skills/fellipeutaka/leon/ai-sdk/Gen Agent Trust Hub

ai-sdk

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches model lists and documentation from official Vercel and AI SDK domains (ai-gateway.vercel.sh, ai-sdk.dev). These resources are well-known services provided by the platform maintainers for environment configuration.
  • [COMMAND_EXECUTION]: Uses system utilities including grep, curl, and jq to search local source code and fetch model metadata. These commands are correctly scoped to documentation lookup and project configuration tasks.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied queries through search operations and URL parameters. The capability surface is limited to documentation retrieval and model identification, with no instructions to execute untrusted data as logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 09:02 AM