youtube-brief
Warn
Audited by Snyk on May 12, 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 required workflow (Steps 4 and 5 in SKILL.md) explicitly instructs the agent to use a research agent and WebSearch to ingest content from open/public, user-generated sources (e.g., Reddit, forums, Quora, YouTube comments) and to use that material to shape points, proofs, and decisions, which clearly exposes the agent to untrusted third-party content that can influence its actions.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata