domain-profiles
Warn
Audited by Snyk on Feb 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: 0.90). The skill's domain profiles (e.g., technology.md and language.md Search Strategy sections) explicitly direct researcher/research-collector agents to fetch and use open/public, user-generated sources such as Stack Overflow, Reddit, GitHub, Wikipedia, and YouTube as part of their research workflow, so the agent will read/interpret untrusted third-party content.
Audit Metadata