OpenClaw Agent Skill
A skill with references teaches an agent to install, configure and repair the OpenClaw gateway, channels and security.
- 313 ★ on the repository
- Last commit
Extended access
Why this level
- The agent changes the gateway configuration with channel tokens and can apply automatic security fixes.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/openclaw-agent-skillDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The agent changes the gateway configuration with channel tokens and can apply automatic security fixes.
This is third-party code. Review the repository files before installing.
What it does
OpenClaw Agent Skill gives an agent knowledge of OpenClaw, a self-hosted gateway that connects AI agents to dozens of messengers at once. The main `SKILL.md` describes core workflows, commands and troubleshooting, while the `references/` folder holds references on architecture, channels, models, memory, security and gateway operations. The agent handles installing and updating OpenClaw, editing `openclaw.json`, connecting channels such as WhatsApp, Telegram, Discord and Slack, gateway management and routing between several agents. The skill covers launching external runtimes through ACP (Codex, Claude Code, Gemini CLI), browser control, cron, hooks and security audits. On failures the agent follows a diagnostic ladder: `openclaw status`, logs, `openclaw doctor` and the channel check `openclaw channels status --probe`.
Who it is for. Administrators and enthusiasts who run an OpenClaw gateway themselves and want an agent to set up channels and find the cause of failures.
Good fit when
- The OpenClaw gateway must be updated and checked afterwards with `openclaw doctor`.
- A Telegram bot is connected to the gateway with a token and a delivery check.
- The gateway does not respond or the port is busy and you need sequential diagnostics.
- A second agent gets its own workspace and channel bindings.
Not a fit when
- OpenClaw is not installed and not planned: the skill is only about it.
- You want third-party plugins with code you have not read.
- Channel tokens and configuration must not reach a chat with the agent.
Example request
Update OpenClaw to the latest version, run `openclaw doctor` and show which channels fail the check.Limitations
The skill is useful only together with OpenClaw, and its facts match the version available when it was written. During setup the agent reads and changes `openclaw.json`, which holds channel tokens and keys, so review its commands before they run. The `openclaw security audit --fix` mode applies changes automatically. No licence is stated in the repository. The README covers only installation for Antigravity; for other agents the files must be wired in by hand. The last commit dates from November 2026.
How to disable. Delete the skill folder from the agent's skills directory, for example `~/.gemini/antigravity/skills/openclaw`. OpenClaw itself and its configuration remain and are removed separately.
FAQ
How do I install the skill?
Run `git clone https://github.com/win4r/OpenClaw-Skill.git` and copy the folder into the skills directory; for Antigravity that is `~/.gemini/antigravity/skills/openclaw`. For other agents, wire in `SKILL.md` and `references/` by hand.
What if the gateway does not respond?
The agent follows a ladder: `openclaw gateway status`, logs, `openclaw doctor`, then `openclaw channels status --probe`.
What about an EADDRINUSE error?
The skill finds the port conflict and either runs `openclaw gateway --force` or changes the port.
Which channels does the skill cover?
The references describe WhatsApp, Telegram, Discord, Slack, iMessage and more than fifteen other channels.
Can the skill harden security?
Yes. `openclaw security audit` checks posture and `--fix` repairs findings automatically. Review the report first.
Discussion
No questions or reviews yet.
Related
SkillsA skill that wires Agent-S into OpenClaw: the agent hands over a task in words and Agent-S clicks through desktop windows itself.
Thirteen skills turn an OpenClaw agent into a worker for paid jobs, while Guard caps its spending and stops loops
A language and runtime for skills an agent writes and a human approves: execution through allowlisted connectors and signatures
Skills for coding agents that build and verify workspaces on the Lemma platform, with their agents and apps
A skill that equips an agent with Kody CLI for local code execution using shared memory, packages and cloud account data.
Yandex Smart Home for OpenClaw
Yandex Smart Home CLI for OpenClaw
Python CLI and OpenClaw skill to switch Yandex Smart Home lights and sockets and set brightness, color temperature and color
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| OpenClaw Agent Skill (this card) | no | 313 | |
| OpenClaw Documentation Skill | no | 32 | gives the agent official OpenClaw docs as a daily snapshot; this skill drives operations: installation, channels and diagnostics |
| OpenClaw Agent Skills | yes | 1,119 | skills for agents that develop the OpenClaw project itself; this one is for those who operate it |
README in short
The skill is built as an OpenClaw reference: one main file with workflows and a folder of separate topics, from the agent loop and command queue to media and polls.
