tanstack-start
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- [Prompt Injection] (SAFE): The instructions are purely functional and do not attempt to override AI safety constraints or system behaviors.- [Data Exposure & Exfiltration] (SAFE): There are no hardcoded secrets, API keys, or commands that access sensitive user data or perform network requests.- [External Downloads] (SAFE): The skill does not reference external URLs for downloading scripts, binaries, or unverifiable packages.- [Indirect Prompt Injection] (SAFE): The skill serves as a guideline for development and does not implement data ingestion surfaces for untrusted content.- [Command Execution] (SAFE): No shell commands, subprocess calls, or script execution patterns are present.
Audit Metadata