content-research

Warn

Audited by Snyk on Feb 19, 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.md workflow explicitly requires using WebSearch and scraping public sources (Phase 1.1 "Use WebSearch to find recent news" and Phase 2.1 which instructs searching Twitter/X, LinkedIn, podcasts, Reddit/Quora) and directs the agent to read and incorporate that user-generated, open-web content into research and content decisions, so untrusted third-party material can materially influence agent behavior.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 19, 2026, 03:51 PM