receive-review

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

• Third-party content exposure detected (high risk: 0.70). The skill requires “READ every item” and later “RESPOND/VERIFY/EVALUATE” based on review feedback items, but the source of those items (e.g., external reviewers’ text) is not constrained, so at runtime the agent may ingest outsider-authored free-form review feedback into the LLM context.