Battlecard
Warn
Audited by Snyk on Mar 3, 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's required workflow (Layer 2: "Competitor Research" in SKILL.md and the "Web Search" required capability) explicitly instructs the agent to run web searches (Google grounding) and ingest content from public sources including G2/Capterra/TrustRadius, news sites, forums/Reddit and other open web pages (see references/competitive-intel-guide.md), so untrusted user-generated third-party content is read, interpreted, and used to drive battlecard conclusions and next actions.
Audit Metadata