implementing-code

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

• Third-party content exposure detected (high risk: 0.70). SKILL.md Step 2/5 requires reading “plan.md” and other dependency artifacts from the provided artifact path (task breakdown, feature spec, constitution, etc.), which are outsider-authored free-text documents if the operating user did not author them; that text is then ingested into the agent’s LLM context for matching/extraction and requirement re-reading.