research

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

• Third-party content exposure detected (high risk: 1.00). The skill's Step 3 explicitly uses Claude's WebSearch/WebFetch and Gemini's "web grounding" to fetch and interpret content from public websites/URLs, and those retrieved third‑party sources are read and synthesized into actionable recommendations that can influence subsequent tool use and decisions.