update-with-comments

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

• Third-party content exposure detected (high risk: 0.95). The workflow fetches PR conversation and inline comments at runtime via gh pr view <pr-id> --comments and gh api repos/:owner/:repo/pulls/<pr-id>/comments, which are outsider-authored free-form text from other contributors, and then “analyzes all comments” to feed into the agent’s LLM context.