Skinny Jeans FAQs

Question 1

Is Skinny Jeans compatible with other AI coding tools besides Claude Code?

Accepted Answer

While Skinny Jeans is designed for seamless integration with Claude Code as an MCP server, it can also be used as a standalone CLI tool to optimize files or estimate token savings for any LLM-related task. Its core optimizations are generally beneficial for any LLM context.

Question 2

Can I estimate token savings before processing my entire project?

Accepted Answer

Yes, Skinny Jeans provides tools like `toon_estimate_tokens` to quickly check the token count and potential savings for individual files, and `toon_batch_estimate` to generate comprehensive token savings reports across multiple files or entire directories within your project.

Question 3

How does Skinny Jeans achieve token savings for different file types?

Accepted Answer

It employs TOON encoding for JSON files, achieving 40-60% token savings with improved LLM comprehension. Markdown files are minified (15-30% savings), and code files have comments stripped (10-25% savings). Other text files benefit from general whitespace cleanup.

Question 4

What is TOON encoding, and why is it better for LLMs than standard JSON?

Accepted Answer

TOON (Token-Oriented Object Notation) is a data format specifically designed for LLM consumption. It represents structured data in a highly compact, tabular form that uses 40-60% fewer tokens than JSON, while also improving LLM comprehension accuracy compared to standard JSON.

Question 5

What is Skinny Jeans and how does it help with AI coding?

Accepted Answer

Skinny Jeans is an MCP server that transparently optimizes file content (JSON, Markdown, code) for large language models used in AI coding assistants like Claude Code. It significantly reduces token consumption, leading to lower API costs and more efficient interactions with your AI assistant.

Skinny Jeans

Skinny Jeans

About

Key Features

Use Cases

About

Key Features

Use Cases