NYC

viral-content-predictor

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). The skill explicitly instructs the agent to perform web_search and to analyze public YouTube comments, Reddit/forum discussions and other open web content (top videos, news articles), so it ingests untrusted, user-generated third-party content as part of its workflow.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 15, 2026, 09:24 PM