ai-sdk-ui

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Indirect Prompt Injection] (SAFE): The provided implementation patterns ingest user messages via an API route. This is a standard surface for chatbot applications and follows official Vercel AI SDK examples without introducing additional vulnerabilities.
  • [Metadata Poisoning] (SAFE): The README lists templates and scripts that are not present in the current file. This is consistent with a documentation-only view of a larger skill repository.
  • [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials or unauthorized network requests were detected. Links to external sites are provided for documentation and validation purposes only.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:10 PM