caveman-compress

Compress memory Markdown files into caveman-speak to reduce input tokens.

2|Updated Mar 16, 2026
One-click install
npx skills add https://github.com/hamzaPixl/pixl-ai --skill caveman-compress
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: caveman-compress
Source: https://github.com/hamzaPixl/pixl-ai/tree/main/packages/crew/skills/caveman-compress
Command: npx skills add https://github.com/hamzaPixl/pixl-ai --skill caveman-compress

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Compress natural-language memory content to caveman style to reduce token usage while preserving essential structure like code blocks, URLs, and file paths. It enables efficient long-term memory management for AI sessions.

Core Features & Use Cases

  • Caveman-compress memory Markdown files (CLAUDE.md, rules, references, todos) to minimize token load during sessions.
  • Preserves code blocks, URLs, file paths, and headings to allow safe rollback and auditability.
  • Includes a rollback mechanism: the original file is saved as <filename>.original.md before compression.

Quick Start

Compress the target memory Markdown file using caveman-compress to save tokens while preserving code blocks, URLs, and file paths.

Frequently Asked Questions about caveman-compress

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

FAQPage Schema
How do I reduce token usage in markdown memory files for AI sessions?

You can compress markdown memory files into caveman-speak to reduce token usage during AI sessions. This technique preserves essential structure like code blocks, URLs, file paths, and headings while minimizing the overall token load.

Does compressing markdown content preserve code blocks and file paths?

Yes, compressing markdown content into caveman-speak preserves code blocks, URLs, file paths, and headings. It validates the compressed output against the original file to ensure essential structural elements remain intact for safe rollback.

Can I roll back markdown files after compressing them to save tokens?

Yes, you can roll back compressed markdown files because the original content is saved as a .original.md backup before compression occurs. This mechanism ensures safe rollback and auditability across sessions.

What is the best way to compress CLAUDE.md and rules files for token savings?

The best way to compress CLAUDE.md and rules files is by converting natural-language content into caveman-speak. This approach reduces input tokens while maintaining the validity of references and todos across sessions.

Do I need Python CLI to compress markdown memory content into caveman-speak?

Yes, you need the vendored Python CLI and Claude integration to compress markdown memory content. This setup applies the caveman-speak transformation targetting CLAUDE.md, rules, references, and todos across sessions.

When should I not use caveman-speak compression for markdown files?

You should avoid caveman-speak compression when natural-language readability of markdown files is critical for non-technical auditing. Although it preserves code blocks and URLs, the compressed text format alters standard prose into abbreviated caveman style.