ai-sdk
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill utilizes shell commands like
grepandglobto search within thenode_modulesdirectory for documentation and source code. While intended for searching SDK references, it involves shell interaction based on user queries. - [EXTERNAL_DOWNLOADS] (SAFE): The skill fetches content from
https://ai-sdk.dev. Per the [TRUST-SCOPE-RULE], Vercel is a trusted organization, so these external references are downgraded to safe/low severity. - [PROMPT_INJECTION] (LOW): Category 8 (Indirect Prompt Injection) surface identified. The skill ingests data from external URLs and local files which could contain malicious instructions. Evidence: 1. Ingestion points:
ai-sdk.devand localnode_modules. 2. Boundary markers: Absent. 3. Capability inventory:grep,glob,generateText. 4. Sanitization: Absent.
Audit Metadata