academic-deep-research

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

• Third-party content exposure detected (high risk: 1.00). The skill's required Phase 3 workflow and Tool Configuration mandate using web_search and web_fetch to retrieve and deeply analyze open/public third‑party sources (see "Tool Configuration" and Phase 3: "web_search" count=20 and "web_fetch on primary sources" in SKILL.md; example.md even fetches TechCrunch and healthcare IT forum discussions), so the agent will consume untrusted user-generated/public web content that can materially influence its actions and recommendations.