repomix

Package code repositories into single AI-friendly files with token-aware formatting.

Updated Feb 2, 2026
One-click install
npx skills add https://github.com/vuminhieu/spexor-client --skill repomix-vuminhieu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: repomix
Source: https://github.com/vuminhieu/spexor-client/tree/main/.agent/skills/skills/repomix
Command: npx skills add https://github.com/vuminhieu/spexor-client --skill repomix-vuminhieu

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Repomix packages entire code repositories into AI-friendly single-file formats, enabling efficient AI consumption and analysis.

Core Features & Use Cases

  • AI-optimized repository packaging: preserve file structure and context across formats (XML, Markdown, JSON, plain text).
  • Token-aware outputs: supports token counting and AI-friendly separators for large codebases.
  • Use Case: create repository snapshots for LLM context, security audits, documentation, or cross-repo comparisons.

Quick Start

Install and run Repomix to package a repository; start with defaults and adjust include/exclude patterns as needed.

Frequently Asked Questions about repomix

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

FAQPage Schema
How do I package a code repository into a single file for AI analysis?

To package a code repository for AI analysis, you can use Repomix to consolidate your codebase into a single AI-friendly file. This process preserves the file structure and context while applying token-aware formatting to optimize the output for large language models.

What output formats are supported for AI-friendly repository snapshots?

AI-friendly repository snapshots support multiple output formats including XML, Markdown, JSON, and plain text. These formats ensure that the packaged code repository retains its structure and context for efficient AI consumption and structured analysis.

How does token counting work when packaging large codebases for LLM context?

Token counting for large codebases works by applying token-aware formatting and AI-optimized separators during the repository packaging process. This ensures the generated single-file snapshot fits within LLM context limits while preserving the necessary code structure.

Can I include or exclude specific files when creating a repository snapshot?

Yes, you can use include and exclude patterns when creating a repository snapshot. This allows you to selectively package specific files or directories, ensuring the generated AI-friendly output contains only the relevant code for your analysis.

Does this approach work for both local and remote repositories?

Yes, this approach works for packaging both local and remote repositories. It enables you to generate AI-friendly single-file snapshots for AI-assisted code review, security audits, and documentation generation regardless of where the codebase is hosted.

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

The best way to prepare a codebase for an AI security audit is to package it into an AI-friendly single file using Repomix. This consolidates the repository with token-aware formatting and preserved structure, enabling the AI to efficiently analyze the entire codebase.