deep-research
Warn
Audited by Snyk on Feb 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). This skill explicitly performs live web searches via Claude WebSearch or MCP search tools and, in scripts/research.py, uses ddgs (Bing backend) to fetch titles/bodies/URLs from arbitrary public websites and includes those snippets in the Markdown report, thereby ingesting untrusted third‑party content that could carry indirect prompt injection.
Audit Metadata