rlxp-define-metrics

What problem does it solve?

This Skill helps define clear, task-scoped reinforcement-learning metrics when training reward is not enough to judge success or compare experiments fairly.

Core Features & Use Cases

• Primary Metric Design: Establish the main outcome metric for a specific task and study.
• Guardrails and Diagnostics: Add component metrics, worst-bin checks, lower-quantile checks, and failure-mode diagnostics.
• Scorecard Readiness: Produce metric specs and notes that support candidate validation, monitoring, result audits, and accept or reject decisions.
• Use Case: Use this Skill when a baseline exists but the team needs a defensible scorecard for evaluating RL improvements without changing the protocol mid-study.

Quick Start

Use this skill to define the primary metric, guardrails, and scorecard for the current RL task and study.