ai-sdk

Pass

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches model lists and documentation from official Vercel domains (ai-gateway.vercel.sh, ai-sdk.dev). These are well-known technical resources and are considered safe for the skill's primary purpose.
  • [COMMAND_EXECUTION]: Uses pnpm for package installation and curl for metadata retrieval. These operations are limited to the scope of the AI SDK and target trusted endpoints for development purposes.
  • [PROMPT_INJECTION]: The skill instructs the agent to favor external documentation over its internal training data to ensure technical accuracy. This is a standard functional directive for a documentation tool and does not attempt to bypass safety guidelines.
  • [DATA_EXFILTRATION]: No sensitive files or credentials are accessed or transmitted. References to API keys in documentation are provided as placeholders for user configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 15, 2026, 03:00 PM