marketing-refresh
Warn
Audited by Snyk on Mar 23, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 1.00). The skill's Phase 1 and Phase 2 explicitly require web searches and ingestion of public third‑party content (e.g., "Use web search... Check G2, Capterra, Product Hunt, AlternativeTo" in Phase 1 and "Reddit, Hacker News, and forum discussions" in Phase 2), which are untrusted/user‑generated sources the agent must read and that directly influence its analysis and next actions.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata