deep-research

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly instructs the agent to "Fetch top sources" by running WebFetch on external URLs (see "Step 2: Fetch Top Sources — Run WebFetch with the URL" and "Fetch top 5-8 Tier 1-2 sources"), meaning it ingests untrusted public web content and uses those findings to drive decisions and actions.