product-antifraud

Warn

Audited by Snyk on Feb 25, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.80). The SKILL's "Trend Awareness Protocol" explicitly tells the agent to perform web searches (e.g., queries like "fintech fraud detection patterns 2026") before answering, which requires fetching and interpreting open/public third-party web content that could materially influence decisions and tool use.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 25, 2026, 02:44 AM