audit-github-actions

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly clones and audits public GitHub repos ("If invoked with a GitHub URL ... git clone ...") and instructs the agent to read every workflow YAML and transitive local files (SKILL.md §1–§3 and §Scope) and to fetch IOCs from external blogs (historical-audit.md), meaning it ingests untrusted, user-generated third‑party content that the agent must interpret and which can influence tool use and next actions.