babysitting-prs

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required workflow reads outsider-authored free text from GitHub pull request review threads and comments at runtime via scripts/pr_babysit.sh status -> review_threads_graphql_json (GraphQL reviewThreads.comments.body) and state_json (issue/pr comments body, review body) that are then included in the JSON fed to the LLM for deciding how to respond/fix.