voice-of-customer-synthesizer
Warn
Audited by Snyk on Jul 10, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.85). SKILL.md describes runtime ingestion of outsider-authored free text from external sources (e.g., “G2/Capterra reviews” scraped at runtime and “Social mentions” from Twitter/Reddit), which would be read by the LLM as extracted review/comment text for clustering and quote generation.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata