maintain-project-context

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

• Third-party content exposure detected (low risk: 0.10). The workflow reads issue-tracker tickets and PR bodies referenced by specific IDs found in local repository files, representing low-risk targeted fetching of external content.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs installation of an external package from an arbitrary GitHub repository (toy-crane/skills) via npx skills@latest add at runtime, creating an unverified operational dependency with weak provenance.