AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

CLAUDE.md Progressive Disclosurer

Restructures a bloated CLAUDE.md/AGENTS.md into a lean Level 1 plus on-demand Level 2 references with guaranteed zero information loss.

Dev & CodingAdvanced★ 1,354⑂ 215AI score 9/10Last updated: Aug 26, 2026

What it does

  • Reframes "my CLAUDE.md is too long" from cut lines into relocate information. Line count is explicitly banned as a goal, KPI, or reported result — it may only be a diagnostic symptom.
  • Runs a 6-step workflow: backup → hotspot profiling (/context, per-heading byte table, long-line distribution, downstream truncation limits) → signal vs anti-signal triage → tiering → reference file creation → three-part verification.
  • Enforces verbatim moves (no "tidying while moving") and ships scripts (scripts/profile_claude_md.py, scripts/sink_sections.py) that sink many sections at once, verify each full string landed, and auto-rollback on failure.
  • Provides a carrier-selection matrix: which rules truly belong in always-loaded Level 1 vs a Skill, a blocking/injecting Hook, or .claude/rules/ with paths: frontmatter.
  • Warns about 9 anti-patterns, including fake pointers, believing @import saves context, and hiding loss behind "intentional deletion", plus a detailed catalogue of grep/shell pitfalls that make verification lie to you.

Who it's for

  • Anyone whose global ~/.claude/CLAUDE.md has grown so large that Claude keeps ignoring its rules.
  • Teams cleaning up project CLAUDE.md/AGENTS.md without silently dropping critical conventions.
  • Users sharing one AGENTS.md with Codex who are hitting silent project_doc_max_bytes truncation.

Examples

  1. You ask "is my CLAUDE.md too big?" → it profiles sections by bytes, works the largest contributor first, and refuses to report "reduced 39%" as a win.
  2. Sinking 10 SOP sections out of a 160KB global file into ~/.claude/references/, then proving 10/10 survived via whole-string containment checks (not grep, which false-positives on multi-line text).
  3. Hitting a rule like "always run lint before commit" → it consults the carrier table, recommends a Skill or Hook instead of prose, and lists it for your confirmation instead of deleting it.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/daymade/claude-code-skills/HEAD/daymade-claude-code/claude-md-progressive-disclosurer/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the daymade-claude-code/claude-md-progressive-disclosurer folder from the GitHub repo daymade/claude-code-skills into my ~/.claude/skills/claude-md-progressive-disclosurer/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/daymade/claude-code-skills.git && mkdir -p ~/.claude/skills && cp -r claude-code-skills/daymade-claude-code/claude-md-progressive-disclosurer ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and change into a folder where you keep source code.
  2. Clone the repository: git clone https://github.com/daymade/claude-code-skills.git
  3. Create your skills directory: mkdir -p ~/.claude/skills
  4. Copy just this skill: cp -r claude-code-skills/daymade-claude-code/claude-md-progressive-disclosurer ~/.claude/skills/
  5. Confirm the bundled references/ and scripts/ folders came along: ls ~/.claude/skills/claude-md-progressive-disclosurer
  6. Check python3 is available, since the integrity checks rely on it: python3 --version
  7. Restart Claude Code and ask: "Optimize my CLAUDE.md using progressive disclosure."
  8. Before approving edits, make sure a backup exists (CLAUDE.md.bak.<timestamp>) — it is the reference point for the content-integrity audit.
View source on GitHub ↗License: MIT