ai-sdk
Warn
Audited by Snyk on Aug 26, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill instructs the agent at runtime to fetch documentation pages (e.g. https://ai-sdk.dev/api/search-docs?q=your_query and https://ai-sdk.dev/docs/agents/building-agents.md) and use their plain-text .md content for up-to-date instructions, which can directly control the agent's prompts/context.
Issues (1)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata