markdown-compression

Compress markdown documents using lossless and lossy techniques to reduce token count.

28|3|Updated Feb 22, 2026

Search Tags:#documentation #markdown #prompt engineering #llm #compression #token optimization

Ready to run

npx skills add https://github.com/oborchers/fractional-cto --skill markdown-compression

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: markdown-compression
Source: https://github.com/oborchers/fractional-cto/tree/main/markdown-compressor/skills/markdown-compression
Command: npx skills add https://github.com/oborchers/fractional-cto --skill markdown-compression

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and examples (resource) components.

What problem does it solve?

This Skill tackles the challenge of excessive token usage in LLM-facing documents, making prompts and documentation more concise and cost-effective without losing critical information.

Core Features & Use Cases

Lossless Compression: Optimizes markdown structure (whitespace, formatting) without altering content.
Lossy Compression: Rewrites content for density, removing redundancy and filler while preserving essential instructions.
Use Case: You have a lengthy CLAUDE.md file for an agent. Use this Skill to significantly reduce its token count, making it faster and cheaper to load into the LLM's context, while ensuring the agent still understands all its instructions.

Quick Start

Use the markdown-compression skill to compress the provided markdown document.

Frequently Asked Questions about markdown-compression

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

FAQPage Schema

How do I reduce the token count of a markdown file for an LLM prompt?▼

You can reduce markdown token count by applying lossless structural formatting and lossy semantic compression to eliminate redundancy. This Skill analyzes document structure and rewrites content for semantic density, significantly lowering token usage and cost while preserving critical LLM instructions.

What is the difference between lossless and lossy markdown compression?▼

Lossless markdown compression optimizes whitespace and formatting without altering content, whereas lossy compression rewrites content for density by removing filler. Both techniques reduce token count for LLM consumption while maintaining essential instructions.

How do I compress a lengthy CLAUDE.md file to make agent loading faster?▼

To compress a lengthy CLAUDE.md file, use the automatic activation workflow to apply structural and semantic transformations. This reduces token count, making the document faster and cheaper to load into the LLM's context while ensuring the agent still understands all instructions.

Does markdown compression remove critical instructions from LLM documentation?▼

Markdown compression preserves critical instructions by applying semantic density transformations that remove redundancy and filler. The process analyzes document structure to ensure that while token count is significantly reduced, the LLM still understands all essential instructions.

Can I optimize prompt engineering documentation without losing context?▼

You can optimize prompt engineering documentation using guided compression workflows that apply lossless and lossy techniques. This reduces token count and cost while explicitly preserving essential instructions and maintaining the necessary context for LLM consumption.

When should I not use lossy compression for markdown documents?▼

You should avoid lossy compression when exact wording is legally or technically required, as it rewrites content for density. Instead, use lossless compression to optimize whitespace and formatting, which reduces token count without altering the original text.