instrument-existing-agent-with-prefactor-sdk
Pass
Audited by Gen Agent Trust Hub on Feb 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructs on installing and using vendor-provided packages (@prefactor/langchain, @prefactor/ai, etc.) and a CLI tool (prefactor), which is standard for SDK integration.
- [SAFE]: Instructions for setting environment variables for authentication (PREFACTOR_API_TOKEN) follow security best practices by using placeholders rather than hardcoded secrets.
- [SAFE]: The documentation explicitly mandates the redaction of secrets and truncation of data when capturing inputs and outputs, demonstrating a security-conscious design.
Audit Metadata