Skills
.Work. You
Rest

agents-md-evals

Community

Optimize AI instructions with A/B testing.

Software Engineering#prompt engineering#optimization#claude.md#agents.md#a/b testing#ai instructions
Authorvltansky
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill helps you identify and remove redundant or ineffective instructions from your AI configuration files (like AGENTS.md or CLAUDE.md), saving valuable context tokens and improving AI performance.

Core Features & Use Cases

  • A/B Testing: Rigorously tests instruction rules by comparing performance with and without them.
  • Rule Categorization: Identifies rules related to coding style, workflow, decision style, and more.
  • Performance Benchmarking: Grades and benchmarks results to pinpoint rules that don't earn their tokens.
  • Use Case: You suspect your extensive AGENTS.md file is wasting tokens. This Skill will run controlled tests to show you exactly which rules are essential and which can be safely removed.

Quick Start

Use the agents-md-evals skill to evaluate your AGENTS.md file.

Dependency Matrix

Required Modules

None required

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: agents-md-evals
Download link: https://github.com/vltansky/skills/archive/main.zip#agents-md-evals

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 471,000+ vetted skills library on demand.