vrr/openclaw
2
0
Fork 0
mirror of https://github.com/openclaw/openclaw.git synced 2026-05-19 07:42:04 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 19

docs: reorganize tools automation nav (#80116)

Browse source 
* docs: reorganize tools automation nav

* docs: add nav spec glossary terms

* docs: refresh nav spec validation

* docs: keep capabilities nav grouped

* docs: refactor tools overview

* docs: restore tools overview coverage

* add doc refactor skill

* docs: mark refactored docs schema

* docs: remove refactor specs from pr

* docs: rename tools overview header
This commit is contained in:
Kevin Lin 2026-05-11 15:59:27 -07:00 • committed by GitHub
parent b348cdd1f1
commit f5b0eca12a
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
16 changed files with 219 additions and 214 deletions
Download patch file Download diff file Expand all files Collapse all files

2
.gitignore vendored
View file

@ -118,6 +118,8 @@ USER.md
!.agents/skills/gitcrawl/
!.agents/skills/gitcrawl/**
!.agents/skills/openclaw-docs/**
!.agents/skills/openclaw-refactor-docs/
!.agents/skills/openclaw-refactor-docs/**
!.agents/skills/openclaw-debugging/
!.agents/skills/openclaw-debugging/**
!.agents/skills/openclaw-ghsa-maintainer/

1
CHANGELOG.md
View file

@ -15,6 +15,7 @@ Docs: https://docs.openclaw.ai
- Runtime/Fly: detect Fly Machines as container environments from their runtime env vars, so gateway bind and Bonjour defaults match remote container launches. (#80209) Thanks @liorb-mountapps.
- Providers/fal: route GPT Image 2 and Nano Banana 2 reference-image edit requests to `/edit` with `image_urls` array, enforce NB2 edit geometry using `aspect_ratio` and `resolution` params, lift Fal edit mode input-image caps to 10 for GPT Image 2 and 14 for Nano Banana 2, and allow aspect-ratio hints in edit mode. (#77295) Thanks @leoge007.
- Control UI: show a plain HTML recovery panel when the app module never registers, giving blank dashboard pages a retry path and browser-extension troubleshooting link. Fixes #44107. Thanks @BunsDev.
- Docs: rename the broad tools nav to Capabilities, keep automation and agent coordination as sections, and keep the tools overview focused on tools, skills, and plugins. https://docs.openclaw.ai/tools
- Build: enable additional low-churn oxlint rules for promise, TypeScript, and runtime footgun checks.
- Build: enable stricter Vitest lint rules for focused, disabled, conditional, hook, matcher, and expectation hazards.

60
docs/.i18n/glossary.zh-CN.json
View file

@ -867,6 +867,66 @@
"source": "/cli/config",
"target": "/cli/config"
},
{
"source": "Automation",
"target": "自动化"
},
{
"source": "Tools, skills, and plugins",
"target": "工具、技能和插件"
},
{
"source": "Capabilities",
"target": "能力"
},
{
"source": "Overview",
"target": "概览"
},
{
"source": "Tools, skills, and plugins overview",
"target": "工具、技能和插件概览"
},
{
"source": "Tools overview",
"target": "工具概览"
},
{
"source": "Automation overview",
"target": "自动化概览"
},
{
"source": "Sub-agents",
"target": "子智能体"
},
{
"source": "ACP agents",
"target": "ACP 智能体"
},
{
"source": "Tools and custom providers",
"target": "工具和自定义提供商"
},
{
"source": "Exec approvals",
"target": "Exec 审批"
},
{
"source": "Elevated exec",
"target": "提升权限的 Exec"
},
{
"source": "Sandbox vs tool policy vs elevated",
"target": "沙箱、工具策略和提升权限"
},
{
"source": "Per-agent sandbox and tool restrictions",
"target": "按 Agent 配置的沙箱和工具限制"
},
{
"source": "Agents",
"target": "智能体"
},
{
"source": "fs-safe Cleanup Plan",
"target": "fs-safe Cleanup Plan"

