catalog-babysitter-users

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). Yes—this skill runs gh search code and gh api repos/<owner>/<name> against public GitHub repositories, then ingests the returned repository metadata and matching file paths (outsider-authored free text) into the generated markdown catalog that is fed back into the agent’s LLM context during the run.