code-review

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads GitHub pull request diffs, commits, comments, and issue text via GitHub MCP tools (list_pull_requests, gh pr view), which ingest outsider-authored free text.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill references GitHub raw URLs for optional design and engineering principles documents, which have unknown publisher provenance and represent an unverifiable external runtime dependency.