2
docs/automation/cron-jobs.md
View file

@ -486,7 +486,7 @@ openclaw doctor
## Related
- [Automation & Tasks](/automation) — all automation mechanisms at a glance
- [Automation](/automation) — all automation mechanisms at a glance
- [Background Tasks](/automation/tasks) — task ledger for cron executions
- [Heartbeat](/gateway/heartbeat) — periodic main-session turns
- [Timezone](/concepts/timezone) — timezone configuration

2
docs/automation/cron-vs-heartbeat.md
View file

@ -3,7 +3,7 @@ summary: "Redirect to /automation"
title: "Cron vs heartbeat"
---
The decision guide for cron vs heartbeat lives under [Automation and tasks](/automation).
The decision guide for cron vs heartbeat lives under [Automation](/automation).
## Related

3
docs/automation/index.md
View file

@ -1,10 +1,11 @@
---
doc-schema-version: 1
summary: "Overview of automation mechanisms: tasks, cron, hooks, standing orders, and Task Flow"
read_when:
- Deciding how to automate work with OpenClaw
- Choosing between heartbeat, cron, commitments, hooks, and standing orders
- Looking for the right automation entry point
title: "Automation and tasks"
title: "Automation"
---
OpenClaw runs work in the background through tasks, scheduled jobs, inferred

2
docs/automation/standing-orders.md
View file

@ -243,7 +243,7 @@ Each program should have:
## Related
- [Automation and tasks](/automation): all automation mechanisms at a glance.
- [Automation](/automation): all automation mechanisms at a glance.
- [Cron jobs](/automation/cron-jobs): schedule enforcement for standing orders.
- [Hooks](/automation/hooks): event-driven scripts for agent lifecycle events.
- [Webhooks](/automation/cron-jobs#webhooks): inbound HTTP event triggers.

4
docs/automation/tasks.md
View file

@ -9,7 +9,7 @@ sidebarTitle: "Background tasks"
---
<Note>
Looking for scheduling? See [Automation and tasks](/automation) for choosing the right mechanism. This page is the activity ledger for background work, not the scheduler.
Looking for scheduling? See [Automation](/automation) for choosing the right mechanism. This page is the activity ledger for background work, not the scheduler.
</Note>
Background tasks track work that runs **outside your main conversation session**: ACP runs, subagent spawns, isolated cron job executions, and CLI-initiated operations.
@ -367,7 +367,7 @@ A sweeper runs every **60 seconds** and handles four things:
## Related
- [Automation & Tasks](/automation) - all automation mechanisms at a glance
- [Automation](/automation) - all automation mechanisms at a glance
- [CLI: Tasks](/cli/tasks) - CLI command reference
- [Heartbeat](/gateway/heartbeat) - periodic main-session turns
- [Scheduled Tasks](/automation/cron-jobs) - scheduling background work

4
docs/docs.json
View file

@ -1179,7 +1179,7 @@
]
},
{
"tab": "Tools & Plugins",
"tab": "Capabilities",
"groups": [
{
"group": "Overview",
@ -1250,7 +1250,7 @@
]
},
{
"group": "Automation and tasks",
"group": "Automation",
"pages": [
"automation/index",
"automation/cron-jobs",

4
docs/gateway/heartbeat.md
View file

@ -8,7 +8,7 @@ sidebarTitle: "Heartbeat"
---
<Note>
**Heartbeat vs cron?** See [Automation & Tasks](/automation) for guidance on when to use each.
**Heartbeat vs cron?** See [Automation](/automation) for guidance on when to use each.
</Note>
Heartbeat runs **periodic agent turns** in the main session so the model can surface anything that needs attention without spamming you.
@ -485,7 +485,7 @@ Current heartbeats preserve the shared session's existing runtime model after th
## Related
- [Automation & Tasks](/automation) — all automation mechanisms at a glance
- [Automation](/automation) — all automation mechanisms at a glance
- [Background Tasks](/automation/tasks) — how detached work is tracked
- [Timezone](/concepts/timezone) — how timezone affects heartbeat scheduling
- [Troubleshooting](/automation/cron-jobs#troubleshooting) — debugging automation issues

