Skills

OpenClaw uses AgentSkills-compatible skill folders to teach the agent how to use tools. Each skill is a directory containing a

with YAML frontmatter and instructions. OpenClaw loads bundled skills plus optional local overrides, and filters them at load time based on environment, config, and binary presence.

Locations and precedence

OpenClaw loads skills from these sources, highest precedence first:

If a skill name conflicts, the highest source wins.

Codex CLI's native

directory is not one of these OpenClaw skill roots. In Codex harness mode, local app-server launches use isolated per-agent Codex homes, so personal Codex CLI skills are not loaded implicitly. Use to inventory them and to choose skill directories with an interactive checkbox prompt before copying them into the current OpenClaw agent workspace. For non-interactive runs, repeat for the exact skills to copy.

Per-agent vs shared skills

In multi-agent setups each agent has its own workspace:

Same name in multiple places → highest source wins. Workspace beats project-agent, beats personal-agent, beats managed/local, beats bundled, beats extra dirs.

Agent skill allowlists

Skill location and skill visibility are separate controls. Location/precedence decides which copy of a same-named skill wins; agent allowlists decide which skills an agent can actually use.

json5
Copy
{
  agents: {
    defaults: {
      skills: ["github", "weather"],
    },
    list: [
      { id: "writer" }, // inherits github, weather
      { id: "docs", skills: ["docs-search"] }, // replaces defaults
      { id: "locked-down", skills: [] }, // no skills
    ],
  },
}

Allowlist rules

Plugins and skills

Plugins can ship their own skills by listing

directories in (paths relative to the plugin root). Plugin skills load when the plugin is enabled. This is the right place for tool-specific operating guides that are too long for the tool description but should be available whenever the plugin is installed — for example, the browser plugin ships a skill for multi-step browser control.

Plugin skill directories are merged into the same low-precedence path as

, so a same-named bundled, managed, agent, or workspace skill overrides them. You can gate them via on the plugin's config entry.

See Plugins for discovery/config and Tools for the tool surface those skills teach.

Skill Workshop

The optional, experimental Skill Workshop plugin can create or update workspace skills from reusable procedures observed during agent work. It is disabled by default and must be explicitly enabled via

.

Skill Workshop writes only to

, scans generated content, supports pending approval or automatic safe writes, quarantines unsafe proposals, and refreshes the skill snapshot after successful writes so new skills become available without a Gateway restart.

Use it for corrections such as "next time, verify GIF attribution" or hard-won workflows such as media QA checklists. Start with pending approval; use automatic writes only in trusted workspaces after reviewing its proposals. Full guide: Skill Workshop plugin.

ClawHub (install and sync)

ClawHub is the public skills registry for OpenClaw. Use native

commands for discover/install/update, or the separate CLI for publish/sync workflows. Full guide: ClawHub.

Native

installs into the active workspace directory. The separate CLI also installs into under your current working directory (or falls back to the configured OpenClaw workspace). OpenClaw picks that up as on the next session. Configured skill roots also support one grouping level, such as , so related third-party skills can be kept under a shared folder without broad recursive scanning.

ClawHub skill pages expose the latest security scan state before install, with scanner detail pages for VirusTotal, ClawScan, and static analysis.

remains only the install path; publishers recover false positives through the ClawHub dashboard or .

Security

• Workspace and extra-dir skill discovery only accepts skill roots and
  text

  Copy

  SKILL.md
  files whose resolved realpath stays inside the configured root.
• Gateway-backed skill dependency installs (
  text

  Copy

  skills.install
  , onboarding, and the Skills settings UI) run the built-in dangerous-code scanner before executing installer metadata.
  text

  Copy

  critical
  findings block by default unless the caller explicitly sets the dangerous override; suspicious findings still warn only.
• text

  Copy

  openclaw skills install <slug>
  is different — it downloads a ClawHub skill folder into the workspace and does not use the installer-metadata path above.
• text

  Copy

  skills.entries.*.env
  and
  text

  Copy

  skills.entries.*.apiKey
  inject secrets into the host process for that agent turn (not the sandbox). Keep secrets out of prompts and logs.

