AutoDev Codebase
A vector embedding-based code semantic search tool with MCP server and multi-model integration. Can be used as a pure CLI tool. Supports Ollama for fully local embedding and reranking, enabling complete offline operation and privacy protection for yo
What is AutoDev Codebase?
AutoDev Codebase is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to vector embedding-based code semantic search tool with mcp server and multi-model integration. can be used as a pure cli tool. supports ollama for fully local embedding and reranking, enabling complete...
A vector embedding-based code semantic search tool with MCP server and multi-model integration. Can be used as a pure CLI tool. Supports Ollama for fully local embedding and reranking, enabling complete offline operation and privacy protection for yo
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- A vector embedding-based code semantic search tool with MCP
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx autodev-codebaseConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use AutoDev Codebase
autodev-codebase (@autodev/codebase) is a vector embedding-based semantic code search and analysis tool that exposes an MCP server for IDE and AI assistant integration. It indexes your codebase using tree-sitter for 40+ languages and stores embeddings in a local Qdrant vector database, enabling natural-language code discovery, call graph analysis, and code outline extraction. When paired with Ollama, it operates entirely offline with no cloud services required, making it suitable for proprietary codebases where privacy is a concern.
Prerequisites
- Node.js 18 or later and npm for installing the CLI
- Ollama installed and running locally (for offline/private mode); alternatively an OpenAI, Jina, Gemini, or other cloud embedding provider API key
- Qdrant vector database running locally via Docker (port 6333)
- ripgrep (rg) installed for fast file search (brew install ripgrep on macOS)
- An MCP-compatible AI client (Claude Desktop, Cursor, or similar)
Install system dependencies
Install Ollama (for local embeddings) and ripgrep (for fast file search), then start Ollama and pull the embedding model.
# macOS
brew install ollama ripgrep
ollama serve &
ollama pull nomic-embed-textStart Qdrant vector database
Run Qdrant in Docker. The autodev-codebase CLI will store all code embeddings here.
docker run -d -p 6333:6333 -p 6334:6334 --name qdrant qdrant/qdrantInstall the autodev-codebase CLI
Install the npm package globally to get the codebase command available system-wide.
npm install -g @autodev/codebaseConfigure the embedding provider
Set the embedding provider and model. For fully offline use, choose ollama with the nomic-embed-text model.
codebase config --set embedderProvider=ollama,embedderModelId=nomic-embed-textIndex your codebase
Navigate to your project root and run the index command. This processes all source files, generates embeddings, and stores them in Qdrant. Large codebases may take several minutes.
cd /path/to/your/project
codebase indexStart the MCP server and configure your client
Launch the MCP server in SSE mode on a local port, then add it to your MCP client config.
# Start MCP server
codebase index --serve --port=3001
# Client config (SSE transport)
{
"mcpServers": {
"autodev-codebase": {
"transport": "sse",
"url": "http://localhost:3001/sse"
}
}
}AutoDev Codebase Examples
Client configuration
MCP config connecting to the autodev-codebase SSE server running on port 3001.
{
"mcpServers": {
"autodev-codebase": {
"transport": "sse",
"url": "http://localhost:3001/sse"
}
}
}Prompts to try
Use semantic search and call graph tools to navigate your codebase via natural language.
- "Find all code related to user authentication and session management"
- "Show me the call graph for the addUser function"
- "Generate an outline of the src/api/routes.ts file with AI summaries"
- "Search for code that handles database connection pooling"
- "Find all places where the payment service is invoked"Troubleshooting AutoDev Codebase
Embedding errors or connection refused to Qdrant
Verify Qdrant is running: docker ps | grep qdrant. If the container is not listed, start it again with the docker run command. The default qdrantUrl is http://localhost:6333 — confirm this matches your container port mapping.
Ollama embedding model not found
Run ollama pull nomic-embed-text to download the model, and make sure ollama serve is running (check with ollama list). If Ollama is not on the default port, set qdrantUrl or configure embedderOpenAiCompatibleBaseUrl to point at your Ollama instance.
Indexing is very slow on large codebases
Add a .codebaseignore file (gitignore syntax) to the project root to exclude node_modules, dist, build, and other generated directories. You can also increase throughput by using a cloud embedding provider like OpenAI instead of local Ollama.
Frequently Asked Questions about AutoDev Codebase
What is AutoDev Codebase?
AutoDev Codebase is a Model Context Protocol (MCP) server that vector embedding-based code semantic search tool with mcp server and multi-model integration. can be used as a pure cli tool. supports ollama for fully local embedding and reranking, enabling complete offline operation and privacy protection for yo It connects AI assistants to external tools and data sources through a standardized interface.
How do I install AutoDev Codebase?
Follow the installation instructions on the AutoDev Codebase GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with AutoDev Codebase?
AutoDev Codebase works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is AutoDev Codebase free to use?
Yes, AutoDev Codebase is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
AutoDev Codebase Alternatives — Similar Developer Tools Servers
Looking for alternatives to AutoDev Codebase? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up AutoDev Codebase in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use AutoDev Codebase?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.