repomix

Package entire code repositories into AI-friendly single-file formats.

Updated Feb 12, 2026
One-click install
npx skills add https://github.com/duc01226/SoftwareDevCareers --skill repomix-duc01226
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: repomix
Source: https://github.com/duc01226/SoftwareDevCareers/tree/main/.claude/skills/repomix
Command: npx skills add https://github.com/duc01226/SoftwareDevCareers --skill repomix-duc01226

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Repomix consolidates large codebases into AI-friendly formats, enabling fast, scalable AI analysis without parsing many files or dealing with noisy context.

Core Features & Use Cases

  • AI-optimized repository packaging with metadata and separators that improve AI digestion
  • Output formats include XML, Markdown, JSON, and plain text
  • Gitignore-aware processing with token counting to optimize context window
  • Remote repository support to package code without cloning locally
  • Security checks for sensitive data and potential exposure during packaging
  • Use cases: preparing AI-ready code contexts for reviews, security audits, and documentation

Quick Start

Install Repomix and begin packaging a repository to generate an AI-friendly context snapshot.

Frequently Asked Questions about repomix

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

FAQPage Schema
How do I package an entire code repository into a single file for LLM context?

To package a code repository for LLM context, you can consolidate the codebase into AI-optimized single-file formats like XML, Markdown, or JSON, applying include/exclude patterns and token counting to optimize the context window.

What is the best way to prepare a codebase for AI security audits?

Preparing a codebase for AI security audits involves packaging the repository into a single file while running security checks to detect sensitive data and potential exposure during the consolidation process.

Can I package a remote GitHub repository without cloning it locally first?

Yes, remote repository packaging is supported, allowing you to package code directly from a remote source without needing to clone the entire repository to your local machine first.

Does repository packaging respect Gitignore rules and custom file filters?

Yes, repository packaging is Gitignore-aware and supports custom include/exclude patterns, ensuring only relevant files are consolidated while respecting your existing version control ignore configurations.

What output formats are available when consolidating a codebase for AI analysis?

When consolidating a codebase for AI analysis, available output formats include XML, Markdown, JSON, and plain text, all structured with metadata and separators to improve AI digestion.

How do I check token counts before feeding a packaged repository snapshot to an LLM?

You can check token counts during the repository packaging process, as the tool includes built-in token counting to help optimize and manage the context window before feeding the snapshot to an LLM.