2
docs/help/faq-first-run.md
View file

@ -74,7 +74,7 @@ and troubleshooting see the main [FAQ](/help/faq).
In task mode, due timestamps are only advanced after a real heartbeat run
completes. Skipped runs do not mark tasks as completed.
Docs: [Heartbeat](/gateway/heartbeat), [Automation & Tasks](/automation).
Docs: [Heartbeat](/gateway/heartbeat), [Automation](/automation).
</Accordion>

4
docs/help/faq.md
View file

@ -258,7 +258,7 @@ lives on the [First-run FAQ](/help/faq-first-run).
openclaw cron runs --id <jobId> --limit 50
```
Docs: [Cron jobs](/automation/cron-jobs), [Automation & Tasks](/automation).
Docs: [Cron jobs](/automation/cron-jobs), [Automation](/automation).
</Accordion>
@ -344,7 +344,7 @@ lives on the [First-run FAQ](/help/faq-first-run).
- **Heartbeat** for "main session" periodic checks.
- **Isolated jobs** for autonomous agents that post summaries or deliver to chats.
Docs: [Cron jobs](/automation/cron-jobs), [Automation & Tasks](/automation),
Docs: [Cron jobs](/automation/cron-jobs), [Automation](/automation),
[Heartbeat](/gateway/heartbeat).
</Accordion>

4
docs/refactor/ingress-core.md
View file

@ -335,7 +335,7 @@ Each work package records:
- compatibility code is isolated to SDK/core seams
- bundled plugins consume ingress projections or generic outcomes directly
- plugin production LOC is at least 1,500 net negative against `origin/main`
- core production LOC is <= +1,500, or any excess is paid for while total stays
<= +2,000
- core production LOC is `<= +1,500`, or any excess is paid for while total
stays `<= +2,000`
- representative tests cover redaction, route, command/event, activation,
access-group, and channel-specific fallback behavior

2
docs/start/hubs.md
View file

@ -109,7 +109,7 @@ Use these hubs to discover every page, including deep dives and reference docs t
- [PDF tool](/tools/pdf)
- [Elevated mode](/tools/elevated)
- [Cron jobs](/automation/cron-jobs)
- [Automation & Tasks](/automation)
- [Automation](/automation)
- [Thinking + verbose](/tools/thinking)
- [Models](/concepts/models)
- [Sub-agents](/tools/subagents)

335
docs/tools/index.md
View file

@ -1,237 +1,178 @@
---
summary: "OpenClaw tools and plugins overview: what the agent can do and how to extend it"
doc-schema-version: 1
summary: "OpenClaw tools, skills, and plugins overview: what agents can call and how to extend them"
read_when:
- You want to understand what tools OpenClaw provides
- You need to configure, allow, or deny tools
- You are deciding between built-in tools, skills, and plugins
title: "Tools and plugins"
- You need the right docs entry point for tool policy, automation, or agent coordination
title: "Overview"
---
Everything the agent does beyond generating text happens through **tools**.
Tools are how the agent reads files, runs commands, browses the web, sends
messages, and interacts with devices.
Use this page to choose the right Capabilities surface. **Tools** are callable
actions, **skills** teach agents how to work, and **plugins** add runtime
capabilities such as tools, providers, channels, hooks, and packaged skills.
## Tools, skills, and plugins
This is an overview and routing page. For exhaustive tool policy, defaults,
group membership, provider restrictions, and configuration fields, use
[Tools and custom providers](/gateway/config-tools).
OpenClaw has three layers that work together:
## Start here
For most agents, start with the built-in tool categories, then adjust policy
only when the agent should see fewer tools or needs explicit host access.
| If you need to... | Use this first | Then read |
| ------------------------------------------- | ---------------------------------------------- | ----------------------------------------------------------------------- |
| Let an agent act with existing capabilities | [Built-in tools](#built-in-tool-categories) | [Tool categories](#built-in-tool-categories) |
| Control what an agent can call | [Tool policy](#configure-access-and-approvals) | [Tools and custom providers](/gateway/config-tools) |
