review-analyst-agent
Warn
Audited by Snyk on Feb 16, 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 Review Scraper and "Step 2: Collect Reviews" explicitly instruct the agent to scrape and ingest user-generated reviews from public sites (e.g., Amazon, App Store/Google Play, G2/Capterra, Reddit, Twitter/X, YouTube comments), so the agent will read untrusted third‑party content that could contain indirect prompt injections.
Audit Metadata