skills.rest
austinwilcox
austinwilcoxCOMMUNITY
Explore all skills by this author →
Website𝕏@Seconds21iveGitHub

compress

Compress markdown memory files into denser phrasing while preserving code, URLs, and headings.

1|1|Updated Jul 31, 2023
Search Tags:#markdown#validation#Claude#natural language#memory compression#token reduction#backup safety
One-click install
npx skills add https://github.com/austinwilcox/dotfiles --skill compress-austinwilcox
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: compress
Source: https://github.com/austinwilcox/dotfiles/tree/main/.claude/skills/compress
Command: npx skills add https://github.com/austinwilcox/dotfiles --skill compress-austinwilcox

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires anthropic, subprocess, and includes scripts (resource) components.

What problem does it solve?

This Skill reduces input-token usage by transforming natural-language memory files into a shorter “caveman” format while preserving all technical substance.

Core Features & Use Cases

  • Token-saving compression: Converts markdown/text prose (e.g., CLAUDE.md, todos, preferences) into denser phrasing.
  • Hard preservation guarantees: Keeps code blocks, inline code, URLs, headings, and structural elements intact.
  • Safety via validation + targeted fixes: Validates the compressed output and retries by applying only the listed issues (no full recompression).

Quick Start

Ask for compression with /caveman:compress <filepath> to overwrite the file and save a human-readable backup as <filename>.original.md.

Frequently Asked Questions about compress

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

FAQPage Schema
How do I reduce input token usage for my Claude markdown memory files?▼

To reduce input token usage, you can compress markdown memory files into a denser format that preserves technical content, structure, and references while cutting down natural-language prose.

How does markdown memory compression handle code blocks and URLs?▼

Markdown memory compression guarantees hard preservation of code blocks, inline code, URLs, and headings by never altering them during the token reduction process.

How do I compress a CLAUDE.md file without losing technical content?▼

You can compress a CLAUDE.md file by applying a caveman compression command to the filepath, which transforms natural language into denser phrasing while keeping all technical substance intact.

Is my original markdown file safe when using a token reduction tool?▼

Your original file remains safe during token reduction because the tool saves a human-readable backup as <filename>.original.md and only overwrites the original after passing validation guardrails.

What happens if markdown compression validation fails?▼

If markdown compression validation fails, the tool retries by applying targeted fixes to the listed issues rather than performing a full recompression of the entire file.

Can I use natural language compression for long-lived chat assistant preferences?▼

Yes, natural language compression is designed for workflows that maintain long-lived memory such as todos, preferences, and CLAUDE.md files for chat-based assistants.

Related Skills

Explore More Personal & Entrepreneur Skills→
guhans6guhans6

compress

Compress markdown files into caveman format using the Claude API.

Community
Advanced
SamuelSahaSamuelSaha

Token Optimization

Reduce AI token consumption via shallow skill loading, context pruning, and output compression.

Community
Intermediate
markshustmarkshust
55

project-update

Synchronize .claude configuration files and migrate legacy pipeline definitions to frontmatter-based agent enrollment.

Community
Advanced
ebrahim-saniebrahim-sani

web-reader

Convert web pages into clean Markdown text with JSON output.

Community
Basic