Generator

🚀 CTX Context Generator

The CTX tool is designed to organize project code and related content into documents for sharing with Large Language Models (LLMs). By describing the project structure in a configuration file, CTX automatically generates .md files suitable for uploading to ChatGPT or Claude.

🚀 Quick Start

📦 Installation

Use the installation script to download and install the tool:

curl -sSL https://raw.githubusercontent.com/context-hub/generator/main/download-latest.sh | sh

This command installs the ctx command to the system (usually in the /usr/local/bin directory).

💻 Initializing the Configuration File

Create a new configuration file in the project directory:

ctx init

This generates a basic context.yaml file to help you get started quickly.

💻 Configuring the Project Structure Description

Edit the generated context.yaml file to specify the code or content to include. For example:

$schema: 'https://raw.githubusercontent.com/context-hub/generator/refs/heads/main/json-schema.json'

documents:
  - description: "User authentication system"
    outputPath: "auth-context.md"
    sources:
      - type: file
        description: "Authentication controller"
        sourcePaths:
          - src/Auth
        filePattern: "*.php"

      - type: file
        description: "Authentication model"
        sourcePaths:
          - src/Models
        filePattern: "*User*.php"

This configuration collects all PHP files in the src/Auth directory and any PHP files in the src/Models directory with names containing "User".

💻 Building the Context

Run the following command to generate the context file:

ctx

CTX will generate the specified output file (such as auth-context.md in the example) based on your configuration.

💻 Sharing the Context with an LLM

Upload or paste the generated context file into your preferred LLM (e.g., ChatGPT or Claude). Now you can ask specific questions about the codebase, and the LLM will have the necessary background information to provide accurate assistance.

Example Prompt:

I've shared my user registration system code with you. Can you help me identify potential security vulnerabilities in the user registration process?

💻 Optional Step: Connecting to Claude AI

Connect directly to Claude AI through CTX's built-in MCP server for a more seamless experience: Add the following to the configuration file:

{
  "mcpServers": {
    "ctx": {
      "command": "ctx server -c /path/to/your/project"
    }
  }
}

This allows Claude to directly access your codebase!

📚 Documentation

For more features and configuration options, visit: https://docs.ctxgithub.com

📄 License

This project is licensed under the MIT License.