openclaw-cheatsheet

Quick reference for every OpenClaw CLI command, sub-command, and flag. Use when the user asks about OpenClaw CLI usage (agent, message, config, gateway, cron...

This cheatsheet appears coherent and low-risk: it only reads its own included documentation and provides CLI examples. Before installing or using it: 1) Confirm you do not want the agent to execute commands automatically — the skill may suggest commands but does not itself run them. 2) Be aware the examples reference local scripts (~/ .openclaw/...), node, and obsidian-cli; ensure those tools are present before copying commands. 3) The reference mentions environment variables (e.g., DISCORD_PANEL_WEBHOOK_URL) — do not paste real secrets into chat when asking for help. 4) If you allow the agent to run commands, restrict its shell access and always double-check destructive commands (the cheatsheet already instructs to confirm dangerous actions).

Type: OpenClaw Skill Name: openclaw-cheatsheet Version: 1.0.0 The skill bundle provides a cheatsheet for powerful OpenClaw CLI commands (e.g., `config set`, `gateway install/uninstall`, `cron rm`, `agents delete`) and documents the use of external CLI tools like `obsidian-cli` and local Node.js scripts. While these commands inherently carry risk due to their system interaction capabilities (file system access, service management, configuration changes), the `SKILL.md` and `README.md` files explicitly include strong safety rules, instructing the agent to 'Never run destructive or state-changing commands... without confirming with the user first.' This mitigates direct malicious intent from the skill itself. However, the documentation of such powerful commands and the agent's capability to execute them (even with a safety prompt) makes the skill 'suspicious' rather than 'benign' due to the potential for misuse or exploitation if the agent's safety mechanisms were bypassed by a user's prompt, despite the skill's own defensive instructions.