ck:repomix

Converts selected Git repositories into a single, transportable file for AI consumption.

Updated Mar 7, 2026
One-click install
npx skills add https://github.com/MinhHoangDono/antigravity-kit --skill ck-repomix-minhhoangdono
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck:repomix
Source: https://github.com/MinhHoangDono/antigravity-kit/tree/main/.agent/skills/repomix
Command: npx skills add https://github.com/MinhHoangDono/antigravity-kit --skill ck-repomix-minhhoangdono

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Repomix eliminates the tedious work of manually collecting and formatting large codebases for AI review by packaging an entire repository into a single AI-friendly output.

Core Features & Use Cases

  • AI-friendly repository snapshots: Generates XML, Markdown, JSON, or plain-text representations with clear separators to improve LLM readability.
  • Curated context preparation: Supports include/exclude glob filtering (and respects .gitignore by default) to focus on relevant files and reduce token usage.
  • Security-aware packaging: Runs sensitive-data detection (Secretlint) and supports comment removal to reduce noise for audits and analysis.
  • Use cases: Use for codebase snapshots, architecture/code review context, third-party library evaluation, and security audits across large repositories.

Quick Start

Tell the AI assistant: Package the current repository into a single Markdown file for AI code review with comments removed.

Frequently Asked Questions about ck:repomix

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

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

To package a repository for AI code review, you can snapshot the codebase into a single XML, Markdown, JSON, or plain-text file. This consolidates all relevant content into an AI-friendly format with clear separators to improve LLM readability.

Can I filter specific files when creating an AI-ready repository snapshot?

Yes, you can filter specific files when creating a repository snapshot by using include and exclude glob patterns. The packaging process also respects .gitignore rules by default to help reduce token usage and focus on relevant modules.

Does repository packaging support secret scanning for security audits?

Yes, repository packaging supports security audits by running sensitive-data detection using Secretlint. This security-aware packaging identifies sensitive data while optionally removing code comments to reduce noise during analysis.

What is the best way to prepare large codebases for LLM context without manual file collection?

The best way to prepare large codebases for LLM context is automated repository packaging. It eliminates manual file collection by generating a single deliverable artifact containing the repository’s content, formatted specifically to improve LLM readability.

Can I package remote repositories or does it only work with local codebases?

You can package both local and remote repositories into consolidated artifacts. The process generates a single deliverable file containing the repository’s content, making it suitable for third-party library evaluation and architecture review prep.