agent-md-refactor

Refactor monolithic AGENTS.md files into a minimal root with linked, categorized documentation.

1|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/Thomashighbaugh/opencode --skill agent-md-refactor-thomashighbaugh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-md-refactor
Source: https://github.com/Thomashighbaugh/opencode/tree/main/skills/agent-md-refactor
Command: npx skills add https://github.com/Thomashighbaugh/opencode --skill agent-md-refactor-thomashighbaugh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The root problem is oversized, hard-to-navigate agent instruction files (AGENTS.md, CLAUDE.md, and similar) that burden teams with scattered guidance. Progressive disclosure reduces cognitive load by embedding only the essentials at the root and linking the rest in a organized, categorized structure.

Core Features & Use Cases

  • Progressive disclosure at the root keeps essential guidance visible while longer instructions live in linked, topic-specific files.
  • Clear categorization of related content into dedicated sections, enabling scalable documentation maintenance.
  • Use Case: a team with a monolithic agent guide can transform it into a modular doc set with quick access to core policies and deeper references.

Quick Start

Refactor the monolithic agent instruction files into a minimal root and linked, categorized documentation.

Frequently Asked Questions about agent-md-refactor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I refactor a monolithic AGENTS.md file into modular documentation?

Progressive disclosure reduces cognitive load by embedding only essential guidance at the root and linking the rest in organized, categorized files, which is necessary when monolithic documents become hard to navigate.

What is progressive disclosure for agent instruction files?

Progressive disclosure for agent instruction files keeps essential guidance visible at the root while moving longer instructions into linked, topic-specific files to reduce cognitive load and enable scalable documentation maintenance.

How do I split CLAUDE.md into categorized linked files?

Split CLAUDE.md by applying root minimalism to keep core policies visible, then categorize related content into dedicated sections that link out to deeper reference files for scalable maintenance.

Does this approach work for any monolithic agent guide?

Yes, this approach works for any monolithic agent guide by transforming it into a modular doc set, provided the goal is quick access to core policies alongside deeper references.

When should I restructure agent documentation into a modular doc set?

You should restructure agent documentation into a modular doc set when oversized instruction files scatter guidance and burden teams with navigation difficulties during maintenance.