MCP.so
Sign In

Code-Index-MCP (Local-first Code Indexer)

@ViperJuice

Basic information

Category

Developer Tools

Transports

stdio

Author

ViperJuice

About Code-Index-MCP

Code indexing MCP server to provide context to coding agents.

Code-Index-MCP config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "Code-Index-MCP": {
      "command": "python",
      "args": [
        "-m",
        "venv",
        "venv"
      ]
    }
  }
}

Code-Index-MCP tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Code-Index-MCP?

Code-Index-MCP is a local-first code indexer designed to enhance Claude Code and other LLMs with deep code understanding capabilities. It is built on the Model Context Protocol (MCP) for seamless integration with AI assistants and is intended for developers who need fast, private code analysis.

How to use Code-Index-MCP?

Clone the repository, create a Python virtual environment, install dependencies from requirements.txt, and start the server with uvicorn mcp_server.gateway:app --reload --host 0.0.0.0 --port 8000. Test the API using curl endpoints like /status and /search. Optionally configure via a .env file (e.g., VOYAGE_AI_API_KEY for semantic search). Docker deployment options are also provided.

Key features of Code-Index-MCP

  • Local-first architecture for speed and privacy
  • Plugin-based design for extensibility
  • Multi-language support (Python, C/C++, JavaScript, Dart, HTML/CSS)
  • Real-time file system monitoring for instant index updates
  • AI-powered semantic search with Voyage AI embeddings
  • Rich code intelligence: symbol resolution, type inference, dependency tracking

Use cases of Code-Index-MCP

  • Enhance AI assistants with deep understanding of local codebases
  • Search for code patterns across multiple languages using regex or semantic queries
  • Get symbol definitions and references for navigation and refactoring
  • Index projects in real-time for up-to-date AI-assisted development
  • Deploy as a private, self-hosted code intelligence backend

Frequently asked questions

1

What dependencies does Code-Index-MCP require?

Python 3.8+ and several pip packages (listed in `requirements.txt`). For semantic search, a Voyage AI API key is optional. Docker is optional for architecture visualization or containerized deployment.

2

Where does indexed data live?

All indexing happens locally on your machine by default. The local index uses SQLite with FTS5; no data is sent to external servers unless you configure a cloud feature (optional).

3

What languages are supported and how complete is the indexer?

Six languages are fully implemented: Python, C, C++, JavaScript/TypeScript, Dart, and HTML/CSS. The project is 95% complete with a comprehensive testing framework. Rust, Go, Ruby, Swift, Kotlin, Java, and TypeScript are planned.

4

What performance targets does the server meet?

Target SLAs include symbol lookup under 100ms (p95), code search under 500ms (p95), and file indexing of 10,000 files per minute. Memory usage is expected under 2GB for 100,000 files. Core functionality is implemented; benchmark results are pending publication.

5

Which transport and authentication mechanisms are supported?

The server exposes a RESTful API over HTTP (FastAPI) and speaks the MCP protocol. Authentication is not mentioned in the README; the server can be configured with host/port settings in the `.env` file. Security measures include local-first processing, path validation, input sanitization, and secret detection.

Comments

More Developer Tools MCP servers