draft-spec

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

• Third-party content exposure detected (high risk: 0.80). The skill's Step 2 explicitly tells the agent to "look up library or framework docs" and to use "WebSearch as a fallback," which requires fetching and interpreting open/public third‑party web content (including user‑generated docs/forums) and using those findings to influence architecture and decisions.