI realtime STT paths; ElevenLabs also gains Scribe v2 batch audio transcription for inbound media.</li>
<li>TUI: add local embedded mode for running terminal chats without a Gateway while keeping plugin approval gates enforced. (#66767) Thanks @fuller-stack-dev.</li>
<li>Onboarding: auto-install missing provider and channel plugins during setup so first-run configuration can complete without manual plugin recovery.</li>
<li>OpenAI/Responses: use OpenAI's native <code>web_search</code> tool automatically for direct OpenAI Responses models when web search is enabled and no managed search provider is pinned; explicit providers such as Brave keep the managed <code>web_search</code> tool.</li>
<li>Models/commands: add <code>/models add <provider> <modelId></code> so you can register a model from chat and use it without restarting the gateway; keep <code>/models</code> as a simple provider browser while adding clearer add guidance and copy-friendly command examples. (#70211) Thanks @Takhoffman.</li>
<li>WhatsApp: add configurable native reply quoting with replyToMode for WhatsApp conversations. Thanks @mcaxtr.</li>
<li>WhatsApp/groups+direct: forward per-group and per-direct <code>systemPrompt</code> config into inbound context <code>GroupSystemPrompt</code> so configured per-chat behavioral instructions are injected on every turn. Supports <code>"*"</code> wildcard fallback and account-scoped overrides under <code>channels.whatsapp.accounts.<id>.{groups,direct}</code>; account maps fully replace root maps (no deep merge), matching the existing <code>requireMention</code> pattern. Closes #7011. (#59553) Thanks @Bluetegu.</li>
<li>Agents/sessions: add mailbox-style <code>sessions_list</code> filters for label, agent, and search plus visibility-scoped derived title and last-message previews. (#69839) Thanks @dangoZhang.</li>
<li>Control UI/settings+chat: add a browser-local personal identity for the operator (name plus local-safe avatar), route user identity rendering through the shared chat/avatar path used by assistant and agent surfaces, and tighten Quick Settings, agent fallback chips, and narrow-screen chat layouts so personalization no longer wastes space or clips controls. (#70362) Thanks @BunsDev.</li>
<li>Gateway/diagnostics: enable payload-free stability recording by default and add a support-ready diagnostics export with sanitized logs, status, health, config, and stability snapshots for bug reports. (#70324) Thanks @gumadeiras.</li>
<li>Providers/Tencent: add the bundled Tencent Cloud provider plugin with TokenHub onboarding, docs, <code>hy3-preview</code> model catalog entries, and tiered Hy3 pricing metadata. (#68460) Thanks @JuniperSling.</li>
<li>Providers/Amazon Bedrock Mantle: add Claude Opus 4.7 through Mantle's Anthropic Messages route with provider-owned bearer-auth streaming, so the model is actually callable without treating AWS bearer tokens like Anthropic API keys. Thanks @wirjo.</li>
<li>Providers/GPT-5: move the GPT-5 prompt overlay into the shared provider runtime so compatible GPT-5 models receive the same behavior and heartbeat guidance through OpenAI, OpenRouter, OpenCode, Codex, and other GPT providers; add <code>agents.defaults.promptOverlays.gpt5.personality</code> as the global friendly-style toggle while keeping the OpenAI plugin setting as a fallback.</li>
<li>Providers/OpenAI Codex: remove the Codex CLI auth import path from onboarding and provider discovery so OpenClaw no longer copies <code>~/.codex</code> OAuth material into agent auth stores; use browser login or device pairing instead. (#70390) Thanks @pashpashpash.</li>
<li>CLI/Claude: default <code>claude-cli</code> runs to warm stdio sessions, including custom configs that omit transport fields, and resume from the stored Claude session after Gateway restarts or idle exits. (#69679) Thanks @obviyus.</li>
<li>Pi/models: update the bundled pi packages to <code>0.68.1</code> and let the OpenCode Go catalog come from pi instead of plugin-maintained model aliases, adding the refreshed <code>opencode-go/kimi-k2.6</code>, Qwen, GLM, MiMo, and MiniMax entries.</li>
<li>Tokenjuice: add bundled native OpenClaw support for tokenjuice as an opt-in plugin that compacts noisy <code>exec</code> and <code>bash</code> tool results in Pi embedded runs. (#69946) Thanks @vincentkoc.</li>
<li>ACPX: add an explicit <code>openClawToolsMcpBridge</code> option that injects a core OpenClaw MCP server for selected built-in tools, starting with <code>cron</code>.</li>
<li>CLI/doctor plugins: lazy-load doctor plugin paths and prefer installed plugin <code>dist/*</code> runtime entries over source-adjacent JavaScript fallbacks, reducing the measured <code>doctor --non-interactive</code> runtime by about 74% while keeping cold doctor startup on built plugin artifacts. (#69840) Thanks @gumadeiras.</li>
<li>CLI/debugging: add an opt-in temporary debug timing helper for local CLI performance investigations, with readable stderr output, JSONL capture, and docs for removing probes before landing fixes. (#70469) Thanks @shakkernerd.</li>
<li>Docs/i18n: add Thai translation support for the docs site.</li>
<li>Providers/OpenAI-compatible: mark known local backends such as vLLM, SGLang, llama.cpp, LM Studio, LocalAI, Jan, TabbyAPI, and text-generation-webui as streaming-usage compatible, so their token accounting no longer degrades to unknown/stale totals. (#68711) Thanks @gaineyllc.</li>
<li>Providers/OpenAI-compatible: recover streamed token usage from llama.cpp-style <code>timings.prompt_n</code> / <code>timings.predicted_n</code> metadata and sanitize usage counts before accumulation, fixing unknown or stale totals when compatible servers do not emit an OpenAI-shaped <code>usage</code> object. (#41056) Thanks @xaeon2026.</li>
<li>Plugins/startup: prefer native Jiti loading for built bundled plugin dist modules on supported runtimes, cutting measured bundled plugin load time by 82-90% while keeping source TypeScript on the transform path. (#69925) Thanks @aauren.</li>
<li>Plugin SDK/STT: share realtime transcription WebSocket transport and multipart batch transcription form helpers across bundled STT providers, reducing provider plugin boilerplate while preserving proxy capture, reconnects, audio queueing, close flushing, upload filename normalization, and ready handshakes.</li>
<li>Plugin SDK/Pi embedded runs: add a bundled-plugin embedded extension factory seam so native plugins can extend Pi embedded runs with async runtime hooks such as <code>tool_result</code> handling instead of falling back to the older synchronous persistence path. (#69946) Thanks @vincentkoc.</li>
<li>Codex harness/hooks: route native Codex app-server turns through <code>before_prompt_build</code> and emit <code>before_compaction</code> / <code>after_compaction</code> for native compaction items so prompt and compaction hooks stop drifting from Pi. Thanks @vincentkoc.</li>
<li>Codex harness/plugins: add a bundled-plugin Codex app-server extension seam for async <code>tool_result</code> middleware, fire <code>after_tool_call</code> for Codex tool runs, and route mirrored Codex transcript writes through <code>before_message_write</code> so tool integrations stop diverging from Pi. Thanks @vincentkoc.</li>
<li>Codex harness/hooks: fire <code>llm_input</code>, <code>llm_output</code>, and <code>agent_end</code> for native Codex app-server turns so lifecycle hooks stop drifting from Pi. Thanks @vincentkoc.</li>
<li>QA/Telegram: record per-scenario reply RTT in the live Telegram QA report and summary, starting with the canary response. (#70550) Thanks @obviyus.</li>
<li>Status: add an explicit <code>Runner:</code> field to <code>/status</code> so sessions now report whether they are running on embedded Pi, a CLI-backed provider, or an ACP harness agent/backend such as <code>codex (acp/acpx)</code> or <code>gemini (acp/acpx)</code>. (#70595)</li>
</ul>
<h3>Fixes</h3>
<ul>
<li>Thinking defaults/status: raise the implicit default thinking level for reasoning-capable models from legacy <code>off</code>/<code>low</code> fallback behavior to a safe provider-supported <code>medium</code> equivalent when no explicit config default is set, preserve configured-model reasoning metadata when runtime catalog loading is empty, and make <code>/status</code> report the same resolved default as runtime.</li>
<li>Gateway/model pricing: fetch OpenRouter and LiteLLM pricing asynchronously at startup and extend catalog fetch timeouts to 30 seconds, reducing noisy timeout warnings during slow upstream responses.</li>
<li>Agents/sessions: keep daily reset and idle-maintenance bookkeeping from bumping session activity or pruning freshly active routes, so active conversations no longer look newer or disappear for maintenance-only updates.</li>
<li>Plugins/install: add newly installed plugin ids to an existing <code>plugins.allow</code> list before enabling them, so allowlisted configs load installed plugins after restart.</li>
<li>Status: show <code>Fast</code> in <code>/status</code> when fast mode is enabled, including config/default-derived fast mode, and omit it when disabled.</li>
<li>OpenAI/image generation: detect Azure OpenAI-style image endpoints, use Azure <code>api-key</code> auth plus deployment-scoped image URLs, honor <code>AZURE_OPENAI_API_VERSION</code>, and document the Azure setup path so image generation and edits work against Azure-hosted OpenAI resources. (#70570) Thanks @zhanggpcsu.</li>
<li>Telegram/forum topics: cache recovered forum metadata with bounded expiry so supergroup updates no longer need repeated <code>getChat</code> lookups before topic routing.</li>
<li>Onboarding/WeCom: show the official WeCom channel plugin with its native Enterprise WeChat display name and blurb in the external channel catalog.</li>
<li>Models/auth: merge provider-owned default-model additions from <code>openclaw models auth login</code> instead of replacing <code>agents.defaults.models</code>, so re-authenticating an OAuth provider such as OpenAI Codex no longer wipes other providers' aliases and per-model params. Migrations that must rename keys (Anthropic -> Claude CLI) opt in with <code>replaceDefaultModels</code>. Fixes #69414. (#70435) Thanks @neeravmakwana.</li>
<li>Media understanding/audio: prefer configured or key-backed STT providers before auto-detected local Whisper CLIs, so installed local transcription tools no longer shadow API providers such as Groq/OpenAI in <code>tools.media.audio</code> auto mode. Fixes #68727.</li>
<li>Providers/OpenAI: lock the auth picker wording for OpenAI API key, Codex browser login, and Codex device pairing so the setup choices no longer imply a mixed Codex/API-key auth path. (#67848) Thanks @tmlxrd.</li>
<li>Agents/BTW: route <code>/btw</code> side questions through provider stream registration with the session workspace, so Ollama provider URL construction and workspace-scoped hooks apply correctly. Fixes #68336. (#70413) Thanks @suboss87.</li>
<li>Agents/sessions: make session transcript write locks non-reentrant by default, so same-process transcript writers contend unless a helper explicitly opts into nested lock ownership.</li>
<li>ACPX/probe: expose an optional <code>probeAgent</code> plugin config field so the embedded ACP runtime health probe can target a configured agent (for example <code>opencode</code> or <code>claude</code>) instead of hardcoding <code>codex</code>, and stop marking the entire ACP runtime backend unavailable when the default probe agent is simply not installed or not authenticated. (#68409) Thanks @lyfuci.</li>
<li>Memory search: use sqlite-vec KNN for vector recall while preserving full post-filter result limits in multi-model indexes. Fixes #69666. (#69680) Thanks @aalekh-sarvam.</li>
<li>Providers/OpenAI Codex: stop stale per-agent <code>openai-codex:default</code> OAuth profiles from shadowing a newer main-agent identity-scoped profile, and let <code>openclaw doctor</code> offer the matching cleanup. (#70393) Thanks @pashpashpash.</li>