| Teach an agent a workflow | [Skills](#choose-tools-skills-or-plugins) | [Skills](/tools/skills) and [Creating skills](/tools/creating-skills) |
| Add a new integration or runtime surface | [Plugins](#extend-capabilities) | [Plugins](/tools/plugin) and [Build plugins](/plugins/building-plugins) |
| Run work later or in the background | [Automation](/automation) | [Automation overview](/automation) |
| Coordinate multiple agents or harnesses | [Sub-agents](/tools/subagents) | [ACP agents](/tools/acp-agents) and [Agent send](/tools/agent-send) |
| Search a large PI tool catalog | [Tool Search](/tools/tool-search) | [Tool Search](/tools/tool-search) |
## Choose tools, skills, or plugins
<Steps>
<Step title="Tools are what the agent calls">
A tool is a typed function the agent can invoke (e.g. `exec`, `browser`,
`web_search`, `message`). OpenClaw ships a set of **built-in tools** and
plugins can register additional ones.
<Step title="Use a tool when the agent needs to act">
A tool is a typed function the agent can call, such as `exec`, `browser`,
`web_search`, `message`, or `image_generate`. Use tools when the agent
needs to read data, change files, send messages, call a provider, or operate
another system. Visible tools are sent to the model as structured function
definitions.
The agent sees tools as structured function definitions sent to the model API.
The model only sees tools that survive the active profile, allow/deny
policy, provider restrictions, sandbox state, channel permissions, and
plugin availability.
</Step>
<Step title="Skills teach the agent when and how">
A skill is a markdown file (`SKILL.md`) injected into the system prompt.
Skills give the agent context, constraints, and step-by-step guidance for
using tools effectively. Skills live in your workspace, in shared folders,
or ship inside plugins.
<Step title="Use a skill when the agent needs instructions">
A skill is a `SKILL.md` instruction pack loaded into the agent prompt. Use a
skill when the agent already has the tools it needs, but needs a repeatable
workflow, review rubric, command sequence, or operating constraint.
[Skills reference](/tools/skills) | [Creating skills](/tools/creating-skills)
Skills can live in a workspace, shared skill directory, managed OpenClaw
skill root, or plugin package.
[Skills](/tools/skills) | [Creating skills](/tools/creating-skills) | [Skills config](/tools/skills-config)
</Step>
<Step title="Plugins package everything together">
A plugin is a package that can register any combination of capabilities:
channels, model providers, tools, skills, speech, realtime transcription,
realtime voice, media understanding, image generation, video generation,
web fetch, web search, and more. Some plugins are **core** (shipped with
OpenClaw), others are **external** (published on npm by the community).
<Step title="Use a plugin when OpenClaw needs a new capability">
A plugin can add tools, skills, channels, model providers, speech, realtime
voice, media generation, web search, web fetch, hooks, and other runtime
capabilities. Use a plugin when the capability has code, credentials,
lifecycle hooks, manifest metadata, or installable packaging. Existing
plugins can be installed from ClawHub, npm, git, local directories, or
archives.
[Install and configure plugins](/tools/plugin) | [Build your own](/plugins/building-plugins)
[Install and configure plugins](/tools/plugin) | [Build plugins](/plugins/building-plugins) | [Plugin SDK](/plugins/sdk-overview)
</Step>
</Steps>
## Built-in tools
## Built-in tool categories
These tools ship with OpenClaw and are available without installing any plugins:
The table lists representative tools so you can recognize the surface. It is
not the full policy reference. For exact groups, defaults, and allow/deny
semantics, use [Tools and custom providers](/gateway/config-tools).
| Tool | What it does | Page |
