code-review-excellence

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

• Third-party content exposure detected (high risk: 0.85). This skill’s runtime workflow is for reviewing a user-provided PR (PR description/linked issue and code diffs), which are outsider-authored free text/code; that content is ingested into the agent’s LLM context via the PR review inputs (e.g., PR description, issue body, and changed file text).