component

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly enables fetching and ingesting public third‑party content (e.g., commands/research.md and agents/researcher.md allow WebFetch and mcp__scholar__search and the literature-search skill references PubMed/Semantic Scholar), so the agent is expected to read and interpret untrusted external content that could carry indirect prompt injection.