compress

Report headroom status and compression savings for Claude Code sessions.

Updated Mar 13, 2026
One-click install
npx skills add https://github.com/jtucker9/mystuff --skill compress-jtucker9
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: compress
Source: https://github.com/jtucker9/mystuff/tree/main/skills/compress
Command: npx skills add https://github.com/jtucker9/mystuff --skill compress-jtucker9

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Headroom is a transparent proxy between Claude Code and the Anthropic API that compresses tool outputs by removing redundant boilerplate. It extends the effective context window by 30–40%.

Core Features & Use Cases

  • Status checks and health diagnostics to verify proxy and compression performance.
  • Session reporting that quantifies tokens saved and compression ratios for cost and efficiency.
  • Configuration guidance to tune headroom settings and ensure stable CC interactions.

Quick Start

Launch the Headroom proxy and run a status check to verify compression gains.

Frequently Asked Questions about compress

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

FAQPage Schema
How do I reduce token usage in Claude Code sessions?

You can reduce token usage in Claude Code by using a proxy to compress tool outputs, which removes redundant boilerplate and extends the effective context window by 30-40%.

What is context window compression and how does it work?

Context window compression works by intercepting tool outputs through a transparent proxy and stripping redundant boilerplate, effectively expanding the available context headroom for ongoing interactions.

How do I check proxy health and compression performance?

You can check proxy health and compression performance by running a status check diagnostic, which verifies the proxy connection and quantifies current token savings and compression ratios.

How do I tune proxy settings to maximize context savings?

You can tune proxy settings for context savings by using configuration reference guidance to adjust compression parameters, ensuring stable Claude Code interactions while achieving 30-40% token reduction.

Does context compression work with ongoing Claude Code sessions?

Yes, context compression is applied during ongoing Claude Code sessions, intercepting API traffic through a transparent proxy to optimize token usage and maintain context headroom in real-time.

Why does my Claude Code session run out of context headroom?

Your Claude Code session runs out of context headroom because uncompressed tool outputs contain redundant boilerplate; applying proxy compression removes this excess data to free up context space.