review-pr

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

• Third-party content exposure detected (high risk: 0.85). The skill ingests outsider-authored free text from the PR itself—e.g., gh pr view <N> --json ... ,body,... and git log ... --format="%h %s%n%n%b" read the PR description and commit messages written by the contributor (non-operating user) into the agent’s LLM context.