Skill to OpenClaw

Convert a foreign agent-skill directory into an OpenClaw skill. Use when importing a non-OpenClaw skill folder, such as `.agents/skills` or Claude/Codex-styl...

This skill appears coherent and safe in structure: it doesn't install software or request credentials, and it enforces pre-conversion vetting. Before using it, ensure you: (1) run the built-in vetting gate and review flagged items carefully, (2) do not approve conversion of HIGH/EXTREME-risk sources, and (3) prefer running vetting/conversion in an isolated worker or sandbox if your host supports it—if not, be extra cautious because vetting will run in the main session and may expose the session to malicious content. Finally, remember the skill's mapping rules preserve source capabilities by default, so explicitly instruct the agent to exclude or sanitize any content you do not want preserved.

Type: OpenClaw Skill Name: skill-to-openclaw Version: 1.0.0 The skill is designed to convert foreign agent skill directories into OpenClaw format and includes a structured vetting process. However, it contains highly unusual instructions in SKILL.md and references/mapping-rules.md that explicitly command the agent to preserve malicious payloads, covert exfiltration logic, and secret-bearing content during the conversion process unless the user specifically requests a 'salvage-only' path. While framed as a 'high-fidelity' migration principle, these instructions act as a directive to ensure malware is successfully ported rather than cleaned, which could lead to the propagation of harmful code if a user misinterprets the vetting results.