| ------------------------------------------ | --------------------------------------------------------------------- | ------------------------------------------------------------ |
| `exec` / `process` | Run shell commands, manage background processes | [Exec](/tools/exec), [Exec Approvals](/tools/exec-approvals) |
| `code_execution` | Run sandboxed remote Python analysis | [Code Execution](/tools/code-execution) |
| `browser` | Control a Chromium browser (navigate, click, screenshot) | [Browser](/tools/browser) |
| `web_search` / `x_search` / `web_fetch` | Search the web, search X posts, fetch page content | [Web](/tools/web), [Web Fetch](/tools/web-fetch) |
| `read` / `write` / `edit` | File I/O in the workspace | |
| `apply_patch` | Multi-hunk file patches | [Apply Patch](/tools/apply-patch) |
| `message` | Send messages across all channels | [Agent Send](/tools/agent-send) |
| `nodes` | Discover and target paired devices | |
| `cron` / `gateway` | Manage scheduled jobs; inspect, patch, restart, or update the gateway | |
| `image` / `image_generate` | Analyze or generate images | [Image Generation](/tools/image-generation) |
| `music_generate` | Generate music tracks | [Music Generation](/tools/music-generation) |
| `video_generate` | Generate videos | [Video Generation](/tools/video-generation) |
| `tts` | One-shot text-to-speech conversion | [TTS](/tools/tts) |
| `sessions_*` / `subagents` / `agents_list` | Session management, status, and sub-agent orchestration | [Sub-agents](/tools/subagents) |
| `session_status` | Lightweight `/status`-style readback and session model override | [Session Tools](/concepts/session-tool) |
For image work, use `image` for analysis and `image_generate` for generation or editing. If you target `openai/*`, `google/*`, `fal/*`, or another non-default image provider, configure that provider's auth/API key first.
For music work, use `music_generate`. If you target `google/*`, `minimax/*`, or another non-default music provider, configure that provider's auth/API key first.
For video work, use `video_generate`. If you target `qwen/*` or another non-default video provider, configure that provider's auth/API key first.
For workflow-driven audio generation, use `music_generate` when a plugin such as
ComfyUI registers it. This is separate from `tts`, which is text-to-speech.
`session_status` is the lightweight status/readback tool in the sessions group.
It answers `/status`-style questions about the current session and can
optionally set a per-session model override; `model=default` clears that
override. Like `/status`, it can backfill sparse token/cache counters and the
active runtime model label from the latest transcript usage entry.
`gateway` is the owner-only runtime tool for gateway operations:
- `config.schema.lookup` for one path-scoped config subtree before edits
- `config.get` for the current config snapshot + hash
- `config.patch` for partial config updates with restart
- `config.apply` only for full-config replacement
- `update.run` for explicit self-update + restart
For partial changes, prefer `config.schema.lookup` then `config.patch`. Use
`config.apply` only when you intentionally replace the entire config.
For broader config docs, read [Configuration](/gateway/configuration) and
[Configuration reference](/gateway/configuration-reference).
The tool also refuses to change `tools.exec.ask` or `tools.exec.security`;
legacy `tools.bash.*` aliases normalize to the same protected exec paths.
### Plugin-provided tools
Plugins can register additional tools. Some examples:
- [Canvas](/plugins/reference/canvas) — experimental bundled plugin for node Canvas control and A2UI rendering
- [Diffs](/tools/diffs) — diff viewer and renderer
- [LLM Task](/tools/llm-task) — JSON-only LLM step for structured output
- [Lobster](/tools/lobster) — typed workflow runtime with resumable approvals
- [Music Generation](/tools/music-generation) — shared `music_generate` tool with workflow-backed providers
- [OpenProse](/prose) — markdown-first workflow orchestration