README badge
Original SKILL.md
--- name: openclaw description: Comprehensive guide for installing, configuring, operating, and troubleshooting OpenClaw — a self-hosted, multi-channel AI agent gateway. Use when the user asks about OpenClaw setup, configuration, channel management (WhatsApp/Telegram/Discord/Slack/iMessage/etc.), model provider setup, Gateway operations, multi-agent routing, security hardening, troubleshooting, or any maintenance task related to their local OpenClaw installation. Also use when encountering errors from `openclaw` CLI commands or the Gateway daemon. --- # OpenClaw Maintenance Skill OpenClaw is a self-hosted, open-source (MIT) gateway that routes AI agents across WhatsApp, Telegram, Discord, Slack, iMessage, Signal, and 15+ other channels simultaneously. It runs on macOS, Linux, or Windows. ## Reference Files | Reference | Coverage | |---|---| | [channels.md](references/channels.md) | Per-channel setup (WhatsApp, Telegram, Discord, etc.) | | [channel_troubleshooting.md](references/channel_troubleshooting.md) | Per-channel failure signatures and walkthroughs | | [tools.md](references/tools.md) | Tools inventory (profiles, groups, all built-in tools) | | [exec.md](references/exec.md) | Exec tool: parameters, config, PATH, security, process tool | | [exec_approvals.md](references/exec_approvals.md) | Exec approvals: allowlists, safe bins, approval flow | | [browser.md](references/browser.md) | Browser plugin: profiles, CDP, Chrome MCP, snapshots, SSRF, Control API | | [web_tools.md](references/web_tools.md) | Web tools: Brave, Perplexity, Gemini search providers | | [pdf_tool.md](references/pdf_tool.md) | PDF tool: native/fallback modes, config, page filtering | | [elevated.md](references/elevated.md) | Elevated mode: /elevated directives, sandbox breakout | | [lobster.md](references/lobster.md) | Lobster: typed workflow runtime with approvals | | [llm_task.md](references/llm_task.md) | LLM Task: JSON-only LLM step for structured output | | [openprose.md](references/openprose.md) | OpenProse: multi-agent program runtime | | [plugins.md](references/plugins.md) | Plugins: capability model, official list, config, hooks, SDK, authoring | | [skills.md](references/skills.md) | Skills: locations, config, ClawHub, watcher, token impact | | [providers.md](references/providers.md) | Model provider setup | | [multi_agent.md](references/multi_agent.md) | Multi-agent routing | | [nodes.md](references/nodes.md) | Nodes (iOS/Android/macOS/headless) | | [security.md](references/security.md) | Security hardening | | [secrets.md](references/secrets.md) | Secrets management (SecretRef, vault) | | [sandboxing.md](references/sandboxing.md) | Sandboxing (Docker isolation) | | [config_reference.md](references/config_reference.md) | Full config field reference | | [gateway_ops.md](references/gateway_ops.md) | Gateway operations | | [remote_access.md](references/remote_access.md) | Remote access, SSH, Tailscale, web dashboard | | [sessions.md](references/sessions.md) | Session management, DM isolation, lifecycle, compaction | | [hooks.md](references/hooks.md) | Hooks: internal event hooks, HTTP webhooks, authoring, CLI | | [automation.md](references/automation.md) | Cron jobs, webhooks, Gmail Pub/Sub, background tasks, standing orders | | [acp_agents.md](references/acp_agents.md) | ACP agents: spawn external AI runtimes (Codex, Claude, Gemini, 14+ harnesses) | | [install.md](references/install.md) | Installation, updating, rollback, migration, uninstall | | [web_ui.md](references/web_ui.md) | Web surfaces: Dashboard, Control UI, WebChat | | [slash_commands.md](references/slash_commands.md) | Chat slash commands (/new, /model, /acp, etc.) | | [platforms.md](references/platforms.md) | Platform-specific guides (macOS, iOS, Android, Linux, Windows) | | [diffs_firecrawl.md](references/diffs_firecrawl.md) | Diffs plugin + Firecrawl anti-bot fallback |