investment-analysis
Warn
Audited by Snyk on Apr 4, 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 (SKILL.md Step 4 and references/agent-prompts.md) explicitly instructs the sector agents to use WebSearch/WebFetch on open web sources — including CoinGecko/CoinMarketCap, news sites, Twitter/X and Reddit — and to read/interpret that untrusted, user-generated content to generate recommendations, so third‑party pages can materially influence agent decisions.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata