Confidence Check
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- Indirect Prompt Injection (SAFE): The skill retrieves content from external documentation and code repositories via WebFetch and Tavily MCP. This ingestion is a core part of its verification logic and no malicious instructions were found.
- Command Execution (SAFE): The skill utilizes standard developer tools like grep and glob to perform local searches for existing code patterns.
- Unverifiable Dependencies (SAFE): The documentation refers to an external file named confidence.ts which was not provided, but its described functionality is consistent with the skill's purpose.
Audit Metadata