code-review

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

• Third-party content exposure detected (medium risk: 0.65). Yes—this workflow can ingest outsider-authored free text from a PR comment/thread or prior synthesis when fix-loop applies (“Prior Action findings exist in thread or PR” → “paste summary or link” into member prompts), which is then read by the LLM as part of the contextual Full re-review overlay.