<li>ACPX: route OpenClaw ACP bridge commands through the MCP-free runtime path even when the command is wrapped with <code>env</code>, has bridge flags, or is resumed from persisted session state, so documented <code>acpx openclaw</code> setups no longer fail on per-session MCP injection. (#68741) Thanks @alexlomt.</li>
<li>Codex harness: route Codex-tagged MCP tool approval elicitations through OpenClaw plugin approvals, including current empty-schema app-server requests, while leaving generic user-input prompts fail-closed. (#68807) Thanks @kesslerio.</li>
<li>WhatsApp/outbound: hold an in-memory active-delivery claim while a live outbound send is in flight, so a concurrent reconnect drain no longer re-drives the same pending queue entry and duplicates cron sends 7-12x after the 30-minute inbound-silence watchdog fires mid-delivery. Crash-replay of fresh queue entries left behind by a dead process is preserved because the claim is intentionally process-local. Fixes #70386. (#70428) Thanks @neeravmakwana.</li>
<li>Matrix/commands: keep Matrix DM allowlist state out of room control-command authorization, so trusted DM senders do not accidentally gain room-command access.</li>
<li>Providers/SDK retry: cap long <code>Retry-After</code> sleeps in Stainless-based Anthropic/OpenAI model SDKs so 60s+ retry windows surface immediately for OpenClaw failover instead of blocking the run. (#68474) Thanks @jetd1.</li>
<li>Agents/TTS: preserve spoken text in TTS tool results while defusing reply directives in transcript content, so future turns remember voice replies without treating spoken <code>MEDIA:</code> or voice tags as delivery metadata. (#68869) Thanks @zqchris.</li>
<li>Providers/OpenAI: harden Voice Call realtime transcription against OpenAI Realtime session-update drift, forward language and prompt hints, and add live coverage for realtime STT.</li>
<li>Agents/Pi embedded runs: suppress the "⚠️ Agent couldn't generate a response" warning when the assistant already delivered user-visible content through a messaging tool and the turn ended cleanly (<code>stopReason=stop</code>). Real failure modes (tool errors, provider <code>stopReason=error</code>, interrupted tool use) still surface the existing "verify before retrying" warning. Fixes #70396. (#70425) Thanks @neeravmakwana.</li>
<li>Gateway/Linux: wrap gateway-managed supervisor, PTY, MCP stdio, and browser child processes in a tiny <code>/bin/sh</code> shim that raises the child's own <code>oom_score_adj</code> on Linux, so under cgroup memory pressure the kernel prefers transient workers over the long-lived gateway. Opt out with <code>OPENCLAW_CHILD_OOM_SCORE_ADJ=0</code>. Fixes #70404. (#70419) Thanks @neeravmakwana.</li>
<li>Providers/Moonshot: stop strict-sanitizing Kimi's native tool_call IDs (shaped like <code>functions.<name>:<index></code>) on the OpenAI-compatible transport, so multi-turn agentic flows through Kimi K2.6 no longer break after 2-3 tool-calling rounds when the serving layer fails to match mangled IDs against the original tool definitions. Adds a <code>sanitizeToolCallIds</code> opt-out to the shared <code>openai-compatible</code> replay family helper and wires Moonshot to it. Fixes #62319. (#70030) Thanks @LeoDu0314.</li>
<li>Dependencies/security: override transitive <code>uuid</code> to <code>14.0.0</code>, clearing the runtime advisory across dependencies.</li>
<li>Codex harness: ignore dynamic tool descriptions when deciding whether to reuse a native app-server thread while still fingerprinting tool schemas, so channel-specific copy changes no longer reset otherwise compatible Codex conversations. (#69976) Thanks @chen-zhang-cs-code.</li>
<li>Codex harness: expose the Codex app-server model catalog in <code>models list/status</code>, avoid startup hangs from app-server discovery timeouts, and accept current Codex turn-completion notifications so Docker live gateway turns finish reliably.</li>
<li>Codex harness: drop invalid legacy app-server <code>serviceTier</code> values such as <code>"priority"</code> before native thread and turn requests, while keeping supported Codex tiers limited to <code>"fast"</code> and <code>"flex"</code>. Fixes #64815.</li>
<li>Codex harness: show bounded, sanitized permission target samples in app-server approval prompts, so native permission requests keep their specific hosts, roots, and paths visible without leaking home usernames or URL credentials. (#70340) Thanks @Lucenx9.</li>
<li>Docs/Codex harness: narrow native compaction docs to the current start/completion signals, without promising a readable summary or kept-entry audit list yet. (#69612) Thanks @91wan.</li>
<li>Providers/Amazon Bedrock: use known context-window metadata for discovered models while keeping the unknown-model fallback conservative, so compaction and overflow handling improve for newer Bedrock models without overstating unlisted model limits. Thanks @wirjo.</li>
<li>Providers/Amazon Bedrock Mantle: refresh IAM-backed bearer tokens at runtime instead of baking discovery-time tokens into provider config, so long-lived Mantle sessions keep working after the initial token ages out. Thanks @wirjo.</li>
<li>Config/includes: write through single-file top-level includes for isolated OpenClaw-owned mutations, so <code>plugins install</code> and <code>plugins update</code> update an included <code>plugins.json5</code> file instead of flattening modular <code>$include</code> configs. Fixes #41050 and #66048.</li>
<li>Config/reload: plan gateway reloads from source-authored config instead of runtime-materialized snapshots, so plugin update writes no longer trigger false restarts from derived provider/plugin config paths. Fixes #68732.</li>
<li>Plugins/update: skip npm plugin reinstall/config rewrites when the installed version and recorded artifact identity already match the registry target, let bare npm package names resolve back to tracked install records, and point already-installed <code>plugins install</code> attempts at <code>plugins update</code> / <code>--force</code> instead of a hook-pack fallback. Fixes #46955, #67957, and #68073.</li>
<li>Agents/MCP: keep <code>mcp.servers</code> and bundle MCP tools available in Pi embedded <code>coding</code> and <code>messaging</code> sessions while preserving <code>minimal</code> profile and <code>tools.deny: ["bundle-mcp"]</code> opt-out behavior. Fixes #68875 and #68818.</li>
<li>Plugins/startup: tolerate transient bundled-channel catalog/metadata drift while auto-enabling configured plugins, so CLI and gateway startup no longer crash when a channel id is known but its display metadata is unavailable.</li>
<li>CLI/Claude: report CLI-backed reply runs as streaming while Claude/Codex CLI turns are still in flight, so WebChat keeps visible response state until the backend finishes. Fixes #70125.</li>
<li>Slack/streaming: fall back to normal Slack replies for Slack Connect streams rejected before the SDK flushes its local buffer, so short replies no longer disappear or report success before Slack acknowledges delivery. Fixes #70295. (#70370) Thanks @mvanhorn.</li>
<li>Codex harness: rotate the shared app-server websocket client when the configured bearer token changes, so auth-token refreshes reconnect with the new <code>Authorization</code> header instead of reusing a stale socket. (#70328) Thanks @Lucenx9.</li>
<li>Channels/sandbox: derive runtime policy keys for external direct messages that share the main conversation, so sandbox/tool policy no longer treats channel-originated DMs as local main-session runs.</li>
<li>Config/models: merge provider-scoped model allowlist updates and protect model/provider map writes from accidental full replacement, adding <code>config set --merge</code> for additive updates and <code>--replace</code> for intentional clobbers. Fixes #65920, #68392, and #68653.</li>
<li>Agents/Pi auth: preserve AWS SDK-authenticated Bedrock runs for IMDS and task-role setups, clear stale refresh timers on sentinel fallback, and log unexpected runtime-auth prep failures instead of silently leaving the provider unauthenticated. Thanks @wirjo.</li>
<li>Config/gateway: restore last-known-good config on critical clobber signatures such as missing metadata, missing <code>gateway.mode</code>, or sharp size drops, preventing gateway crash loops when a valid backup exists. Fixes #70336.</li>
<li>Config/gateway: recover configs accidentally prefixed with non-JSON output during gateway startup or <code>openclaw doctor --fix</code>, preserving the clobbered file as a backup while leaving normal config reads read-only.</li>
<li>Agents/GitHub Copilot: normalize connection-bound Responses item IDs in the Copilot provider wrapper so replayed histories no longer fail after the upstream connection changes. (#69362) Thanks @Menci.</li>
<li>Pi embedded runs: pass real built-in tools into Pi session creation and then narrow active tool names after custom tool registration, so the runner and compaction paths compile cleanly and keep OpenClaw-managed custom tool allowlists without feeding string arrays into <code>createAgentSession</code>. Thanks @vincentkoc.</li>
<li>Agents/OpenAI websocket: route native OpenAI websocket metadata and session-header decisions through the shared endpoint classifier so local mocks and custom <code>models.providers.openai.baseUrl</code> endpoints stay out of the native OpenAI path consistently across embedded-runner and websocket transport code. Thanks @vincentkoc.</li>
<li>Cron/MCP: retire bundled MCP runtimes through one shared cleanup path for isolated cron run ends, persistent cron session rollover, and direct cron <code>deleteAfterRun</code> fallback cleanup. Fixes #69145, #68623, and #68827.</li>
<li>MCP/gateway: tear down stdio MCP process trees on transport close and dispose bundled MCP runtimes during session delete/reset, preventing orphaned wrapper/server processes from accumulating. Fixes #68809 and #69465.</li>
<li>Agents/MCP: retire bundled MCP runtimes after completed one-shot subagent cleanup and nested <code>sessions_send</code> steps, while keeping persistent subagent sessions warm.</li>
<li>Config: render validation warnings with real line breaks instead of a literal <code>\n</code> sequence in CLI/audit output. Fixes #70140.</li>
<li>Cron/doctor: repair malformed persisted cron job IDs through <code>openclaw doctor</code>, including legacy <code>jobId</code>, non-string <code>id</code>, and missing <code>id</code> rows, so <code>cron list</code> no longer needs display-layer coercion for corrupt store data. Fixes #70128.</li>
<li>Discord: normalize prefixed channel targets only at the thread-binding API boundary, so <code>sessions_spawn({ runtime: "acp", thread: true })</code> can create child threads from Discord channels without breaking current-channel ACP bindings. (#68034) Thanks @Zetarcos.</li>
<li>Discord: harden inbound thread metadata handling against partial Carbon channel getters, so non-command thread messages and queued jobs no longer crash when <code>name</code>, <code>parentId</code>, <code>parent</code>, or <code>ownerId</code> requires fetched raw data.</li>
<li>Discord: let <code>message</code> tool reactions resolve <code>user:<id></code> DM targets and preserve <code>channels.discord.guilds.<guild>.channels.<channel>.requireMention: false</code> during reply-stage activation fallback. Fixes #70165 and #69441.</li>
<li>Plugins/startup: pre-normalize and cache Jiti alias maps before creating plugin loaders, so module-scoped loader filenames do not reintroduce per-plugin alias-normalization startup cost. Fixes #70186.</li>
<li>ACP/Codex: run the bundled Codex ACP harness with an isolated <code>CODEX_HOME</code> and avoid writing incomplete ChatGPT auth bridge files, so Codex ACP sessions no longer clobber the user's real Codex CLI auth. Fixes #70234. Thanks @Lonobers88.</li>
<li>Gateway/client: keep long-running RPCs such as ACP <code>agent.wait</code> calls in charge of their own timeout instead of closing the websocket on a missed app-level tick while work is still pending.</li>
<li>Telegram/webhooks: lower the grammY webhook callback timeout to 5s so Telegram gets an early 200 response instead of retrying long-running updates as read timeouts. (#70146) Thanks @friday-james.</li>
<li>Telegram/polling: rebuild the polling HTTP transport after <code>getUpdates</code> 409 conflicts, so retries use a fresh TCP connection instead of looping on a Telegram-terminated keep-alive socket. (#69873) Thanks @hclsys.</li>
<li>Media delivery: strip persisted base64 audio payloads from webchat history, resolve stored <code>media://inbound/*</code> attachments before local-root checks, suppress duplicate Telegram voice/audio sends when TTS emits the same media twice, and support custom image-model IDs that already include their provider prefix.</li>
<li>Slack/files: resolve <code>downloadFile</code> bot tokens from the runtime config when callers provide <code>cfg</code> without an explicit token or prebuilt client, preserving cfg-only file downloads outside the action runtime path. (#70160) Thanks @martingarramon.</li>
<li>Slack/HTTP: dispatch registered Request URL webhooks through the same handler registry used by Slack monitor setup, so HTTP-mode Slack events no longer 404 after successful route registration. (#70275) Thanks @FroeMic.</li>
<li>Slack/runtime bindings: route focused Slack thread replies through their bound ACP session instead of preparing replies against the default agent shell. Fixes #67739. Thanks @Frankla20.</li>
<li>CLI/Claude: keep stored Claude CLI sessions through OAuth refresh-token rotation by keying auth epochs on stable account identity instead of mutable OAuth token material. (#70452) Thanks @obviyus.</li>
<li>CLI/Claude: verify stored Claude CLI session ids have a readable project transcript before resuming, clearing phantom bindings with <code>reason=transcript-missing</code> instead of silently starting fresh under <code>--resume</code>. Fixes #70177.</li>
<li>CLI sessions: persist CLI session clearing through the atomic session-store merge path, so expired Claude/Codex CLI bindings are actually removed before retrying without the stale session id. (#70298) Thanks @HFConsultant.</li>
<li>ACP/sessions_spawn: honor explicit <code>model</code> overrides for ACP child sessions instead of silently falling back to the target agent default model. (#70210) Thanks @felix-miao.</li>
<li>Diffs/viewer: re-read remote viewer access policy from live runtime config on each request, so toggling <code>plugins.entries.diffs.config.security.allowRemoteViewer</code> closes proxied viewer access immediately instead of waiting for a restart. Thanks @vincentkoc.</li>
<li>Diffs/tooling: re-read <code>viewerBaseUrl</code>, presentation defaults, and viewer access policy from live runtime config, and fail closed when the live <code>diffs</code> plugin entry disappears instead of reviving startup viewer settings. Thanks @vincentkoc.</li>
<li>Memory/LanceDB: stop resurrecting removed live <code>memory-lancedb</code> hook config from startup snapshots, so deleting or disabling the plugin entry shuts off auto-recall and auto-capture without a restart. Thanks @vincentkoc.</li>
<li>Memory/LanceDB: keep auto-recall and auto-capture hooks wired when those settings start disabled, so turning them on in live config starts recall and capture without waiting for a restart. Thanks @vincentkoc.</li>
<li>Skill Workshop: keep the tool plus <code>before_prompt_build</code> / <code>agent_end</code> hooks wired while the plugin is disabled at startup, so turning the plugin back on in live config starts guidance and capture without waiting for a restart. Thanks @vincentkoc.</li>
<li>Active Memory: stop reviving removed live <code>active-memory</code> config from startup snapshots, so removing the plugin entry turns the hook off immediately instead of waiting for a restart. Thanks @vincentkoc.</li>
<li>GitHub Copilot: re-read plugin discovery config from the live runtime snapshot, so toggling <code>plugins.entries.github-copilot.config.discovery.enabled</code> takes effect without a restart. Thanks @vincentkoc.</li>
<li>Ollama: re-read plugin discovery config from the live runtime snapshot, so toggling <code>plugins.entries.ollama.config.discovery.enabled</code> takes effect without a restart. Thanks @vincentkoc.</li>
<li>OpenAI: re-read the plugin prompt-overlay personality from live runtime config, so GPT-5 system prompt contributions update without a restart when <code>plugins.entries.openai.config.personality</code> changes. Thanks @vincentkoc.</li>
<li>Amazon Bedrock: re-read live discovery and guardrail plugin config, so toggling <code>plugins.entries.amazon-bedrock.config.discovery</code> or <code>plugins.entries.amazon-bedrock.config.guardrail</code> takes effect without a restart. Thanks @vincentkoc.</li>
<li>Codex: re-read the plugin discovery config from the live runtime snapshot, so toggling <code>plugins.entries.codex.config.discovery</code> takes effect without a restart. Thanks @vincentkoc.</li>
<li>Agents/subagents: drop bare <code>NO_REPLY</code> from the parent turn when the session still has pending spawned children, so direct-conversation surfaces such as Telegram DMs no longer rewrite the sentinel into visible fallback chatter while waiting for the child completion event. (#69942) Thanks @neeravmakwana.</li>
<li>Plugins/install: keep bundled plugin dependencies off npm install while repairing them when plugins activate from a packaged install, including Feishu/Lark, Browser, and direct bundled channel setup-entry loads.</li>
<li>CLI/channels: skip and cache bundled channel plugin, setup, and secrets load failures during read-only discovery, so one broken unused bundled channel cannot crash <code>openclaw status</code> or bootstrap secret scans.</li>
<li>Memory/LanceDB: retry initialization after a failed LanceDB load and report unsupported Intel macOS native runtime clearly instead of caching the failure or repeatedly attempting an install that cannot work.</li>
<li>CLI/Claude: hash only static extra system prompt parts when deciding whether to reuse a CLI session, so per-message inbound metadata no longer resets Claude CLI conversations on every turn. (#70122) Thanks @zijunl.</li>
<li>Hooks/Slack: standardize shared message hook routing fields (<code>threadId</code> / <code>replyToId</code>) and stop Slack outbound delivery from re-running <code>message_sending</code> inside the channel adapter, so plugins like thread-ownership make one outbound routing decision per reply. Thanks @vincentkoc.</li>
<li>Auto-reply/media: share one run-scoped reply media context between streamed block delivery and final payload filtering, so a local <code>MEDIA:</code> attachment is staged once and duplicate media sends are suppressed reliably. (#68111) Thanks @ayeshakhalid192007-dev.</li>
<li>Plugins/gateway hooks: expose startup config, workspace dir, and a live cron getter on the typed <code>gateway_start</code> hook, and move memory-core managed dreaming off the internal <code>gateway:startup</code> bridge so cron reconciliation stays on the public plugin hook path. Thanks @vincentkoc.</li>
<li>Plugins/config: read plugin trust decisions from the source config snapshot when a resolved runtime snapshot is active, so <code>plugins.allow</code> remains enforced and <code>doctor</code>/gateway startup no longer warn that the allowlist is empty when it is configured. Fixes #70161. Also fixes #70141.</li>
<li>Agents/openai-completions: enable malformed streamed tool-call argument repair for self-hosted OpenAI-compatible backends such as Kimi/SGLang, so fragmented tool-call arguments no longer reach tools as empty or unusable objects. Fixes #69672. (#70294) Thanks @MonkeyLeeT.</li>
<li>Gateway/restart: preserve group and channel chat context when resuming an agent turn after a Gateway restart, so continuation replies keep the same prompt, routing, and tool-status behavior as the original conversation.</li>
<li>Gateway/pairing: shared-secret loopback CLI clients now silently auto-approve <code>metadata-upgrade</code> pairing (platform / device family refresh) instead of being disconnected with <code>1008 pairing required</code>. This matches the scope-upgrade and role-upgrade behavior added in #69431 and unblocks non-interactive CLI automation when a paired-device record has a stale platform string (e.g. device key replicated across hosts, install migrated between OSes, or platform-string format changed between OpenClaw versions). Browser / Control-UI clients keep the existing approval-required flow for metadata changes.</li>
<li>Gateway/pairing: treat any forwarded-header evidence (<code>Forwarded</code>, <code>X-Forwarded-*</code>, or <code>X-Real-IP</code>) as proxied WebSocket traffic before pairing locality checks, so reverse-proxy topologies cannot use the loopback shared-secret helper auto-pairing path.</li>
<li>Agents/OpenAI: treat exact <code>NO_REPLY</code> assistant output as a deliberate silent reply in embedded runs, so GPT-5.4 turns with signed reasoning plus a silent final no longer surface a false incomplete-turn error.</li>
<li>Auto-reply/streaming: preserve streamed reply directives through chunk boundaries and phase-aware <code>final_answer</code> delivery, so split <code>MEDIA:<path></code> lines, voice tags, and reply targets reach channel delivery instead of leaking as text or being dropped. (#70243) Thanks @zqchris.</li>
<li>Anthropic/Claude Opus 4.7: normalize Opus 4.7 and <code>claude-cli</code> Opus 4.7 variants to a 1M context window in resolved runtime metadata and active-agent status/context reporting, so they no longer inherit the stale 200k fallback. Thanks @BunsDev.</li>
<li>Gateway/pairing webchat: render <code>/pair qr</code> replies as structured media instead of raw markdown text, preserve inline reply threading and silent-control handling on media replies, avoid persisting sensitive QR images into transcript history, and keep local webchat media embedding behind internal-only trust markers. (#70047) Thanks @BunsDev.</li>
<li>Codex harness: default app-server runs to unchained local execution, so OpenAI heartbeats can use network and shell tools without stalling behind native Codex approvals or the workspace-write sandbox.</li>
<li>Codex harness: fail closed for unknown native app-server approval methods instead of routing unsupported future approval shapes through OpenClaw approval grants. (#70356) Thanks @Lucenx9.</li>
<li>Codex harness: apply the GPT-5 behavior and heartbeat prompt overlay to native Codex app-server runs, so <code>codex/gpt-5.x</code> sessions get the same follow-through, tool-use, and proactive heartbeat guidance as OpenAI GPT-5 runs.</li>
<li>Codex harness: add an explicit Guardian mode for Codex app-server approvals, plus a Docker live probe for approved and ask-back Guardian decisions, while keeping default app-server runs unchained for unattended local heartbeats. The legacy <code>OPENCLAW_CODEX_APP_SERVER_GUARDIAN</code> shortcut is removed; use plugin config <code>appServer.mode: "guardian"</code> or <code>OPENCLAW_CODEX_APP_SERVER_MODE=guardian</code>. Thanks @pashpashpash.</li>
<li>OpenAI/Responses: keep embedded OpenAI Responses runs on HTTP when <code>models.providers.openai.baseUrl</code> points at a local mock or other non-public endpoint, so mocked/custom endpoints no longer drift onto the hardcoded public websocket transport. (#69815) Thanks @vincentkoc.</li>
<li>Channels/config: require resolved runtime config on channel send/action/client helpers and block runtime helper <code>loadConfig()</code> calls, so SecretRefs are resolved at startup/boundaries instead of being re-read during sends.</li>
<li>Discord: pass resolved runtime config through guild and moderation action helpers, so thread-originated Discord commands can run channel, member, role, and guild actions without falling back to runtime config reads. (#70215) Thanks @szponeczek.</li>
