ai-sdk
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches available model identifiers from Vercel's official AI Gateway endpoint (ai-gateway.vercel.sh).
- [COMMAND_EXECUTION]: Executes shell commands to manage development dependencies (pnpm add ai), check for package updates (npm view ai version), and retrieve model metadata using curl and jq.
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from documentation files and external websites to answer queries.
- Ingestion points: Reads markdown files from node_modules/ai/docs/ and fetches content from https://ai-sdk.dev/docs.
- Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the documentation content.
- Capability inventory: The agent possesses capabilities for file system access, package management, and network operations.
- Sanitization: No explicit sanitization or validation of documentation content is performed before it is processed by the agent.
Audit Metadata