<li>Agents/Browser: add <code>browser.target</code> (sandbox/host/custom) with sandbox host-control gating via <code>agents.defaults.sandbox.browser.allowHostControl</code>, allowlists for custom control URLs/hosts/ports, and expand browser tool docs (remote control, profiles, internals).</li>
<li>Onboarding/Models: add catalog-backed default model picker to onboarding + configure. (#611) — thanks @jonasjancarik.</li>
<li>Agents/OpenCode Zen: update fallback models + defaults, keep legacy alias mappings. (#669) — thanks @magimetal.</li>
<li>Providers: unify group history context wrappers across providers with per-provider/per-account <code>historyLimit</code> overrides (fallback to <code>messages.groupChat.historyLimit</code>). Set <code>0</code> to disable. (#672) — thanks @steipete.</li>
<li>Docker: allow optional home volume + extra bind mounts in <code>docker-setup.sh</code>. (#679) — thanks @gabriel-trigo.</li>
</ul>
<h3>Fixes</h3>
<ul>
<li>CLI/Status: expand tables to full terminal width; clarify provider setup vs runtime warnings; richer per-provider detail; token previews in <code>status</code> while keeping <code>status --all</code> redacted; add troubleshooting link footer; keep log tails pasteable; show gateway auth used when reachable; surface provider runtime errors (Signal/iMessage/Slack); harden <code>tailscale status --json</code> parsing; make <code>status --all</code> scan progress determinate; and replace the footer with a 3-line “Next steps” recommendation (share/debug/probe).</li>
<li>CLI/Gateway: clarify that <code>clawdbot gateway status</code> reports RPC health (connect + RPC) and shows RPC failures separately from connect failures.</li>
<li>CLI/Update: gate progress spinner on stdout TTY and align clean-check step label. (#701) — thanks @bjesuiter.</li>
<li>Telegram: add <code>/whoami</code> + <code>/id</code> commands to reveal sender id for allowlists; allow <code>@username</code> and prefixed ids in <code>allowFrom</code> prompts (with stability warning).</li>
<li>Control UI: stop auto-writing <code>telegram.groups["*"]</code> and warn/confirm before enabling wildcard groups.</li>
<li>WhatsApp: send ack reactions only for handled messages and ignore legacy <code>messages.ackReaction</code> (doctor copies to <code>whatsapp.ackReaction</code>). (#629) — thanks @pasogott.</li>
<li>Sandbox/Skills: mirror skills into sandbox workspaces for read-only mounts so SKILL.md stays accessible.</li>
<li>Docker: allow optional apt packages during image build and document the build arg. (#697) — thanks @gabriel-trigo.</li>
<li>Gateway/Heartbeat: deliver reasoning even when the main heartbeat reply is <code>HEARTBEAT_OK</code>. (#694) — thanks @antons.</li>
<li>macOS: clear unsigned launchd overrides on signed restarts and warn via doctor when attach-only/disable markers are set. (#695) — thanks @jeffersonwarrior.</li>
<li>Agents: enforce single-writer session locks and drop orphan tool results to prevent tool-call ID failures (MiniMax/Anthropic-compatible APIs).</li>
<li>Docs: make <code>clawdbot status</code> the first diagnostic step, clarify <code>status --deep</code> behavior, and document <code>/whoami</code> + <code>/id</code>.</li>
<li>Docs/Testing: clarify live tool+image probes and how to list your testable <code>provider/model</code> ids.</li>
<li>Tests/Live: make gateway bash+read probes resilient to provider formatting while still validating real tool calls.</li>
<li>WhatsApp: detect @lid mentions in groups using authDir reverse mapping + resolve self JID E.164 for mention gating. (#692) — thanks @peschee.</li>
<li>Gateway/Auth: default to token auth on loopback during onboarding, add doctor token generation flow, and tighten audio transcription config to Whisper-only.</li>
<li>Providers: dedupe inbound messages across providers to avoid duplicate LLM runs on redeliveries/reconnects. (#689) — thanks @adam91holt.</li>
<li>Agents: strip <code><thought></code>/<code><antthinking></code> tags from hidden reasoning output and cover tag variants in tests. (#688) — thanks @theglove44.</li>
<li>macOS: save model picker selections as normalized provider/model IDs and keep manual entries aligned. (#683) — thanks @benithors.</li>
<li>Agents: recognize "usage limit" errors as rate limits for failover. (#687) — thanks @evalexpr.</li>
<li>CLI: avoid success message when daemon restart is skipped. (#685) — thanks @carlulsoe.</li>
<li>Commands: disable <code>/config</code> + <code>/debug</code> by default; gate via <code>commands.config</code>/<code>commands.debug</code> and hide from native registration/help output.</li>
<li>Agents/System: clarify that sub-agents remain sandboxed and cannot use elevated host access.</li>
<li>Gateway: disable the OpenAI-compatible <code>/v1/chat/completions</code> endpoint by default; enable via <code>gateway.http.endpoints.chatCompletions.enabled=true</code>.</li>
<li>macOS: stabilize bridge tunnels, guard invoke senders on disconnect, and drain stdout/stderr to avoid deadlocks. (#676) — thanks @ngutman.</li>
<li>Agents/System: clarify sandboxed runtime in system prompt and surface elevated availability when sandboxed.</li>
<li>Auto-reply: prefer <code>RawBody</code> for command/directive parsing (WhatsApp + Discord) and prevent fallback runs from clobbering concurrent session updates. (#643) — thanks @mcinteerj.</li>
<li>WhatsApp: fix group reactions by preserving message IDs and sender JIDs in history; normalize participant phone numbers to JIDs in outbound reactions. (#640) — thanks @mcinteerj.</li>
<li>WhatsApp: expose group participant IDs to the model so reactions can target the right sender.</li>
<li>Cron: <code>wakeMode: "now"</code> waits for heartbeat completion (and retries when the main lane is busy). (#666) — thanks @roshanasingh4.</li>
<li>Agents/OpenAI: fix Responses tool-only → follow-up turn handling (avoid standalone <code>reasoning</code> items that trigger 400 “required following item”) and replay reasoning items in Responses/Codex Responses history for tool-call-only turns.</li>
<li>Sandbox: add <code>clawdbot sandbox explain</code> (effective policy inspector + fix-it keys); improve “sandbox jail” tool-policy/elevated errors with actionable config key paths; link to docs.</li>
<li>Hooks/Gmail: keep Tailscale serve path at <code>/</code> while preserving the public path. (#668) — thanks @antons.</li>
<li>Hooks/Gmail: allow Tailscale target URLs to preserve internal serve paths.</li>
<li>Auth: update Claude Code keychain credentials in-place during refresh sync; share JSON file helpers; add CLI fallback coverage.</li>
<li>Auth: throttle external CLI credential syncs (Claude/Codex), reduce Keychain reads, and skip sync when cached credentials are still fresh.</li>
<li>CLI: respect <code>CLAWDBOT_STATE_DIR</code> for node pairing + voice wake settings storage. (#664) — thanks @azade-c.</li>
<li>Gateway/Control UI: make <code>chat.send</code> non-blocking, wire Stop to <code>chat.abort</code>, and treat <code>/stop</code> as an out-of-band abort. (#653)</li>
<li>Gateway/Control UI: allow <code>chat.abort</code> without <code>runId</code> (abort active runs), suppress post-abort chat streaming, and prune stuck chat runs. (#653)</li>
<li>Gateway/Control UI: sniff image attachments for chat.send, drop non-images, and log mismatches. (#670) — thanks @cristip73.</li>
<li>macOS: force <code>restart-mac.sh --sign</code> to require identities and keep bundled Node signed for relay verification. (#580) — thanks @jeffersonwarrior.</li>
<li>Gateway/Agent: accept image attachments on <code>agent</code> (multimodal message) and add live gateway image probe (<code>CLAWDBOT_LIVE_GATEWAY_IMAGE_PROBE=1</code>).</li>
<li>CLI: <code>clawdbot sessions</code> now includes <code>elev:*</code> + <code>usage:*</code> flags in the table output.</li>
<li>Docs: showcase entries for ParentPay, R2 Upload, iOS TestFlight, and Oura Health. (#650) — thanks @henrino3.</li>
<li>Agents: repair session transcripts by dropping duplicate tool results across the whole history (unblocks Anthropic-compatible APIs after retries).</li>
<li>Tests/Live: reset the gateway session between model runs to avoid cross-provider transcript incompatibilities (notably OpenAI Responses reasoning replay rules).</li>
</ul>
<p><ahref="https://github.com/clawdbot/clawdbot/blob/main/CHANGELOG.md">View full changelog</a></p>
<li>CLI: <code>clawdbot message</code> now subcommands (<code>message send|poll|...</code>) and requires <code>--provider</code> unless only one provider configured.</li>
<li>Commands/Tools: <code>/restart</code> and gateway restart tool disabled by default; enable with <code>commands.restart=true</code>.</li>
</ul>
<h3>New Features and Changes</h3>
<ul>
<li>Models/Auth: OpenCode Zen onboarding (#623) — thanks @magimetal; MiniMax Anthropic-compatible API + hosted onboarding (#590, #495) — thanks @mneves75, @tobiasbischoff.</li>
<li>Models/Auth: setup-token + token auth profiles; <code>clawdbot models auth order {get,set,clear}</code>; per-agent auth candidates in <code>/model status</code>; OAuth expiry checks in doctor/status.</li>
<li>Agent/System: claude-cli runner; <code>session_status</code> tool (and sandbox allow); adaptive context pruning default; system prompt messaging guidance + no auto self-update; eligible skills list injection; sub-agent context trimmed.</li>
<li>Packaging: include MS Teams send module in npm tarball.</li>
<li>Sandbox/Browser: auto-start CDP endpoint; proxy CDP out of container for attachOnly; relax Bun fetch typing; align sandbox list output with config images.</li>
<li>Agents/Runtime: gate heartbeat prompt to default sessions; /stop aborts between tool calls; require explicit system-event session keys; guard small context windows; fix model fallback stringification; sessions_spawn inherits provider; failover on billing/credits; respect auth cooldown ordering; restore Anthropic OAuth tool dispatch + tool-name bypass; avoid OpenAI invalid reasoning replay; harden Gmail hook model defaults.</li>
<li>Agent history/schema: strip/skip empty assistant/error blocks to prevent session corruption/Claude 400s; scrub unsupported JSON Schema keywords + sanitize tool call IDs for Cloud Code Assist; simplify Gemini-compatible tool/session schemas; require raw for config.apply.</li>
<li>Status/Commands: provider prefix in /status model display; usage filtering + provider mapping; auth label + usage snapshots (claude-cli fallback + optional claude.ai); show Verbose/Elevated only when enabled; compact usage/cost line + restore emoji-rich status; /status in directive-only + multi-directive handling; mention-bypass elevated handling; surface provider usage errors; wire /usage to /status; restore hidden gateway-daemon alias; fallback /model list when catalog unavailable.</li>
<li>WhatsApp: vCard/contact cards (prefer FN, include numbers, show all contacts, keep summary counts, better empty summaries); preserve group JIDs + normalize targets; resolve @lid mappings/JIDs (Baileys/auth-dir) + inbound mapping; route queued replies to sender; improve web listener errors + remove provider name from errors; record outbound activity account id; fix web media fetch errors; broadcast group history consistency.</li>
<li>Telegram: keep streamMode draft-only; long-poll conflict retries + update dedupe; grammY fetch mismatch fixes + restrict native fetch to Bun; suppress getUpdates stack traces; include user id in pairing; audio_as_voice handling fixes.</li>
