resolve-pr-feedback

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

• Third-party content exposure detected (high risk: 0.95). Outsider-authored PR review content is fetched at runtime via gh api graphql in scripts/get-pr-comments (review thread comment body, top-level PR comment body, and review body body), then passed into the agent’s LLM context as untrusted “comment text”.