Do I need an existing OpenClaw installation?

Yes. This skill wraps the OpenClaw CLI, which requires a working OpenClaw installation. It can also help with initial setup and configuration.

Is this safe to use on production systems?

It can make real changes to your OpenClaw configuration. Review what your agent is doing before confirming destructive operations, especially on production deployments.

Home/Skills/Development Tools/OpenClaw Anything

OpenClaw Anything

3 versions

doanbactam·Feb 27, 2026

Development Tools

7.1k24

Summary

TL;DR: OpenClaw CLI wrapper - gateway, channels, models, agents, nodes, browser, memory, security, automation.

OpenClaw Anything wraps the OpenClaw CLI so your AI agent can manage the entire OpenClaw ecosystem. Gateway configuration, channel setup, model management, agent deployment, node operations, browser control, memory, security, and automation.

Think of it as a single skill that gives your agent admin access to your OpenClaw installation. Instead of running CLI commands manually, your agent handles the system management for you.

If you are setting up or maintaining an OpenClaw deployment, this skill saves time by letting your agent handle the repetitive configuration and management tasks.

Use cases

Configuring the OpenClaw Gateway and setting up channels
Managing AI models and deploying agents in your OpenClaw instance
Running node operations and monitoring system health
Automating routine OpenClaw administration tasks

Installation

Run this command to install the skill on your OpenClaw agent:

Install with OpenClaw

npx clawhub@latest install openclaw-anything

Downloads

7.1k

Active installs

Stars

Updated

Feb 27, 2026

SKILL.md

---
name: openclaw
description: OpenClaw CLI wrapper — gateway, channels, models, agents, nodes, browser, memory, security, automation.
---

# OpenClaw Skill

CLI wrapper + docs companion. Does NOT contain OpenClaw runtime source.
Wraps `openclaw` CLI and provides local reference docs aligned to `https://docs.openclaw.ai`.

## Prerequisites
- `openclaw` CLI in `PATH` (required)
- Node.js (install/update flows), Playwright deps (browser), Tailscale (remote nodes) — optional

## Quick Reference

| Need | File |
|------|------|
| Find a command | `references/cli-full.md` → search by keyword |
| Security rules | `references/security-policy.md` |
| Config syntax | `references/config-schema.md` |
| Deploy/update | `references/deployment.md` |
| Platform notes | `references/nodes-platforms.md` |
| Doc links | `references/hubs.md` |

## Global Flags
`--dev` `--profile <name>` `--no-color` `--json` `-V`

## Security Model
Default: least privilege. High-risk ops require explicit per-action approval.

### Low-risk (default)
Status, list, health, doctor, logs, config read, docs search, memory search.

### High-risk (require `OPENCLAW_WRAPPER_ALLOW_RISKY=1`)
Shell exec · nodes invoke/run/camera/screen/location · browser automation · cron mutate · plugin/hook install · device pairing · secrets apply · sandbox recreate · webhooks · dns setup.

Wrapper: `bash scripts/openclaw.sh <command> [args]`
Granular gating: plugin gates only install/enable, secrets gates only apply, sandbox gates only recreate.

## Wrapper Command Routes
```
LOW-RISK (pass-through):
  install setup doctor status reset version tui dashboard
  update uninstall health logs configure completion config docs qr
  channel model agent agents message sessions memory skills
  security approvals system directory acp gateway service

HIGH-RISK (OPENCLAW_WRAPPER_ALLOW_RISKY=1):
  cron browser webhooks dns nodes node devices pairing prose
  plugin (install|enable only)
  hooks (install|enable only)
  secrets (apply only)
  sandbox (recreate only)
```

## Non-goals
- Not the OpenClaw runtime source
- Does not provision system packages
- Does not manage networking/VPN
- Does not authorize autonomous privileged execution

---
Last normalization: 2026-02-27 · Source: `https://docs.openclaw.ai`

Version history

v2.0.0Latest

Feb 27, 2026

openclaw-anything 2.0.0 - Expanded and streamlined documentation with improved quick references and global flags list. - Clarified security model, distinguishing low-risk and high-risk commands with specific environmental gates. - Updated and detailed command routing, clearly separating low-risk and high-risk operations. - Simplified prerequisites and non-goals for clearer onboarding and maintenance.

v1.0.2

Feb 17, 2026

- Documentation and policy overhaul: clearer separation of responsibilities, scope, and security boundaries. - Added: references/prerequisites.md and references/security-policy.md for prerequisites and detailed security policy. - Broadened prerequisites listing, including environment variables and typical external dependencies. - New and explicit mapping of wrapper commands to official openclaw CLI actions, with risk-gated areas highlighted. - Emphasized that this package is a CLI wrapper and documentation companion, not the OpenClaw runtime or installer. - Updated and reorganized documentation to align with the latest official OpenClaw docs.

v1.0.1

Feb 3, 2026

Commit

Frequently asked questions

It covers the full CLI: Gateway, Channels, Models, Agents, Nodes, Browser, Memory, Security, and Automation. Your agent can perform any operation available through the OpenClaw command line.

Installation method

Send this prompt to your agent to install the skill

npx clawhub@latest install openclaw-anything

Download ZIP

Skill info

Versionv2.0.0

Authordoanbactam

CategoryDevelopment Tools

UpdatedFeb 27, 2026

Files

SKILL.md2.2 KB

Run OpenClaw in the cloud

Deploy in seconds. Skills pre-installed.

See plans

Skill data sourced from ClawHub