skill-code-review

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

• Third-party content exposure detected (high risk: 0.90). The skill's Pre-Review step explicitly requires comparing the diff against "TODOS.md, PR body, commit messages" (i.e., user-authored PR bodies/commit messages) and the Post-Review flow uses the gh CLI to detect open PRs, so it ingests untrusted, user-generated content from public GitHub PRs which can influence review scope and subsequent actions.