OpenClaw docs capability index

過去レポートのView/ソース規律バックフィルで生成したView。

Generated: 2026-05-10T13:21:19+09:00

Docs tree

.i18n/README.md

AGENTS.md

auth-credential-semantics.md

automation/auth-monitoring.md

automation/clawflow.md

automation/cron-jobs.md

automation/cron-vs-heartbeat.md

automation/gmail-pubsub.md

automation/hooks.md

automation/index.md

automation/poll.md

automation/standing-orders.md

automation/taskflow.md

automation/tasks.md

automation/troubleshooting.md

automation/webhook.md

brave-search.md

channels/access-groups.md

channels/bluebubbles.md

channels/broadcast-groups.md

channels/channel-routing.md

channels/discord.md

channels/feishu.md

channels/googlechat.md

channels/group-messages.md

channels/groups.md

channels/imessage.md

channels/index.md

channels/irc.md

channels/line.md

channels/location.md

channels/matrix-migration.md

channels/matrix-push-rules.md

channels/matrix.md

channels/mattermost.md

channels/msteams.md

channels/nextcloud-talk.md

channels/nostr.md

channels/pairing.md

channels/qqbot.md

channels/signal.md

channels/slack.md

channels/synology-chat.md

channels/telegram.md

channels/tlon.md

channels/troubleshooting.md

channels/twitch.md

channels/wechat.md

channels/whatsapp.md

channels/yuanbao.md

channels/zalo.md

channels/zalouser.md

ci.md

cli/acp.md

cli/agent.md

cli/agents.md

cli/approvals.md

cli/backup.md

cli/browser.md

cli/channels.md

cli/clawbot.md

cli/commitments.md

cli/completion.md

cli/config.md

cli/configure.md

cli/crestodian.md

cli/cron.md

cli/daemon.md

cli/dashboard.md

cli/devices.md

cli/directory.md

cli/dns.md

cli/docs.md

cli/doctor.md

cli/flows.md

cli/gateway.md

cli/health.md

cli/hooks.md

cli/index.md

cli/infer.md

cli/logs.md

cli/mcp.md

cli/memory.md

cli/message.md

cli/migrate.md

cli/models.md

cli/node.md

cli/nodes.md

cli/onboard.md

cli/pairing.md

cli/plugins.md

cli/proxy.md

cli/qr.md

cli/reset.md

cli/sandbox.md

cli/secrets.md

cli/security.md

cli/sessions.md

cli/setup.md

cli/skills.md

cli/status.md

cli/system.md

cli/tasks.md

cli/tui.md

cli/uninstall.md

cli/update.md

cli/voicecall.md

cli/webhooks.md

cli/wiki.md

concepts/active-memory.md

concepts/agent-loop.md

concepts/agent-runtimes.md

concepts/agent-workspace.md

concepts/agent.md

concepts/architecture.md

concepts/channel-docking.md

concepts/commitments.md

concepts/compaction.md

concepts/context-engine.md

concepts/context.md

concepts/delegate-architecture.md

concepts/dreaming.md

concepts/experimental-features.md

concepts/features.md

concepts/mantis.md

concepts/markdown-formatting.md

concepts/memory-builtin.md

concepts/memory-honcho.md

concepts/memory-qmd.md

concepts/memory-search.md

concepts/memory.md

concepts/messages.md

concepts/model-failover.md

concepts/model-providers.md

concepts/models.md

concepts/multi-agent.md

concepts/oauth.md

concepts/openclaw-sdk.md

concepts/parallel-specialist-lanes.md

concepts/presence.md

concepts/progress-drafts.md

concepts/qa-e2e-automation.md

concepts/qa-matrix.md

concepts/queue-steering.md

concepts/queue.md

concepts/retry.md

concepts/session-pruning.md

concepts/session-tool.md

concepts/session.md

concepts/soul.md

concepts/streaming.md

concepts/system-prompt.md

concepts/timezone.md

concepts/typebox.md

concepts/typing-indicators.md

concepts/usage-tracking.md

date-time.md

debug/node-issue.md

diagnostics/flags.md

gateway/authentication.md

gateway/background-process.md

gateway/bonjour.md

gateway/bridge-protocol.md

gateway/cli-backends.md

gateway/config-agents.md

gateway/config-channels.md

gateway/config-tools.md

gateway/configuration-examples.md

gateway/configuration-reference.md

gateway/configuration.md

gateway/diagnostics.md

gateway/discovery.md

gateway/doctor.md

gateway/gateway-lock.md

gateway/health.md

gateway/heartbeat.md

gateway/index.md

gateway/local-models.md

gateway/logging.md

gateway/multiple-gateways.md

gateway/network-model.md

gateway/openai-http-api.md

gateway/openresponses-http-api.md

gateway/openshell.md

gateway/opentelemetry.md

gateway/operator-scopes.md

gateway/pairing.md

gateway/prometheus.md

gateway/protocol.md

gateway/remote-gateway-readme.md

gateway/remote.md

gateway/sandbox-vs-tool-policy-vs-elevated.md

gateway/sandboxing.md

gateway/secrets-plan-contract.md

gateway/secrets.md

gateway/security/audit-checks.md

gateway/security/index.md

gateway/tailscale.md

gateway/tools-invoke-http-api.md

gateway/troubleshooting.md

gateway/trusted-proxy-auth.md

help/debugging.md

help/environment.md

help/faq-first-run.md

help/faq-models.md

help/faq.md

help/gpt55-codex-agentic-parity-maintainers.md

help/gpt55-codex-agentic-parity.md

help/index.md

help/scripts.md

help/testing-live.md

help/testing-updates-plugins.md

help/testing.md

help/troubleshooting.md

index.md

install/ansible.md

install/azure.md

install/bun.md

install/clawdock.md

install/development-channels.md

install/digitalocean.md

install/docker-vm-runtime.md

install/docker.md

install/exe-dev.md

install/fly.md

install/gcp.md

install/hetzner.md

install/hostinger.md

install/index.md

install/installer.md

install/kubernetes.md

install/macos-vm.md

install/migrating-claude.md

install/migrating-hermes.md

install/migrating.md

install/nix.md

install/node.md

install/northflank.mdx

install/oracle.md

install/podman.md

install/railway.mdx

install/raspberry-pi.md

install/render.mdx

install/uninstall.md

install/updating.md

logging.md

network.md

nodes/audio.md

nodes/camera.md

nodes/images.md

nodes/index.md

nodes/location-command.md

nodes/media-understanding.md

nodes/talk.md

nodes/troubleshooting.md

nodes/voicewake.md

perplexity.md

pi-dev.md

pi.md

plan/codex-context-engine-harness.md

plan/ui-channels.md

platforms/android.md

platforms/digitalocean.md

platforms/index.md

platforms/ios.md

platforms/linux.md

platforms/mac/bundled-gateway.md

platforms/mac/canvas.md

platforms/mac/child-process.md

platforms/mac/dev-setup.md

platforms/mac/health.md

platforms/mac/icon.md

platforms/mac/logging.md

platforms/mac/menu-bar.md

platforms/mac/peekaboo.md

platforms/mac/permissions.md

platforms/mac/remote.md

platforms/mac/signing.md

platforms/mac/skills.md

platforms/mac/voice-overlay.md

platforms/mac/voicewake.md

platforms/mac/webchat.md

platforms/mac/xpc.md

platforms/macos.md

platforms/oracle.md

platforms/raspberry-pi.md

platforms/windows.md

plugins/agent-tools.md

plugins/architecture-internals.md

plugins/architecture.md

plugins/building-extensions.md

plugins/building-plugins.md

plugins/bundles.md

plugins/codex-computer-use.md

plugins/codex-harness.md

plugins/community.md

plugins/compatibility.md

plugins/dependency-resolution.md

plugins/google-meet.md

plugins/hooks.md

plugins/manage-plugins.md

plugins/manifest.md

plugins/memory-lancedb.md

plugins/memory-wiki.md

plugins/message-presentation.md

plugins/plugin-inventory.md

plugins/reference.md

plugins/reference/acpx.md

plugins/reference/alibaba.md

plugins/reference/amazon-bedrock-mantle.md

plugins/reference/amazon-bedrock.md

plugins/reference/anthropic-vertex.md

plugins/reference/anthropic.md

plugins/reference/arcee.md

plugins/reference/azure-speech.md

plugins/reference/bluebubbles.md

plugins/reference/bonjour.md

plugins/reference/brave.md

plugins/reference/browser.md

plugins/reference/byteplus.md

plugins/reference/cerebras.md

plugins/reference/chutes.md

plugins/reference/cloudflare-ai-gateway.md

plugins/reference/codex.md

plugins/reference/comfy.md

plugins/reference/copilot-proxy.md

plugins/reference/deepgram.md

plugins/reference/deepinfra.md

plugins/reference/deepseek.md

plugins/reference/diagnostics-otel.md

plugins/reference/diagnostics-prometheus.md

plugins/reference/diffs.md

plugins/reference/discord.md

plugins/reference/document-extract.md

plugins/reference/duckduckgo.md

plugins/reference/elevenlabs.md

plugins/reference/exa.md

plugins/reference/fal.md

plugins/reference/feishu.md

plugins/reference/file-transfer.md

plugins/reference/firecrawl.md

plugins/reference/fireworks.md

plugins/reference/github-copilot.md

plugins/reference/google-meet.md

plugins/reference/google.md

plugins/reference/googlechat.md

plugins/reference/gradium.md

plugins/reference/groq.md

plugins/reference/huggingface.md

plugins/reference/imessage.md

plugins/reference/inworld.md

plugins/reference/irc.md

plugins/reference/kilocode.md

plugins/reference/kimi.md

plugins/reference/line.md

plugins/reference/litellm.md

plugins/reference/llm-task.md

plugins/reference/lmstudio.md

plugins/reference/lobster.md

plugins/reference/matrix.md

plugins/reference/mattermost.md

plugins/reference/memory-core.md

plugins/reference/memory-lancedb.md

plugins/reference/memory-wiki.md

plugins/reference/microsoft-foundry.md

plugins/reference/microsoft.md

plugins/reference/migrate-claude.md

plugins/reference/migrate-hermes.md

plugins/reference/minimax.md

plugins/reference/mistral.md

plugins/reference/moonshot.md

plugins/reference/msteams.md

plugins/reference/nextcloud-talk.md

plugins/reference/nostr.md

plugins/reference/nvidia.md

plugins/reference/ollama.md

plugins/reference/open-prose.md

plugins/reference/openai.md

plugins/reference/opencode-go.md

plugins/reference/opencode.md

plugins/reference/openrouter.md

plugins/reference/openshell.md

plugins/reference/perplexity.md

plugins/reference/qa-channel.md

plugins/reference/qa-lab.md

plugins/reference/qa-matrix.md

plugins/reference/qianfan.md

plugins/reference/qqbot.md

plugins/reference/qwen.md

plugins/reference/runway.md

plugins/reference/searxng.md

plugins/reference/senseaudio.md

plugins/reference/sglang.md

plugins/reference/signal.md

plugins/reference/skill-workshop.md

plugins/reference/slack.md

plugins/reference/stepfun.md

plugins/reference/synology-chat.md

plugins/reference/synthetic.md

plugins/reference/tavily.md

plugins/reference/telegram.md

plugins/reference/tencent.md

plugins/reference/tlon.md

plugins/reference/together.md

plugins/reference/tokenjuice.md

plugins/reference/tts-local-cli.md

plugins/reference/twitch.md

plugins/reference/venice.md

plugins/reference/vercel-ai-gateway.md

plugins/reference/vllm.md

plugins/reference/voice-call.md