- [Tokenjuice](/tools/tokenjuice) — compact noisy `exec` and `bash` tool results
Plugin tools are still authored with `api.registerTool(...)` and declared in
the plugin manifest's `contracts.tools` list. OpenClaw captures the validated
tool descriptor during discovery and caches it by plugin source and contract, so
later tool planning can skip plugin runtime loading. Tool execution still loads
the owning plugin and calls the live registered implementation.
[Tool Search](/tools/tool-search) is the compact surface
for large catalogs. Instead of putting every OpenClaw, MCP, or client tool
schema into the prompt, OpenClaw can give the model an isolated Node runtime
with `openclaw.tools.search`, `openclaw.tools.describe`, and
`openclaw.tools.call`. Calls still flow back through the Gateway, so tool
policy, approvals, hooks, and session logs remain authoritative.
## Tool configuration
### Allow and deny lists
Control which tools the agent can call via `tools.allow` / `tools.deny` in
config. Deny always wins over allow.
```json5
{
tools: {
allow: ["group:fs", "browser", "web_search"],
deny: ["exec"],
},
}
```
OpenClaw fails closed when an explicit allowlist resolves to no callable tools.
For example, `tools.allow: ["query_db"]` only works if a loaded plugin actually
registers `query_db`. If no built-in, plugin, or bundled MCP tool matches the
allowlist, the run stops before the model call instead of continuing as a
text-only run that could hallucinate tool results.
### Tool profiles
`tools.profile` sets a base allowlist before `allow`/`deny` is applied.
Per-agent override: `agents.list[].tools.profile`.
| Profile | What it includes |
| ----------- | ------------------------------------------------------------------------------------------------------------------------------------------------- |
| `full` | All core and optional plugin tools; unrestricted baseline for broader command/control access |
| `coding` | `group:fs`, `group:runtime`, `group:web`, `group:sessions`, `group:memory`, `cron`, `image`, `image_generate`, `music_generate`, `video_generate` |
| `messaging` | `group:messaging`, `sessions_list`, `sessions_history`, `sessions_send`, `session_status` |
| `minimal` | `session_status` only |
| Category | Use when the agent needs to... | Representative tools | Read next |
| ---------------------- | ----------------------------------------------------------------------------- | -------------------------------------------------------------------- | ---------------------------------------------------------------------- |
| Runtime | Run commands, manage processes, or use provider-backed Python analysis | `exec`, `process`, `code_execution` | [Exec](/tools/exec), [Code execution](/tools/code-execution) |
| Files | Read and change workspace files | `read`, `write`, `edit`, `apply_patch` | [Apply patch](/tools/apply-patch) |
| Web | Search the web, search X posts, or fetch readable page content | `web_search`, `x_search`, `web_fetch` | [Web tools](/tools/web), [Web fetch](/tools/web-fetch) |
| Browser | Operate a browser session | `browser` | [Browser](/tools/browser) |
| Messaging and channels | Send replies or channel actions | `message` | [Agent send](/tools/agent-send) |
| Sessions and agents | Inspect sessions, delegate work, steer another run, or report status | `sessions_*`, `subagents`, `agents_list`, `session_status` | [Sub-agents](/tools/subagents), [Session tool](/concepts/session-tool) |
| Automation | Schedule work or respond to background events | `cron`, `heartbeat_respond` | [Automation](/automation) |
| Gateway and nodes | Inspect Gateway state or paired target devices | `gateway`, `nodes` | [Gateway configuration](/gateway/configuration), [Nodes](/nodes) |
| Media | Analyze, generate, or speak media | `image`, `image_generate`, `music_generate`, `video_generate`, `tts` | [Media overview](/tools/media-overview) |
