writing-guidelines

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

• Third-party content exposure detected (high risk: 0.90). The skill uses WebFetch to retrieve guidelines from a public GitHub URL at runtime, and that fetched free-form text is then ingested into the agent’s LLM context to drive the review.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires fetching and using the external guideline file at https://raw.githubusercontent.com/vercel-labs/writing-guidelines/main/command.md at runtime, and that fetched content directly controls the agent's rules/instructions for reviews.