skill-review

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

• Third-party content exposure detected (high risk: 0.80). This skill explicitly fetches and verifies content from open/public third-party sources—e.g., "Official Docs: WebFetch/Context7", GitHub repositories, npm registry, and external documentation links (broken-links checks)—and the agent is expected to read and interpret that external content as part of its audit workflow, exposing it to untrusted, user-generated material.