business-architect
Warn
Audited by Snyk on Apr 16, 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 SKILL.md explicitly instructs using web_search and web_fetch in "Фаза 2: Ресёрч болей" to pull content from public, user-generated sources (Reddit subreddits, X/Twitter, competitor reviews, Google Trends), which the agent must read and use to drive its market-pain analysis and subsequent decisions.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata