Skills
skills/intertwine/dspy-agent-skills/dspy-evaluation-harness/Gen Agent Trust Hub

dspy-evaluation-harness

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation and code examples for the DSPy framework, specifically for building evaluation harnesses.
  • [SAFE]: Usage of external libraries like dspy, pytest, mlflow, and wandb is within the context of model evaluation and experiment tracking.
  • [SAFE]: No suspicious patterns such as prompt injection, data exfiltration, or obfuscation were detected in the instructions or provided scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 02:08 PM
Security Audit — agent-trust-hub — dspy-evaluation-harness