ce-resolve-pr-feedback

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches user-generated PR comments and review threads from GitHub (see scripts/get-pr-comments and SKILL.md step 1/step 5 where agents are given the "full comment text" for each thread), and those untrusted third-party texts are read and used to decide fixes and dispatch tool-driven actions, creating a clear avenue for indirect prompt injection.