pr-to-ready

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow explicitly ingests “live PR state” including “PR title, body… review submissions, unresolved threads, top-level comments, and bot comments” from the PR host, which are outsider-authored free text (e.g., other contributors’ PR bodies/comments) that can be fed into the agent’s LLM context via the PR inspection step.