For a broader threat model and checklists, see Security.

SKILL.md format

must include at least:

markdown
Copy
---
name: image-lab
description: Generate or edit images via a provider-backed image workflow
---

OpenClaw follows the AgentSkills spec for layout/intent. The parser used by the embedded agent supports single-line frontmatter keys only;

should be a single-line JSON object. Use in instructions to reference the skill folder path.

Optional frontmatter keys

URL surfaced as "Website" in the macOS Skills UI. Also supported via `metadata.openclaw.homepage`. When `true`, the skill is exposed as a user slash command. When `true`, the skill is excluded from the model prompt (still available via user invocation). When set to `tool`, the slash command bypasses the model and dispatches directly to a tool. Tool name to invoke when `command-dispatch: tool` is set. For tool dispatch, forwards the raw args string to the tool (no core parsing). The tool is invoked with `{ command: "", commandName: "", skillName: "" }`.

Gating (load-time filters)

OpenClaw filters skills at load time using

(single-line JSON):

markdown
Copy
---
name: image-lab
description: Generate or edit images via a provider-backed image workflow
metadata:
  {
    "openclaw":
      {
        "requires": { "bins": ["uv"], "env": ["GEMINI_API_KEY"], "config": ["browser.enabled"] },
        "primaryEnv": "GEMINI_API_KEY",
      },
  }
---

Fields under

:

When `true`, always include the skill (skip other gates). Optional emoji used by the macOS Skills UI. Optional URL shown as "Website" in the macOS Skills UI. Optional list of platforms. If set, the skill is only eligible on those OSes. Each must exist on `PATH`. At least one must exist on `PATH`. Env var must exist or be provided in config. List of `openclaw.json` paths that must be truthy. Env var name associated with `skills.entries..apiKey`. Optional installer specs used by the macOS Skills UI (brew/node/go/uv/download).

If no

is present, the skill is always eligible (unless disabled in config or blocked by for bundled skills).

Sandboxing notes

• text

  Copy

  requires.bins
  is checked on the host at skill load time.
• If an agent is sandboxed, the binary must also exist inside the container. Install it via
  text

  Copy

  agents.defaults.sandbox.docker.setupCommand
  (or a custom image).
  text

  Copy

  setupCommand
  runs once after the container is created. Package installs also require network egress, a writable root FS, and a root user in the sandbox.
• Example: the
  text

  Copy

  summarize
  skill (
  text

  Copy

  skills/summarize/SKILL.md
  ) needs the
  text

  Copy

  summarize
  CLI in the sandbox container to run there.

Installer specs

markdown
Copy
---
name: gemini
description: Use Gemini CLI for coding assistance and Google search lookups.
metadata:
  {
    "openclaw":
      {
        "emoji": "♊️",
        "requires": { "bins": ["gemini"] },
        "install":
          [
            {
              "id": "brew",
              "kind": "brew",
              "formula": "gemini-cli",
              "bins": ["gemini"],
              "label": "Install Gemini CLI (brew)",
            },
          ],
      },
  }
---

Installer selection rules

Per-installer details

Config overrides

Bundled and managed skills can be toggled and supplied with env values under

in :

json5
Copy
{
  skills: {
    entries: {
      "image-lab": {
        enabled: true,
        apiKey: { source: "env", provider: "default", id: "GEMINI_API_KEY" }, // or plaintext string
        env: {
          GEMINI_API_KEY: "GEMINI_KEY_HERE",
        },
        config: {
          endpoint: "https://example.invalid",
          model: "nano-pro",
        },
      },
      peekaboo: { enabled: true },
      sag: { enabled: false },
    },
  },
}

`false` disables the skill even if it is bundled or installed. The bundled `coding-agent` skill is opt-in: set `skills.entries.coding-agent.enabled: true` before exposing it to agents, then make sure one of `claude`, `codex`, `opencode`, or `pi` is installed and authenticated for its own CLI. Convenience for skills that declare `metadata.openclaw.primaryEnv`. Supports plaintext or SecretRef. Injected only if the variable is not already set in the process. Optional bag for custom per-skill fields. Custom keys must live here. Optional allowlist for **bundled** skills only. If set, only bundled skills in the list are eligible (managed/workspace skills unaffected).

