pulls

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

• Third-party content exposure detected (high risk: 0.75). The required workflow is a GitHub PR command skill that reads PRs/templates and inspects PR diffs/reviews via GitHub tooling (e.g., “critique/review/status” and “create/open” steps), which at runtime can ingest outsider-authored PR body/diff/review text into the LLM context via the PR description and contents.