ai-sdk-expert

Warn

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses local shell utilities including grep, find, and npm to analyze project structure, detect AI SDK versions, and perform code validation such as type-checking and unit testing.
  • [CREDENTIALS_UNSAFE]: The skill performs recursive searches within .env files for keywords related to AI providers (e.g., openai, anthropic). This results in the exposure of sensitive environment configuration files that typically contain private API keys to the agent's context.
  • [EXTERNAL_DOWNLOADS]: Suggests the use of @ai-sdk/codemod and various official AI SDK provider packages. These resources are sourced from established and well-known technology organizations.
  • [REMOTE_CODE_EXECUTION]: Instructs the agent to execute @ai-sdk/codemod using npx, which involves fetching and running a remote script. This is used for automated project migration.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 17, 2026, 02:08 AM
Security Audit — agent-trust-hub — ai-sdk-expert