CodeDox

Provides an API-backed MCP server to index, search, and extract code from documentation with AI-assisted context.

typescript

24

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

CodeDox is an MCP server that enables AI-assisted searching and extraction of code snippets from documentation sites. It combines intelligent crawling, fast full‑text search, and a dedicated MCP interface to help AI assistants query and retrieve relevant code examples with context.

How to use

You connect a client to the CodeDox MCP HTTP endpoint to access tools that expose documentation code data, search results, and live crawl information. Use the MCP URL provided below to register CodeDox as a data source for your AI assistant, then run queries to locate code snippets, explanations, and markdown context across documentation pages.

How to install

Prerequisites you need installed on your system before starting:

Git
Docker
bash
a modern terminal

Step 1. Clone the project repository.

# Clone the repository
git clone https://github.com/chriswritescode-dev/codedox.git
cd codedox

Step 2. Prepare environment configuration.

cp .env.example .env

Then edit the environment file to add your API key if you plan to enable AI-enhanced extraction.


Step 3. Start the services using the recommended Docker setup.

./docker-setup.sh


Step 4. Verify access to the web interfaces.

Web UI: http://localhost:5173 MCP endpoints: http://localhost:8000/mcp

## Configuration notes

CodeDox supports HTTP-based MCP integration by exposing endpoints under the MCP path. The Docker setup provides a default HTTP MCP endpoint you can connect to from your MCP client.

## Security and tips

Keep your CODE\_LLM\_API\_KEY secure. If you enable AI-enhanced extraction, store credentials safely and restrict access to the MCP URL to trusted clients.

## Available tools

### http\_mcp\_integration

Exposes MCP tools via HTTP endpoints with Streamable HTTP transport for MCP 2025-03-26.

### full\_text\_search

PostgreSQL full-text search with fuzzy matching to provide fast, relevant code search results.

### code\_extraction

Smart code extraction with dual-mode titles and descriptions (automatic or LLM-generated).

### documentation\_upload

Upload documentation or fetch content from GitHub repositories with path support for docs.

### dashboard\_ui

Modern React + TypeScript UI for visual management and monitoring of crawl, search, and data.