If the skill name contains hyphens, quote the key (JSON5 allows quoted keys). Config keys match the skill name by default — if a skill defines

, use that key under .

Environment injection

When an agent run starts, OpenClaw:

1. Reads skill metadata.
2. Applies
   text

   Copy

   skills.entries.<key>.env
   and
   text

   Copy

   skills.entries.<key>.apiKey
   to
   text

   Copy

   process.env
   .
3. Builds the system prompt with eligible skills.
4. Restores the original environment after the run ends.

Environment injection is scoped to the agent run, not a global shell environment.

For the bundled

backend, OpenClaw also materializes the same eligible snapshot as a temporary Claude Code plugin and passes it with . Claude Code can then use its native skill resolver while OpenClaw still owns precedence, per-agent allowlists, gating, and env/API key injection. Other CLI backends use the prompt catalog only.

Snapshots and refresh

OpenClaw snapshots the eligible skills when a session starts and reuses that list for subsequent turns in the same session. Changes to skills or config take effect on the next new session.

Skills can refresh mid-session in two cases:

• The skills watcher is enabled.
• A new eligible remote node appears.

Think of this as a hot reload: the refreshed list is picked up on the next agent turn. If the effective agent skill allowlist changes for that session, OpenClaw refreshes the snapshot so visible skills stay aligned with the current agent.

Skills watcher

By default, OpenClaw watches skill folders and bumps the skills snapshot when

files change. Configure under :

json5
Copy
{
  skills: {
    load: {
      watch: true,
      watchDebounceMs: 250,
    },
  },
}

Remote macOS nodes (Linux gateway)

If the Gateway runs on Linux but a macOS node is connected with

allowed (Exec approvals security not set to ), OpenClaw can treat macOS-only skills as eligible when the required binaries are present on that node. The agent should execute those skills via the tool with .

This relies on the node reporting its command support and on a bin probe via

or . Offline nodes do not make remote-only skills visible. If a connected node stops answering bin probes, OpenClaw clears its cached bin matches so agents no longer see skills that cannot currently run there.

Token impact

When skills are eligible, OpenClaw injects a compact XML list of available skills into the system prompt (via

in ). The cost is deterministic:

• Base overhead (only when ≥1 skill): 195 characters.
• Per skill: 97 characters + the length of the XML-escaped
  text

  Copy

  <name>
  ,
  text

  Copy

  <description>
  , and
  text

  Copy

  <location>
  values.

Formula (characters):

text
Copy
total = 195 + Σ (97 + len(name_escaped) + len(description_escaped) + len(location_escaped))

XML escaping expands

into entities (, , etc.), increasing length. Token counts vary by model tokenizer. A rough OpenAI-style estimate is ~4 chars/token, so 97 chars ≈ 24 tokens per skill plus your actual field lengths.

Managed skills lifecycle

OpenClaw ships a baseline set of skills as bundled skills with the install (npm package or OpenClaw.app).

exists for local overrides — for example, pinning or patching a skill without changing the bundled copy. Workspace skills are user-owned and override both on name conflicts.

Looking for more skills?

Browse https://clawhub.ai. Full configuration schema: Skills config.

Related

• ClawHub — public skills registry
• Creating skills — building custom skills
• Plugins — plugin system overview
• Skill Workshop plugin — generate skills from agent work
• Skills config — skill configuration reference
• Slash commands — all available slash commands