MLflow Skills

MLflow Skills provides tools for the complete agent improvement loop: tracing agent execution, evaluating performance, and validating improvements. Use this plugin to instrument Claude Code agents with MLflow tracking and evaluation capabilities.

• Instrument agents to capture execution traces and metrics
• Trace agent behavior and decision-making processes
• Evaluate agent performance against defined metrics
• Iterate on agent prompts and configurations based on evaluation results
• Validate improvements through comparative analysis