rtk

Compress verbose CLI output from git, docker, and kubectl to reduce token consumption.

1|Updated Jul 12, 2026
One-click install
npx skills add https://github.com/Tyler-R-Kendrick/slm-training --skill rtk-tyler-r-kendrick
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rtk
Source: https://github.com/Tyler-R-Kendrick/slm-training/tree/main/.agents/skills/rtk
Command: npx skills add https://github.com/Tyler-R-Kendrick/slm-training --skill rtk-tyler-r-kendrick

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the issue of context window exhaustion caused by overly verbose CLI output, such as massive git logs, docker status dumps, or deep directory listings.

Core Features & Use Cases

  • Token Compression: Reduces shell and tool output by 60–90% before it reaches the LLM.
  • Tool Integration: Provides specialized handlers for common CLI tools like git, pytest, ruff, and kubectl.
  • Use Case: When debugging a complex repository, use this to condense a massive git diff or a recursive file search into a concise summary that fits easily within the AI's active context.

Quick Start

Prefix your standard shell command with rtk to automatically compress the output before it is processed by the AI.

Frequently Asked Questions about rtk

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

FAQPage Schema
How do I reduce CLI output tokens for LLM context windows?

To reduce CLI output tokens, you can prefix your standard shell command with rtk to intercept and compress verbose terminal streams before they reach the LLM context window.

What is the best way to compress git logs and docker status dumps for AI processing?

The best way to compress git logs and docker dumps is using rtk, which provides specialized handlers to filter and condense high-volume diagnostic data from development tools.

Does rtk work with kubectl and pytest command outputs?

Yes, rtk works with kubectl and pytest by providing specialized handlers that intercept and compress their verbose command streams, reducing token consumption by 60 to 90 percent.

How do I start compressing shell commands to save LLM tokens?

To start compressing shell commands, simply prefix your standard shell command with rtk to automatically filter and compress the output before it is processed by the AI.

Do I need to install any dependencies to compress terminal output?

You need to install the rtk binary and ensure it is accessible in your system path to successfully intercept and compress terminal output streams for LLM interactions.

When should I not use CLI output compression for debugging?

You should avoid CLI output compression when your debugging requires the complete, unfiltered diagnostic data from git or docker, as rtk actively filters out verbose command streams.