<li>CLI/channels: preserve bundled setup promotion metadata when a loaded partial channel plugin omits it, so adding a non-default account still moves legacy single-account fields such as Telegram <code>streaming</code> into <code>accounts.default</code>.</li>
<li>Telegram: keep the sent-message ownership cache isolated per configured session store, so own-message reaction filtering remains correct with custom <code>session.store</code> paths.</li>
<li>Security/update: fail closed when exact pinned npm plugin or hook-pack updates detect integrity drift, and expose aborted plugin drift details in <code>openclaw update --json</code>.</li>
<li>Ollama: forward OpenClaw thinking control to native <code>/api/chat</code> requests as top-level <code>think</code>, so <code>/think off</code> and <code>openclaw agent --thinking off</code> suppress thinking on models such as qwen3 instead of idling until the watchdog fires. Fixes #69902. (#69967) Thanks @WZH8898.</li>
<li>Memory-core/dreaming: suppress the startup-only managed dreaming cron unavailable warning when the cron service is still attaching, while preserving the runtime warning if cron genuinely remains unavailable. Fixes #69939. (#69941) Thanks @Sanjays2402.</li>
<li>Mattermost: suppress reasoning-only payloads even when they arrive as blockquoted <code>> Reasoning:</code> text, preventing <code>/reasoning on</code> from leaking thinking into channel posts. (#69927) Thanks @lawrence3699.</li>
<li>Discord: read <code>channel.parentId</code> through a safe accessor in the slash-command, reaction, and model-picker paths so partial <code>GuildThreadChannel</code> prototype getters no longer throw <code>Cannot access rawData on partial Channel</code> when commands like <code>/new</code> run from inside a thread. Fixes #69861. (#69908) Thanks @neeravmakwana.</li>
<li>Discord: use safe channel name and parent accessors across voice command authorization, so <code>/vc</code> commands from partial Discord thread channels no longer crash on Carbon rawData getters. (#70199) Thanks @hanamizuki.</li>
<li>Discord: make auto-thread parent transcript inheritance opt-in via <code>channels.discord.thread.inheritParent</code>, keeping newly created Discord thread sessions isolated by default while preserving explicit inheritance for configured accounts. Fixes #69907. (#69986) Thanks @Blahdude.</li>
<li>Browser/Chrome MCP: reset cached existing-session control sessions when a <code>navigate_page</code> call times out, so one stuck navigation no longer poisons the browser profile until a gateway restart. (#69733) Thanks @ayeshakhalid192007-dev.</li>
<li>Browser/Chrome MCP: propagate click timeouts and abort signals to existing-session actions so a stuck click fails fast and reconnects instead of poisoning the browser tool until gateway restart. (#63524) Thanks @dongseok0.</li>
<li>Amazon Bedrock/prompt caching: resolve opaque application inference profile targets before injecting Bedrock cache points, require every routed target to support explicit cache points, and retry transient profile lookups instead of caching a false negative for the rest of the process. (#69953) Thanks @anirudhmarc and @vincentkoc.</li>
<li>Gateway/channel health: base stale-socket recovery on provider-proven transport activity instead of inbound app-event freshness, preventing quiet Slack, Discord, Telegram, Matrix, and local-style channels from being restarted solely because no user traffic arrived. (#69833) Thanks @bek91.</li>
<li>OpenCode Go: canonicalize stale bundled <code>opencode-go</code> base URLs from <code>/go</code> or <code>/go/v1</code> to <code>/zen/go</code> or <code>/zen/go/v1</code>, so older generated model metadata stops hitting the 404 HTML endpoint. (#69898)</li>
<li>CLI/channels: honor <code>channels.<id>.enabled=false</code> as a hard read-only presence opt-out, so env vars, manifest env vars, or stale persisted auth state no longer make disabled channel plugins appear in status, doctor, or setup-only discovery.</li>
<li>Channels/preview streaming: centralize draft-preview finalization so Slack, Discord, Mattermost, and Matrix no longer flush temporary preview messages for media/error finals, and preserve first-reply threading for normal fallback delivery.</li>
<li>Discord: keep slash command follow-up chunks ephemeral when the command is configured for ephemeral replies, so long <code>/status</code> output no longer leaks fallback model or runtime details into the public channel. (#69869) thanks @gumadeiras.</li>
<li>Gateway/session history: re-check current auth and <code>chat.history</code> scope before later SSE keepalives and transcript updates, so active session-history streams close before delivering post-revocation events.</li>
<li>Plugins/discovery: reject package plugin source entries that escape the package directory before explicit runtime entries or inferred built JavaScript peers can be used. (#69868) thanks @gumadeiras.</li>
<li>CLI/channels: resolve channel presence through a shared policy that keeps ambient env vars and stale persisted auth from surfacing disabled bundled plugins in status, doctor, security audit, and cron delivery validation unless the channel or plugin is effectively enabled or explicitly configured. (#69862) Thanks @gumadeiras.</li>
<li>Doctor/plugins: hydrate legacy partial interactive handler state before plugin reload clears dedupe caches, so <code>openclaw doctor</code> and post-update doctor runs no longer crash with <code>Cannot read properties of undefined (reading 'clear')</code>. (#70135) Thanks @ngutman.</li>
<li>Control UI/config: preserve intentionally empty raw config snapshots when clearing pending updates so reset restores the original bytes instead of synthesizing JSON for blank config files. (#68178) Thanks @BunsDev.</li>
