security-scanner

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

• Third-party content exposure detected (low risk: 0.10). The security-scanner skill ingests GitHub repository content and files from arbitrary public GitHub URLs provided by the user via --url or URL arguments, processing outsider-authored free text during code and natural language analysis.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill uses WebFetch to retrieve remote content from GitHub repository contents and raw file endpoints, which constitute operational runtime dependencies on third-party URLs.