evaluating-skills

Community

Evaluate and benchmark agent skills for behavioral improvement.

AuthorGit-Fg
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill provides a comprehensive framework for evaluating and benchmarking the effectiveness of agent skills. It helps measure the impact of skill improvements on agent behavior.

Core Features & Use Cases

  • Behavioral Comparison: Compare agent behavior with and without a specific skill to measure its impact.
  • Benchmarking: Aggregate and analyze the results of multiple evaluations to benchmark skill performance.
  • Optimization: Iterate on skill descriptions and behaviors based on feedback and evaluation results.
  • Use Case: After implementing a new skill, use this Skill to compare its behavior against a baseline to determine its effectiveness and iterate on improvements.

Quick Start

To evaluate the 'pdf-extract' skill, use the command: evaluating-skills run --mode RUN mode pdf-extract --runs 3

Dependency Matrix

Required Modules

python

Components

scriptsreferencesassets

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: evaluating-skills
Download link: https://github.com/Git-Fg/taches-principled-light/archive/main.zip#evaluating-skills

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 620,000+ vetted skills library on demand.