| Large PI catalogs | Search and call many eligible tools without sending every schema to the model | `tool_search_code`, `tool_search`, `tool_describe` | [Tool Search](/tools/tool-search) |
<Note>
`tools.profile: "messaging"` is intentionally narrow for channel-focused
agents. It leaves out broader command/control tools such as filesystem, runtime,
browser, canvas, nodes, cron, and gateway control. Use `tools.profile: "full"`
as the unrestricted baseline for broader command/control access, then trim
access with `tools.allow` / `tools.deny` when needed.
Tool Search is an experimental PI-agent surface. Codex harness runs use
Codex-native code mode, native tool search, deferred dynamic tools, and nested
tool calls instead of `tools.toolSearch`.
</Note>
`coding` includes lightweight web tools (`web_search`, `web_fetch`, `x_search`)
but not the full browser-control tool. Browser automation can drive real
sessions and logged-in profiles, so add it explicitly with
`tools.alsoAllow: ["browser"]` or a per-agent
`agents.list[].tools.alsoAllow: ["browser"]`.
## Plugin-provided tools
<Note>
Configuring `tools.exec` or `tools.fs` under a restrictive profile (`messaging`, `minimal`) does not implicitly widen the profile's allowlist. Add explicit `tools.alsoAllow` entries (for example `["exec", "process"]` for exec, or `["read", "write", "edit"]` for fs) when you want a restrictive profile to use those configured sections. OpenClaw logs a startup warning when a config section is present without a matching `alsoAllow` grant.
</Note>
Plugins can register additional tools. Plugin authors wire tools through
`api.registerTool(...)` and the manifest's `contracts.tools`; use
[Plugin SDK](/plugins/sdk-overview) and [Plugin manifest](/plugins/manifest)
for contract details.
The `coding` and `messaging` profiles also allow configured bundle MCP tools
under the plugin key `bundle-mcp`. Add `tools.deny: ["bundle-mcp"]` when you
want a profile to keep its normal built-ins but hide all configured MCP tools.
The `minimal` profile does not include bundle MCP tools.
Common plugin-provided tools include:
Example (broadest tool surface by default):
- [Diffs](/tools/diffs) for rendering file and markdown diffs
- [LLM Task](/tools/llm-task) for JSON-only workflow steps
- [Lobster](/tools/lobster) for typed workflows with resumable approvals
- [Tokenjuice](/tools/tokenjuice) for compacting noisy `exec` and `bash` tool
output
- [Tool Search](/tools/tool-search) for discovering and calling large tool
catalogs without putting every schema in the prompt
- [Canvas](/plugins/reference/canvas) for node Canvas control and A2UI
rendering
```json5
{
tools: {
profile: "full",
},
}
```
## Configure access and approvals
### Tool groups
Tool policy is enforced before the model call. If policy removes a tool, the
model does not receive that tool's schema for the turn. A run can lose tools
because of global config, per-agent config, channel policy, provider
restrictions, sandbox rules, owner-only gating, or plugin availability.
Use `group:*` shorthands in allow/deny lists:
- [Tools and custom providers](/gateway/config-tools) documents tool profiles,
allow/deny lists, provider-specific restrictions, loop detection, and
provider-backed tool settings.
- [Exec approvals](/tools/exec-approvals) documents host command approval
policy.
- [Elevated exec](/tools/elevated) documents controlled execution outside the
sandbox.
- [Sandbox vs tool policy vs elevated](/gateway/sandbox-vs-tool-policy-vs-elevated) explains which layer controls file and process access.
- [Per-agent sandbox and tool restrictions](/tools/multi-agent-sandbox-tools)
documents agent-specific restrictions for delegated runs.
| Group | Tools |
| ------------------ | --------------------------------------------------------------------------------------------------------- |
| `group:runtime` | exec, process, code_execution (`bash` is accepted as an alias for `exec`) |
| `group:fs` | read, write, edit, apply_patch |
| `group:sessions` | sessions_list, sessions_history, sessions_send, sessions_spawn, sessions_yield, subagents, session_status |
| `group:memory` | memory_search, memory_get |
| `group:web` | web_search, x_search, web_fetch |
| `group:ui` | browser, canvas when the bundled Canvas plugin is enabled |
| `group:automation` | heartbeat_respond, cron, gateway |
| `group:messaging` | message |
| `group:nodes` | nodes |
| `group:agents` | agents_list, update_plan |
| `group:media` | image, image_generate, music_generate, video_generate, tts |
| `group:openclaw` | All built-in OpenClaw tools (excludes plugin tools) |
## Extend capabilities
`sessions_history` returns a bounded, safety-filtered recall view. It strips
thinking tags, `<relevant-memories>` scaffolding, plain-text tool-call XML
payloads (including `<tool_call>...</tool_call>`,
`<function_call>...</function_call>`, `<tool_calls>...</tool_calls>`,
`<function_calls>...</function_calls>`, and truncated tool-call blocks),
downgraded tool-call scaffolding, leaked ASCII/full-width model control
tokens, and malformed MiniMax tool-call XML from assistant text, then applies
redaction/truncation and possible oversized-row placeholders instead of acting
as a raw transcript dump.
Choose the extension path by the job you need OpenClaw to do:
### Provider-specific restrictions
- Install or manage an existing plugin with [Plugins](/tools/plugin).
- Build a new integration, provider, channel, tool, or hook with
[Build plugins](/plugins/building-plugins).
- Add or tune reusable agent instructions with [Skills](/tools/skills) and
[Creating skills](/tools/creating-skills).
- Package reusable workflow material with
[Skill workshop](/plugins/skill-workshop) when the workflow belongs in a
plugin-distributed skill bundle.
- Use [Plugin SDK](/plugins/sdk-overview) and [Plugin manifest](/plugins/manifest) when you need implementation contracts.
Use `tools.byProvider` to restrict tools for specific providers without
changing global defaults:
## Troubleshoot missing tools
```json5
{
tools: {
profile: "coding",
byProvider: {
"google-antigravity": { profile: "minimal" },
},
},
}
```
If the model cannot see or call a tool, start with the effective policy for the
current turn:
1. Check the active profile, `tools.allow`, and `tools.deny` in
[Tools and custom providers](/gateway/config-tools).
2. Check provider-specific restrictions in
[Tools and custom providers](/gateway/config-tools) and confirm the selected
[model provider](/concepts/model-providers) supports the tool shape.
3. Check channel permissions, sandbox state, and elevated access with
[Sandbox vs tool policy vs elevated](/gateway/sandbox-vs-tool-policy-vs-elevated) and [Elevated exec](/tools/elevated).
4. Check whether the owning plugin is installed and enabled in
[Plugins](/tools/plugin).
5. For delegated runs, check per-agent restrictions in
[Per-agent sandbox and tool restrictions](/tools/multi-agent-sandbox-tools).
6. For large PI catalogs, confirm whether the run uses direct tool exposure or
[Tool Search](/tools/tool-search).
## Related
- [Automation](/automation) for cron, tasks, heartbeat, commitments, hooks, standing orders, and Task Flow
- [Agents](/concepts/agent) for the agent model, sessions, memory, and multi-agent coordination
- [Tools and custom providers](/gateway/config-tools) for the canonical tool policy reference
- [Plugins](/tools/plugin) for plugin installation and management
- [Plugin SDK](/plugins/sdk-overview) for plugin author reference
- [Skills](/tools/skills) for skill load order, gating, and config
- [Tool Search](/tools/tool-search) for compact PI tool catalog discovery

2
docs/tools/lobster.md
View file

@ -360,6 +360,6 @@ One public example: a "second brain" CLI + Lobster pipelines that manage three M
## Related
- [Automation & Tasks](/automation) - scheduling Lobster workflows
- [Automation](/automation) - scheduling Lobster workflows
- [Automation Overview](/automation) - all automation mechanisms
- [Tools Overview](/tools) - all available agent tools