The Agent Skills Directory

[SAFE]: The skill performs metadata auditing and schema validation using local scripts (scripts/skill-lint.js, scripts/generate-manifest.js). These operations are consistent with the skill's stated purpose and do not involve remote code execution or unauthorized data access.
[COMMAND_EXECUTION]: The skill uses the Bash tool to run local validation scripts. This execution is limited to specific repository scripts designed for linting and does not include arbitrary command execution or privilege escalation.
[DATA_EXPOSURE]: File access is restricted to the codebase scope (e.g., skills/**/SKILL.md, schemas/*.json). There are no patterns indicating the collection or exfiltration of sensitive information such as API keys or system credentials.

graph-audit