doc-to-skill

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

• Third-party content exposure detected (low risk: 0.10). The agent skill ingests untrusted source documentation files and Git repositories specified by the user via the source input parameter at runtime.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs cloning remote Git URLs during input resolution, which constitutes an external operational dependency with unknown or arbitrary provenance, resulting in a high risk tier.