Caph-dev

Caph-dev / agents-progressive-disclosure

Public

A skill to refactor bloated AGENTS.md, CLAUDE.md, or similar agent instruction files into a compact routing entrypoint plus focused docs/ reference files.

16
0
100% credibility
Found May 28, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

A skill that helps reorganize bloated AI assistant instruction files into a clean, navigable structure with a compact entrypoint and focused topic documents.

How It Works

1
📄 Your instruction file is getting messy

You notice your AGENTS.md or CLAUDE.md file has grown into a giant wall of text that's hard to maintain.

2
🔍 You search for a solution

You look for a way to organize your AI assistant's instructions and discover this skill that promises to help.

3
⚡ Install it with one click

You run a simple command to add this skill to your AI assistant, and it automatically sets everything up for you.

4
🤖 Ask your assistant to organize everything

You tell your AI assistant to use the skill, and it reads through your messy file and figures out how to organize it.

✨ Your instructions are now clean and organized

Your file transforms into a compact guide with separate topic files, making it easy for your assistant to find exactly what it needs.

Sign up to see the full architecture

3 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 16 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is agents-progressive-disclosure?

This is an agent skill that refactors bloated AGENTS.md or CLAUDE.md files into a compact routing entrypoint with focused documentation files. It solves the common problem where agent instruction files grow unwieldy over time, forcing agents to load low-frequency details on every task. The skill classifies rules into always-on instructions and task-specific guidance, then rewrites the root file as a navigation hub pointing to dedicated docs like response-style.md or project-workflow.md.

Why is it gaining traction?

The hook is clear: context costs money and attention. As teams add rules to agent instruction files, the most critical guidance gets buried alongside rarely-needed details. This skill attacks that bloat directly, offering a structured workflow to extract priority rules while moving detailed task guidance into discoverable docs. Installation is simple via the skills CLI with support for Cursor, Claude Code, Codex, and similar agents. The approach respects existing conventions while providing a concrete methodology for organization.

Who should use this?

Teams maintaining large AGENTS.md or CLAUDE.md files will see immediate value. Tech leads establishing agent guidelines for their codebase benefit most. Developers working with multiple agents across projects can use it to keep instruction files maintainable rather than letting them accumulate rules indefinitely. Anyone frustrated by bloated agent instructions slowing down their workflow should try it.

Verdict

This fills a legitimate gap for teams dealing with agent instruction bloat. At 16 stars and a 1.0% credibility score, the project is early and unproven. Documentation exists in the README, but test coverage and community adoption remain minimal. Worth installing if you struggle with sprawling agent instruction files, but treat it as a starting point rather than a polished solution.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.