<li>memory-core/dreaming: surface a <code>Dreaming status: blocked</code> line in <code>openclaw memory status</code> when dreaming is enabled but the heartbeat that drives the managed cron is not firing for the default agent, and add a Troubleshooting section to the dreaming docs covering the two common causes (per-agent <code>heartbeat</code> blocks excluding <code>main</code>, and <code>heartbeat.every</code> set to <code>0</code>/empty/invalid), so the silent failure described in #69843 becomes legible on the status surface.</li>
<li>Cron/run-log: report generic <code>message</code> tool sends under the resolved delivery channel when they match the cron target, while preserving account-specific mismatch checks for delivery traces. (#69940) Thanks @davehappyminion.</li>
<li>Doctor/channels: merge configured-channel doctor hooks across read-only, loaded, setup, and runtime plugin discovery so partial adapters no longer hide runtime-only compatibility repair or allowlist warnings, preserve disabled-channel opt-outs, and ignore malformed hook values before they can mask valid fallbacks. (#69919) Thanks @gumadeiras.</li>
<li>Models/CLI: show bundled provider-owned static catalog rows in <code>models list --all</code> before auth is configured, including Kimi K2.6 rows for Moonshot, OpenRouter, and Vercel AI Gateway, while keeping local-only and workspace plugin catalog paths isolated. (#69909) Thanks @shakkernerd.</li>
<li>Models/CLI: clarify that <code>models list --provider</code> expects provider ids and reject display labels before loading model discovery. (#70504) Thanks @shakkernerd.</li>
<li>Configure: skip generic CLI startup bootstrap for <code>openclaw configure</code> and bound hint-only gateway probes so the onboarding TUI reaches its first prompt faster when the Gateway is unavailable. (#69984) Thanks @obviyus.</li>
<li>Agents/harness: surface selected plugin harness failures directly instead of replaying the same turn through embedded PI, preventing misleading secondary PI auth errors and avoiding duplicate side effects.</li>
<li>OpenAI Codex: add a ChatGPT device-code auth option beside browser OAuth, so headless or callback-hostile setups can sign in without relying on the localhost browser callback. (#69557) Thanks @vincentkoc.</li>
<li>CLI sessions: keep provider-owned CLI sessions through implicit daily expiry while preserving explicit reset behavior, and retain Claude CLI binding metadata across gateway agent requests. (#70106) Thanks @obviyus.</li>
<li>fix(config): accept truncateAfterCompaction (#68395). Thanks @MonkeyLeeT</li>
<li>CLI/Claude: keep Claude CLI session bindings stable across OAuth access-token refreshes, so gateway restarts continue the same Claude conversation instead of minting a fresh one. (#70132) Thanks @obviyus.</li>
<li>QQBot: add <code>INTERACTION</code> intent (<code>1 << 26</code>) to the gateway constants and include it in the <code>FULL_INTENTS</code> mask so interaction events are received. (#70143) Thanks @cxyhhhhh.</li>
<li>Gateway/restart: preserve one-shot continuation instructions across gateway restarts so agents can resume and reply back to the original chat after reboot. (#63406) Thanks @VACInc.</li>
<li>Gateway/restart: write restart sentinel files atomically so interrupted writes cannot leave a truncated sentinel behind. (#70225) Thanks @obviyus.</li>
<li>Pairing: remove stale pending requests for a device when that paired device is deleted, so an old repair approval cannot recreate the removed device from leftover state.</li>
<li>Security/dotenv: block workspace <code>.env</code> overrides for Matrix, Mattermost, IRC, and Synology endpoint settings so cloned workspaces cannot redirect bundled connector traffic through local endpoint config. (#70240) Thanks @drobison00.</li>
<li>Telegram: require the same <code>/models</code> authorization for group model-picker callbacks, so unauthorized participants can no longer browse or change the session model through inline buttons. (#70235) Thanks @drobison00.</li>
<li>Agents/Pi: keep the filtered tool-name allowlist active for embedded OpenAI/OpenAI Codex GPT-5 runs and compaction sessions, so bundled and client tools still execute after the Pi <code>0.68.1</code> session-tool allowlist change instead of stopping at plan-only replies with no tool call. (#70281) Thanks @jalehman.</li>
<li>Agents/Pi: honor explicit <code>strict-agentic</code> execution contracts for incomplete-turn retry guards across providers, so manually opted-in local or compatible models get the same retry behavior without relying on OpenAI model inference. (#66750) Thanks @ziomancer.</li>
<li>OpenShell/sandbox: pin verified file reads to an already-opened descriptor, walk the ancestor chain for symlinked parents on platforms without fd-path readlink, and re-check file identity so parent symlink swaps cannot redirect in-sandbox reads to host files outside the allowed mount root. (#69798) Thanks @drobison00.</li>
<li>Gateway/Control UI: require authenticated Control UI read access before serving <code>/__openclaw/control-ui-config.json</code> when <code>gateway.auth</code> is enabled, so unauthenticated callers can no longer read bootstrap metadata. (#70247) Thanks @drobison00.</li>
<li>Gateway/restart: default session-scoped restart sentinels to a one-shot agent continuation, so chat-initiated Gateway restarts acknowledge successful boot automatically. (#70269) Thanks @obviyus.</li>
<li>Build/npm publish: fail postpublish verification when root <code>dist/*</code> files import bundled plugin runtime dependencies without mirroring them in the root package manifest, so Slack-style plugin deps cannot silently ship on the wrong module-resolution path again. (#60112) thanks @medns.</li>
</ul>
<p><a href="https://github.com/openclaw/openclaw/blob/main/CHANGELOG.md">View full changelog</a></p>
]]></description>
            <enclosure url="https://github.com/openclaw/openclaw/releases/download/v2026.4.22/OpenClaw-2026.4.22.zip" length="47883836" type="application/octet-stream" sparkle:edSignature="kzJ2j2sWX4H+ZIc4dXEFORYr9tk3w1txpjCJ38cdSFz6yWHU0M6Sx9zN0DB7JGIpv1QC+D+jFbWBkl4SJqW2AA=="/>
        </item>
    </channel>
</rss>