khe-skill-writing

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

• Third-party content exposure detected (medium risk: 0.30). The required testing workflow (SKILL.md step 6 and the cross-model-testing reference) directs the agent to source positive test queries from real user complaints in production logs and user feedback, which end users of the product can submit into without the agent selecting a specific item, so this is medium exposure; no named service URL is specified, so the source is null.