plugins/reference/volcengine.md

plugins/reference/voyage.md

plugins/reference/vydra.md

plugins/reference/web-readability.md

plugins/reference/webhooks.md

plugins/reference/whatsapp.md

plugins/reference/xai.md

plugins/reference/xiaomi.md

plugins/reference/zai.md

plugins/reference/zalo.md

plugins/reference/zalouser.md

plugins/sdk-agent-harness.md

plugins/sdk-channel-plugins.md

plugins/sdk-channel-turn.md

plugins/sdk-entrypoints.md

plugins/sdk-migration.md

plugins/sdk-overview.md

plugins/sdk-provider-plugins.md

plugins/sdk-runtime.md

plugins/sdk-setup.md

plugins/sdk-subpaths.md

plugins/sdk-testing.md

plugins/skill-workshop.md

plugins/voice-call.md

plugins/webhooks.md

plugins/zalouser.md

prose.md

providers/alibaba.md

providers/anthropic.md

providers/arcee.md

providers/azure-speech.md

providers/bedrock-mantle.md

providers/bedrock.md

providers/cerebras.md

providers/chutes.md

providers/claude-max-api-proxy.md

providers/cloudflare-ai-gateway.md

providers/comfy.md

providers/deepgram.md

providers/deepinfra.md

providers/deepseek.md

providers/elevenlabs.md

providers/fal.md

providers/fireworks.md

providers/github-copilot.md

providers/glm.md

providers/google.md

providers/gradium.md

providers/groq.md

providers/huggingface.md

providers/index.md

providers/inferrs.md

providers/inworld.md

providers/kilocode.md

providers/litellm.md

providers/lmstudio.md

providers/minimax.md

providers/mistral.md

providers/models.md

providers/moonshot.md

providers/nvidia.md

providers/ollama.md

providers/openai.md

providers/opencode-go.md

providers/opencode.md

providers/openrouter.md

providers/perplexity-provider.md

providers/qianfan.md

providers/qwen.md

providers/runway.md

providers/senseaudio.md

providers/sglang.md

providers/stepfun.md

providers/synthetic.md

providers/tencent.md

providers/together.md

providers/venice.md

providers/vercel-ai-gateway.md

providers/vllm.md

providers/volcengine.md

providers/vydra.md

providers/xai.md

providers/xiaomi.md

providers/zai.md

reference/AGENTS.default.md

reference/RELEASING.md

reference/api-usage-costs.md

reference/application-modernization-plan.md

reference/credits.md

reference/device-models.md

reference/full-release-validation.md

reference/memory-config.md

reference/openclaw-sdk-api-design.md

reference/prompt-caching.md

reference/rich-output-protocol.md

reference/rpc.md

reference/secretref-credential-surface.md

reference/session-management-compaction.md

reference/templates/AGENTS.dev.md

reference/templates/AGENTS.md

reference/templates/BOOT.md

reference/templates/BOOTSTRAP.md

reference/templates/HEARTBEAT.md

reference/templates/IDENTITY.dev.md

reference/templates/IDENTITY.md

reference/templates/SOUL.dev.md

reference/templates/SOUL.md

reference/templates/TOOLS.dev.md

reference/templates/TOOLS.md

reference/templates/USER.dev.md

reference/templates/USER.md

reference/test.md

reference/token-use.md

reference/transcript-hygiene.md

reference/wizard.md

security/CONTRIBUTING-THREAT-MODEL.md

security/THREAT-MODEL-ATLAS.md

security/formal-verification.md

security/incident-response.md

security/network-proxy.md

start/bootstrapping.md

start/docs-directory.md

start/getting-started.md

start/hubs.md

start/lore.md

start/onboarding-overview.md

start/onboarding.md

start/openclaw.md

start/quickstart.md

start/setup.md

start/showcase.md

start/wizard-cli-automation.md

start/wizard-cli-reference.md

start/wizard.md

superpowers/specs/2026-04-22-tweakcn-custom-theme-import-design.md

tools/acp-agents-setup.md

tools/acp-agents.md

tools/agent-send.md

tools/apply-patch.md

tools/brave-search.md

tools/browser-control.md

tools/browser-linux-troubleshooting.md

tools/browser-login.md

tools/browser-wsl2-windows-remote-cdp-troubleshooting.md

tools/browser.md

tools/btw.md

tools/capability-cookbook.md

tools/clawhub.md

tools/code-execution.md

tools/creating-skills.md

tools/diffs.md

tools/duckduckgo-search.md

tools/elevated.md

tools/exa-search.md

tools/exec-approvals-advanced.md

tools/exec-approvals.md

tools/exec.md

tools/firecrawl.md

tools/gemini-search.md

tools/grok-search.md

tools/image-generation.md

tools/index.md

tools/kimi-search.md

tools/llm-task.md

tools/lobster.md

tools/loop-detection.md

tools/media-overview.md

tools/minimax-search.md

tools/multi-agent-sandbox-tools.md

tools/music-generation.md

tools/ollama-search.md

tools/pdf.md

tools/perplexity-search.md

tools/plugin.md

tools/reactions.md

tools/searxng-search.md

tools/skills-config.md

tools/skills.md

tools/slash-commands.md

tools/steer.md

tools/subagents.md

tools/tavily.md

tools/thinking.md

tools/tokenjuice.md

tools/trajectory.md

tools/tts.md

tools/video-generation.md

tools/web-fetch.md

tools/web.md

tts.md

vps.md

web/control-ui.md

web/dashboard.md

web/index.md

web/tui.md

web/webchat.md

Capability keyword hits

heartbeat

platforms/mac/health.md:34:- You can still use the CLI flow in [Gateway health](/gateway/health) (openclaw status, openclaw status --deep, openclaw health --json) and tail /tmp/openclaw/openclaw-*.log for web-heartbeat / web-reconnect.

