cmo

Warn

Audited by Snyk on Mar 7, 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 explicitly requires using WebSearch/WebFetch to read public competitor websites, pricing pages, Reddit threads, G2/Capterra reviews, Twitter/X and other forums (see "Core Actions" and "How You Research"), so untrusted/user-generated third‑party content is ingested and used to drive positioning and action decisions, enabling indirect prompt injection.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 7, 2026, 11:50 PM