review-pr

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

• Third-party content exposure detected (high risk: 0.90). The skill fetches GitHub pull request review threads via the GraphQL query (queries/pr-review-threads.graphql) and instructs the agent to read and quote unresolved review comments (user-generated content) and to automatically apply fixes based on those comments, so untrusted third-party review text can directly influence tool actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill fetches the Pull Request at runtime (e.g., https://github.com/owner/repo/pull/123 via gh pr view / GitHub GraphQL) and uses review comment bodies to drive agent decisions and automatically apply edits, so external PR URLs/content can directly control prompts and actions.