ai-sdk
Warn
Audited by Snyk on Jul 15, 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 to fetch runtime content that will directly determine behavior — e.g., running curl against https://ai-gateway.vercel.sh/v1/models to choose model IDs and fetching docs from https://ai-sdk.dev/api/search-docs?q=your_query (and linked .md like https://ai-sdk.dev/docs/agents/building-agents.md) to drive code/instruction decisions, so remote content controls the agent at runtime.
Issues (1)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata