The Agent Skills Directory

[SAFE]: No malicious patterns were identified. The skill acts as a meta-tool for capability discovery and adheres to safe operational boundaries.
[COMMAND_EXECUTION]: The skill is designed to output installation command templates (e.g., npx skills add ...) for the user. It explicitly instructs the agent not to execute these commands autonomously, maintaining human-in-the-loop verification for environment changes.
[EXTERNAL_DOWNLOADS]: The skill references external public directories (e.g., SkillsMP) to find matching capabilities. This is a discovery-only feature and does not involve the automatic download or execution of remote scripts or unverified binary content.
[DATA_EXPOSURE]: The skill reads local mapping files like skills/INDEX.md and manifest.json. This access is limited to the skill's own indexing structure and does not involve scanning for sensitive user files, credentials, or private configuration data.

discover-skills