market-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 explicitly directs the agent to consult and extract information from public third-party sources (e.g., "visit their pricing page", "Reddit threads", "Twitter complaints", LinkedIn/SIMILARWEB/G2/Capterra) and to use those user-generated/public-site findings to drive market-sizing and GO/NO-GO decisions, so untrusted external content can influence next actions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 19, 2026, 08:27 PM