tools/thinking.md:119:- Heartbeat probe body is the configured heartbeat prompt (default: Read HEARTBEAT.md if it exists (workspace context). Follow it strictly. Do not infer or repeat old tasks from prior chats. If nothing needs attention, reply HEARTBEAT_OK.). Inline directives in a heartbeat message apply as usual (but avoid changing session defaults from heartbeats).

tools/thinking.md:120:- Heartbeat delivery defaults to the final payload only. To also send the separate Reasoning: message (when available), set agents.defaults.heartbeat.includeReasoning: true or per-agent agents.list[].heartbeat.includeReasoning: true.

tools/exec.md:98:- tools.exec.notifyOnExit (default: true): when true, backgrounded exec sessions enqueue a system event and request a heartbeat on exit.

tools/loop-detection.md:77:When a run id is available, recent tool-call history is evaluated only within that run so scheduled heartbeat cycles and fresh runs do not inherit stale loop counts from earlier runs.

plugins/sdk-channel-plugins.md:32:- Heartbeat typing — optional typing/busy signals for heartbeat delivery targets

plugins/sdk-channel-plugins.md:38:heartbeat.sendTyping(...) on the channel plugin. Core calls it with the

plugins/sdk-channel-plugins.md:39:resolved heartbeat delivery target before the heartbeat model run starts and

plugins/sdk-channel-plugins.md:40:uses the shared typing keepalive/cleanup lifecycle. Add heartbeat.clearTyping(...)

plugins/sdk-migration.md:26: mixed system events, heartbeat state, delivery queues, fetch/proxy helpers,

plugins/sdk-migration.md:314: | Heartbeat event and visibility helpers | openclaw/plugin-sdk/heartbeat-runtime |

plugins/sdk-migration.md:432: | plugin-sdk/heartbeat-runtime | Heartbeat helpers | Heartbeat event and visibility helpers |

plugins/sdk-migration.md:451: | plugin-sdk/reply-runtime | Shared reply runtime | Inbound dispatch, heartbeat, reply planner, chunking |

plugins/codex-harness.md:25:Codex heartbeat turns also get the heartbeat_respond tool by default, so the

plugins/codex-harness.md:30:developer instruction on the heartbeat turn itself. Ordinary chat turns restore

plugins/codex-harness.md:31:Codex Default mode instead of carrying heartbeat philosophy in their normal

plugins/codex-harness.md:498:for autonomous heartbeats: Codex can use shell and network tools without

plugins/codex-harness.md:625:cron, browser, nodes, gateway, heartbeat_respond, and web_search remain

plugins/sdk-overview.md:156: agent_turn_prepare, before_prompt_build, heartbeat_prompt_contribution,

plugins/sdk-overview.md:166:| Background lifecycle monitor | Runtime lifecycle cleanup, agent event subscription, session scheduler ownership/cleanup, heartbeat prompt contribution, UI descriptor |

plugins/hooks.md:110:- heartbeat_prompt_contribution — add heartbeat-only context for background monitor and lifecycle plugins

plugins/hooks.md:228:- heartbeat_prompt_contribution: runs only for heartbeat turns and returns

plugins/sdk-subpaths.md:194: | plugin-sdk/reply-runtime | Shared inbound/reply runtime helpers, chunking, dispatch, heartbeat, reply planner |

plugins/sdk-subpaths.md:238: | plugin-sdk/heartbeat-runtime | Heartbeat event and visibility helpers |

plugins/architecture-internals.md:647:text-runtime, runtime-store, system-event-runtime, heartbeat-runtime,

providers/openai.md:471:The bundled native Codex harness uses the same GPT-5 behavior and heartbeat overlay through Codex app-server developer instructions, so openai/gpt-5.x sessions forced through agentRuntime.id: "codex" keep the same follow-through and proactive heartbeat guidance even though Codex owns the rest of the harness prompt.

start/openclaw.md:25:- Heartbeats now default to every 30 minutes. Disable until you trust the setup by setting agents.defaults.heartbeat.every: "0m".

start/openclaw.md:118:- heartbeats (once you trust it)

start/openclaw.md:131: heartbeat: { every: "0m" },

start/openclaw.md:167:By default, OpenClaw runs a heartbeat every 30 minutes with the prompt:

start/openclaw.md:169:Set agents.defaults.heartbeat.every: "0m" to disable.

start/openclaw.md:171:- If HEARTBEAT.md exists but is effectively empty (only blank lines and markdown headers like # Heading), OpenClaw skips the heartbeat run to save API calls.

start/openclaw.md:172:- If the file is missing, the heartbeat still runs and the model decides what to do.

start/openclaw.md:173:- If the agent replies with HEARTBEAT_OK (optionally with short padding; see agents.defaults.heartbeat.ackMaxChars), OpenClaw suppresses outbound delivery for that heartbeat.

start/openclaw.md:174:- By default, heartbeat delivery to DM-style user:<id> targets is allowed. Set agents.defaults.heartbeat.directPolicy: "block" to suppress direct-target delivery while keeping heartbeat runs active.

start/openclaw.md:180: heartbeat: { every: "30m" },

start/hubs.md:92:- [Heartbeat](/gateway/heartbeat)

cli/doctor.md:50:- Doctor checks openai-codex/* model refs across primary models, fallbacks, heartbeat/subagent/compaction overrides, hooks, channel model overrides, and stale session route pins. --fix rewrites them to openai/* only when the native Codex runtime is installed, enabled, contributes the codex harness, and has usable OAuth, or when direct OpenAI auth is already available and no usable Codex OAuth route would be moved. When openai-codex/* is the working Codex OAuth route through OpenClaw PI, doctor preserves it. If an earlier repair left openai/* GPT-5 routes on PI while only Codex OAuth auth is available, --fix recovers them back to openai-codex/*; when direct OpenAI auth is also usable, doctor warns and leaves the ambiguous mixed-auth route unchanged for explicit confirmation.

cli/doctor.md:52:- Doctor repairs stale plugin config by removing missing plugin ids from plugins.allow/plugins.entries, plus matching dangling channel config, heartbeat targets, and channel model overrides when plugin discovery is healthy.

cli/commitments.md:6: - You are auditing what heartbeat may deliver

commitments

cli/commitments.md:2:summary: "CLI reference for openclaw commitments (inspect and dismiss inferred follow-ups)"

cli/commitments.md:4: - You want to inspect inferred follow-up commitments

cli/commitments.md:7:title: "openclaw commitments"

cli/commitments.md:10:List and manage inferred follow-up commitments.

cli/commitments.md:13:conversation context. See [Inferred commitments](/concepts/commitments) for the

cli/commitments.md:16:With no subcommand, openclaw commitments lists pending commitments.

cli/commitments.md:21:openclaw commitments [--all] [--agent <id>] [--status <status>] [--json]

cli/commitments.md:22:openclaw commitments list [--all] [--agent <id>] [--status <status>] [--json]

cli/commitments.md:23:openclaw commitments dismiss <id...> [--json]

cli/commitments.md:28:- --all: show all statuses instead of only pending commitments.

cli/commitments.md:36:List pending commitments:

cli/commitments.md:39:openclaw commitments

cli/commitments.md:45:openclaw commitments --all

cli/commitments.md:51:openclaw commitments --agent main

cli/commitments.md:54:Find snoozed commitments:

cli/commitments.md:57:openclaw commitments --status snoozed

cli/commitments.md:60:Dismiss one or more commitments:

cli/commitments.md:63:openclaw commitments dismiss cm_abc123 cm_def456

cli/commitments.md:69:openclaw commitments --all --json

cli/commitments.md:87:- [Inferred commitments](/concepts/commitments)

cli/index.md:23:| Models and inference | [models](/cli/models) · [infer](/cli/infer) · capability (alias for [infer](/cli/infer)) · [memory](/cli/memory) · [commitments](/cli/commitments) · [wiki](/cli/wiki) |

cli/index.md:124: commitments

concepts/commitments.md:3:title: "Inferred commitments"

concepts/commitments.md:8: - You want to review or dismiss follow-up commitments

concepts/commitments.md:26:## Enable commitments

concepts/commitments.md:31:openclaw config set commitments.enabled true

concepts/commitments.md:32:openclaw config set commitments.maxPerDay 3

concepts/commitments.md:39: "commitments": {

concepts/commitments.md:46:commitments.maxPerDay limits how many inferred follow-ups can be delivered

concepts/commitments.md:52:separate context. That pass looks only for inferred follow-up commitments. It

concepts/commitments.md:68:If heartbeat is configured with target: "none", due commitments remain

concepts/commitments.md:102:## Manage commitments

concepts/commitments.md:104:Use the CLI to inspect and clear stored commitments:

concepts/commitments.md:107:openclaw commitments

concepts/commitments.md:108:openclaw commitments --all

concepts/commitments.md:109:openclaw commitments --agent main

concepts/commitments.md:110:openclaw commitments --status snoozed

concepts/commitments.md:111:openclaw commitments dismiss cm_abc123

concepts/commitments.md:114:See [openclaw commitments](/cli/commitments) for the command reference.

concepts/commitments.md:123:Stored commitments are local OpenClaw state. They are operational memory, not

taskflow

tools/lobster.md:11:Lobster is one authoring layer above detached background work. For flow orchestration above individual tasks, see [Task Flow](/automation/taskflow) (openclaw tasks flow). For the task activity ledger, see [openclaw tasks](/automation/tasks).

cli/flows.md:18:For full documentation see [Task Flow](/automation/taskflow) and the [tasks CLI reference](/cli/tasks).

automation/clawflow.md:6:ClawFlow was renamed to [Task flow](/automation/taskflow).

automation/clawflow.md:10:- [Task flow](/automation/taskflow)

automation/tasks.md:341: [Task Flow](/automation/taskflow) is the flow orchestration layer above background tasks. A single flow may coordinate multiple tasks over its lifetime using managed or mirrored sync modes. Use openclaw tasks to inspect individual task records and openclaw tasks flow to inspect the orchestrating flow.

automation/tasks.md:343: See [Task Flow](/automation/taskflow) for details.

automation/tasks.md:372:- [Task Flow](/automation/taskflow) — flow orchestration above tasks

automation/index.md:91:See [Task Flow](/automation/taskflow).

automation/index.md:129:- [Task Flow](/automation/taskflow) — durable multi-step flow orchestration

TaskFlow

plugins/reference.md:128:| [webhooks](/plugins/reference/webhooks) | Authenticated inbound webhooks that bind external automation to OpenClaw TaskFlows. | @openclaw/webhooks<br />included in OpenClaw | plugin |

plugins/plugin-inventory.md:136:| [webhooks](/plugins/reference/webhooks) | Authenticated inbound webhooks that bind external automation to OpenClaw TaskFlows. | @openclaw/webhooks<br />included in OpenClaw | plugin |

plugins/sdk-migration.md:733: New: runtime.tasks.managedFlows keeps the managed TaskFlow mutation

plugins/webhooks.md:2:summary: "Webhooks plugin: authenticated TaskFlow ingress for trusted external automation"

plugins/webhooks.md:4: - You want to trigger or drive TaskFlows from an external system

plugins/webhooks.md:12:automation to OpenClaw TaskFlows.

plugins/webhooks.md:15:internal service to create and drive managed TaskFlows without writing a custom

plugins/webhooks.md:46: description: "Zapier TaskFlow bridge",

plugins/webhooks.md:60:- sessionKey: required session that owns the bound TaskFlows

plugins/webhooks.md:75:Each route is trusted to act with the TaskFlow authority of its configured

plugins/webhooks.md:78:This means the route can inspect and mutate TaskFlows owned by that session, so

plugins/webhooks.md:92:- Owner-bound TaskFlow access through api.runtime.tasks.managedFlows.bindSession(...)

plugins/webhooks.md:130:Creates a managed TaskFlow for the route's bound session.

plugins/webhooks.md:145:Creates a managed child task inside an existing managed TaskFlow.

plugins/webhooks.md:183: "error": "TaskFlow not found.",

plugins/sdk-testing.md:141:| createRuntimeTaskFlow | Create isolated runtime task-flow state |

plugins/reference/webhooks.md:2:summary: "Authenticated inbound webhooks that bind external automation to OpenClaw TaskFlows."

plugins/reference/webhooks.md:10:Authenticated inbound webhooks that bind external automation to OpenClaw TaskFlows.

automation/tasks.md:83: # Inspect TaskFlow state

cron

platforms/raspberry-pi.md:75:# Set timezone (important for cron/reminders)

tools/browser.md:212:- tabCleanup is best-effort cleanup for tabs opened by primary-agent browser sessions. Subagent, cron, and ACP lifecycle cleanup still closes their explicit tracked tabs at session end; primary sessions keep active tabs reusable, then close idle or excess tracked tabs in the background.

tools/subagents.md:402:| Source | subagent or cron |

tools/subagents.md:472: deny: ["gateway", "cron"],

tools/index.md:65:| cron / gateway | Manage scheduled jobs; inspect, patch, restart, or update the gateway | |

tools/index.md:150:| coding | group:fs, group:runtime, group:web, group:sessions, group:memory, cron, image, image_generate, music_generate, video_generate |

tools/index.md:157:browser, canvas, nodes, cron, and gateway control. Use tools.profile: "full"

tools/index.md:199:| group:automation | cron, gateway |

tools/acp-agents-setup.md:238:built-in tools such as cron:

tools/acp-agents-setup.md:248:- Exposes selected built-in OpenClaw tools. The initial server exposes cron.

tools/exec.md:93:- For work that should happen later or on a schedule, use cron instead of

install/migrating-hermes.md:80:- cron/

install/raspberry-pi.md:49: # Set timezone (important for cron and reminders)

install/docker.md:282:cron/runs/*.jsonl, installed plugin package roots, and rolling file logs

plugins/sdk-migration.md:355: cron delivery, and session routing. If your plugin owns custom target

plugins/codex-harness.md:625:cron, browser, nodes, gateway, heartbeat_respond, and web_search remain

plugins/sdk-overview.md:319:- cron_changed: observe gateway-owned cron lifecycle changes. Use event.job?.state?.nextRunAtMs and ctx.getCron?.() when syncing external wake schedulers, and keep OpenClaw as the source of truth for due checks and execution.

plugins/manifest.md:1330:- Env-var metadata (setup.providers[].envVars, deprecated providerAuthEnvVars, and channelEnvVars) is declarative only. Status, audit, cron delivery validation, and other read-only surfaces still apply plugin trust and effective activation policy before treating an env var as configured.

plugins/hooks.md:147:- cron_changed — observe gateway-owned cron lifecycle changes (added, updated, removed, started, finished, scheduled)

plugins/hooks.md:159: ctx.runId, ctx.jobId (set on cron-driven runs), and diagnostic ctx.trace

plugins/hooks.md:237:Cron-driven runs also expose ctx.jobId (the originating cron job id) so

plugins/hooks.md:367:cron inspection and updates. Use gateway_stop to clean up long-running

plugins/hooks.md:373:cron_changed fires for gateway-owned cron lifecycle events with a typed

plugins/sdk-subpaths.md:200: | plugin-sdk/cron-store-runtime | Cron store path/load/save helpers |

web/control-ui.md:111: - Cron jobs: list/add/edit/run/enable/disable + run history (cron.*).

web/control-ui.md:140: - Advanced edit controls include delete-after-run, clear agent override, cron exact/stagger options, agent model/thinking overrides, and best-effort delivery toggles.

web/control-ui.md:142: - Set cron.webhookToken to send a dedicated bearer token, if omitted the webhook is sent without an auth header.

web/control-ui.md:143: - Deprecated fallback: stored legacy jobs with notify: true can still use cron.webhook until migrated.

providers/openai.md:326: cron, sub-agent, and configured default-model runs do not fail with

providers/ollama.md:248:Isolated cron jobs do one extra local safety check before they start the agent

providers/ollama.md:250:Ollama provider and /api/tags is unreachable, OpenClaw records that cron run

providers/ollama.md:252:preflight is cached for 5 minutes, so multiple cron jobs pointed at the same

providers/ollama.md:660: localModelLean removes the browser, cron, and message tools from the agent surface, but it does not change Ollama's runtime context or thinking mode. Pair it with explicit params.num_ctx and params.thinking: false for small Qwen-style thinking models that loop or spend their response budget on hidden reasoning.

start/openclaw.md:226:- Cron + wakeups: [Cron jobs](/automation/cron-jobs)

start/docs-directory.md:61:- [Cron jobs](/automation/cron-jobs)

start/docs-directory.md:62:- [Webhooks](/automation/cron-jobs#webhooks)

start/docs-directory.md:63:- [Gmail hooks (Pub/Sub)](/automation/cron-jobs#gmail-pubsub-integration)

start/hubs.md:81:- [Webhooks](/automation/cron-jobs#webhooks)

start/hubs.md:82:- [Gmail Pub/Sub](/automation/cron-jobs#gmail-pubsub-integration)

start/hubs.md:112:- [Cron jobs](/automation/cron-jobs)

sandbox

platforms/mac/webchat.md:49:- The UI is optimized for chat sessions (not a full browser sandbox).

tools/slash-commands.md:144: - /exec host=<auto|sandbox|gateway|node> security=<deny|allowlist|full> ask=<off|on-miss|always> node=<id> shows or sets exec defaults.

tools/slash-commands.md:317:- Execution vs runtime: /status reports Execution for the effective sandbox path and Runtime for who is actually running the session: OpenClaw Pi Default, OpenAI Codex, a CLI backend, or an ACP backend.

tools/lobster.md:32:- Safety policy baked in: Timeouts, output caps, sandbox checks, and allowlists are enforced by the runtime, not each script.

tools/lobster.md:322:- Sandbox-aware — disabled when the tool context is sandboxed.

tools/exec-approvals.md:6: - Reviewing sandbox-escape prompts and their implications

tools/exec-approvals.md:12:a sandboxed agent run commands on a real host (gateway or node). A

tools/exec-approvals.md:186:- tools.exec.host=auto chooses where exec runs: sandbox when available, otherwise gateway.

tools/exec-approvals.md:189:- auto does not make gateway routing a free override from a sandboxed session. A per-call host=node request is allowed from auto; host=gateway is only allowed from auto when no sandbox runtime is active. For a stable non-auto default, set tools.exec.host or use /exec host=... explicitly.

tools/exec-approvals.md:409: <Card title="Sandboxing" href="/gateway/sandboxing" icon="box">

tools/exec-approvals.md:415: <Card title="Sandbox vs tool policy vs elevated" href="/gateway/sandbox-vs-tool-policy-vs-elevated" icon="sliders">

tools/plugin.md:407:runtime config, workspace, agent/session ids, sandbox policy, browser settings,

tools/exec-approvals-advanced.md:159:- For those workflows, prefer sandboxing, a separate host boundary, or an explicit trusted

tools/code-execution.md:2:summary: "code_execution -- run sandboxed remote Python analysis with xAI"

tools/code-execution.md:10:code_execution runs sandboxed remote Python analysis on xAI's Responses API.

tools/code-execution.md:14:- code_execution runs Python in xAI's remote sandbox

tools/browser.md:753: - target (sandbox | host | node) to select where the browser lives.

tools/browser.md:754: - In sandboxed sessions, target: "host" requires agents.defaults.sandbox.browser.allowHostControl=true.

tools/browser.md:755: - If target is omitted: sandboxed sessions default to sandbox, non-sandbox sessions default to host.

tools/browser.md:763:- [Sandboxing](/gateway/sandboxing) — browser control in sandboxed environments

tools/acp-agents.md:54:session must not be sandbox-blocked, and a runtime backend must be

tools/acp-agents.md:205: is enabled, the requester is not sandboxed, and an ACP runtime

tools/acp-agents.md:722:OpenClaw sandbox.

tools/acp-agents.md:728:- OpenClaw's sandbox policy does not wrap ACP harness execution.

tools/acp-agents.md:730:- Use runtime: "subagent" for sandbox-enforced OpenClaw-native work.

tools/acp-agents.md:736:- If the requester session is sandboxed, ACP spawns are blocked for both sessions_spawn({ runtime: "acp" }) and /acp spawn.

tools/acp-agents.md:737:- sessions_spawn with runtime: "acp" does not support sandbox: "require".

tools/acp-agents.md:827:| Sandboxed sessions cannot spawn ACP sessions ... | ACP runtime is host-side; requester session is sandboxed. | Use runtime="subagent" from sandboxed sessions, or run ACP spawn from a non-sandboxed session. |

tools/acp-agents.md:828:| sessions_spawn sandbox="require" is unsupported for runtime="acp" ... | sandbox="require" requested for ACP runtime. | Use runtime="subagent" for required sandboxing, or use ACP with sandbox="inherit" from a non-sandboxed session. |

tools/acp-agents.md:842:- [Multi-agent sandbox tools](/tools/multi-agent-sandbox-tools)

tools/browser-login.md:50:If the agent is sandboxed, the browser tool defaults to the sandbox. To allow host control:

tools/browser-login.md:56: sandbox: {

tools/browser-login.md:73:Or disable sandboxing for the agent that posts updates.

tools/web-fetch.md:139: provider from available credentials. Non-sandboxed web_fetch can use

tools/subagents.md:20:- Keep sub-agents isolated by default (session separation + optional sandboxing).

tools/subagents.md:107: - OpenClaw hides runtime: "acp" until ACP is enabled, the requester is not sandboxed, and a backend plugin such as acpx is loaded. runtime: "acp" expects an external ACP harness id, or an agents.list[] entry with runtime.type="acp"; use the default sub-agent runtime for normal OpenClaw config agents from agents_list.

tools/subagents.md:135:work. Channel/group, provider, sandbox, and per-agent allow/deny policies can

tools/subagents.md:183:<ParamField path="sandbox" type='"inherit" | "require"' default="inherit">

tools/subagents.md:184: require rejects spawn unless the target child runtime is sandboxed.

tools/subagents.md:264:If the requester session is sandboxed, sessions_spawn rejects targets

media

platforms/windows.md:41:- if Scheduled Task creation is denied, OpenClaw falls back to a per-user Startup-folder login item and starts the gateway immediately

platforms/mac/menu-bar.md:21:- Priority: main always wins. If main is active, its state is shown immediately. If main is idle, the most recently active non‑main session is shown. We do not flip‑flop mid‑activity; we only switch when the current session goes idle or main becomes active.

platforms/mac/menu-bar.md:86:- Trigger main session job: verify icon switches immediately and status row shows main label.

platforms/mac/voice-overlay.md:15:- Wake-word alone still auto-sends on silence; push-to-talk sends immediately on release.

platforms/mac/voice-overlay.md:38: - Apply a short cooldown to the wake runtime after push-to-talk finishes so wake-word doesn’t immediately retrigger.

platforms/mac/voicewake.md:13:- Push-to-talk (Right Option hold): hold the right Option key to capture immediately—no trigger needed. The overlay appears while held; releasing finalizes and forwards after a short delay so you can tweak text.

platforms/mac/voicewake.md:42:- Capture pipeline lives in VoicePushToTalk: starts Speech immediately, streams partials to the overlay, and calls VoiceWakeForwarder on release.

platforms/mac/logging.md:56:- Optionally run sudo log config --reload to force logd to drop the override immediately.

platforms/mac/dev-setup.md:40:> Note: Ad-hoc signed apps may trigger security prompts. If the app crashes immediately with "Abort trap 6", see the [Troubleshooting](#troubleshooting) section.

tools/slash-commands.md:32: They run immediately, are stripped before the model sees the message, and the remaining text continues through the normal flow.

tools/slash-commands.md:78: Controls how long bash waits before switching to background mode (0 backgrounds immediately).

tools/slash-commands.md:274: - /model persists the new session model immediately.

tools/slash-commands.md:282: - Fast path: command-only messages from allowlisted senders are handled immediately (bypass queue + model).

tools/slash-commands.md:358:Overrides apply immediately to new config reads, but do not write to openclaw.json. Use /debug reset to clear all overrides and return to the on-disk config.

tools/plugin.md:13:voice, media-understanding, image generation, video generation, web fetch, web

tools/plugin.md:535:explicit install is immediately loadable after restart.

tools/exec-approvals-advanced.md:162:When approvals are required, the exec tool returns immediately with an approval id. Use that id to

tools/thinking.md:125:- Picking another level writes the session override immediately via sessions.patch; it does not wait for the next send and it is not a one-shot thinkingOnce override.

tools/kimi-search.md:84:payload, search_results, or citation URLs. If Kimi stops immediately with a

tools/web-fetch.md:20:call it immediately:

tools/video-generation.md:25:- generate — text-to-video requests with no reference media.

tools/video-generation.md:62:1. OpenClaw submits the request to the provider and immediately returns a task id.

tools/video-generation.md:75:the tool falls back to inline generation and returns the final media path

tools/video-generation.md:78:Generated video files are saved under OpenClaw-managed media storage when

tools/video-generation.md:80:the video media limit, and agents.defaults.mediaMaxMb raises it for

tools/video-generation.md:248:- No reference media → generate

tools/video-generation.md:303:Set agents.defaults.mediaGenerationAutoProviderFallback: false to use

tools/video-generation.md:477:pnpm test:live:media video

tools/video-generation.md:493:pnpm test:live:media video --video-providers fal

tools/video-generation.md:497:transform modes the shared sweep can exercise safely with local media:

tools/subagents.md:78: - The spawn command is non-blocking; it returns a run id immediately.

tools/subagents.md:181: "delete" archives immediately after announce (still keeps the transcript via rename).

tools/subagents.md:278:- cleanup: "delete" archives immediately after announce (still keeps the transcript via rename).

tools/subagents.md:548:- sessions_spawn is always non-blocking: it returns { status: "accepted", runId, childSessionKey } immediately.

tools/music-generation.md:61: the final media path in the tool result.

tools/music-generation.md:178: started/task response immediately, and posts the finished track later in

tools/music-generation.md:242:Set agents.defaults.mediaGenerationAutoProviderFallback: false to use only

tools/music-generation.md:319:pnpm test:live:media music

tools/image-generation.md:12:configured providers. Generated images are delivered automatically as media

tools/image-generation.md:215: agents.defaults.mediaGenerationAutoProviderFallback: false to use only

browser

platforms/android.md:149:no requested scopes. Operator/browser pairing and any role, scope, metadata, or

platforms/linux.md:118:- OpenClaw-launched browser/Chrome processes.

platforms/raspberry-pi.md:176:Then open the printed Dashboard URL in your local browser.

platforms/raspberry-pi.md:294:| Chromium (browser) | ✅ | sudo apt install chromium-browser |

platforms/mac/webchat.md:49:- The UI is optimized for chat sessions (not a full browser sandbox).

platforms/mac/remote.md:32:possible; if you need browser control from that Mac, install/start it with

platforms/mac/remote.md:34:openclaw node run ... in the foreground), then target that browser-capable

platforms/ios.md:63:no requested scopes. Operator/browser pairing and any role, scope, metadata, or

tools/clawhub.md:199: Site base URL (browser login).

tools/clawhub.md:216: clawhub login # browser flow

tools/clawhub.md:225: - --label <label> — label stored for browser login tokens (default: CLI token).

tools/clawhub.md:226: - --no-browser — do not open a browser (requires --token).

tools/plugin.md:242: - browser — bundled browser plugin for the browser tool, openclaw browser CLI, browser.request gateway method, browser runtime, and default browser control service (enabled by default; disable before replacing it)

tools/plugin.md:407:runtime config, workspace, agent/session ids, sandbox policy, browser settings,

tools/plugin.md:524:bundled model providers, bundled speech providers, and the bundled browser

tools/thinking.md:127:- The picker uses thinkingLevels returned by the gateway session row/defaults, with thinkingOptions kept as a legacy label list. The browser UI does not keep its own provider regex list; plugins own model-specific level sets.

tools/browser-wsl2-windows-remote-cdp-troubleshooting.md:5: - Seeing overlapping browser/control-ui errors across WSL2 and Windows

tools/browser-wsl2-windows-remote-cdp-troubleshooting.md:10:In the common split-host setup, OpenClaw Gateway runs inside WSL2, Chrome runs on Windows, and browser control must cross the WSL2 and Windows boundary. The layered failure pattern from [issue #39369](https://github.com/openclaw/openclaw/issues/39369) means several independent problems can show up at once, which makes the wrong layer look broken first.

tools/browser-wsl2-windows-remote-cdp-troubleshooting.md:12:## Choose the right browser mode first

tools/browser-wsl2-windows-remote-cdp-troubleshooting.md:18:Use a remote browser profile that points from WSL2 to a Windows Chrome CDP endpoint.

tools/browser-wsl2-windows-remote-cdp-troubleshooting.md:24:- you need browser control to cross the WSL2/Windows boundary

tools/browser-wsl2-windows-remote-cdp-troubleshooting.md:33:- you want the local signed-in browser state

tools/browser-wsl2-windows-remote-cdp-troubleshooting.md:34:- you do not need cross-host browser transport

tools/browser-wsl2-windows-remote-cdp-troubleshooting.md:45:- Windows opens the Control UI in a normal browser at http://127.0.0.1:18789/`