Vectorcode
A code repository indexing tool to supercharge your LLM experience.
What is Vectorcode?
Vectorcode is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to code repository indexing tool to supercharge your llm experience.
A code repository indexing tool to supercharge your LLM experience.
This server falls under the Developer Tools and Knowledge & Memory categories on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- A code repository indexing tool to supercharge your LLM expe
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx vectorcodeConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Vectorcode
VectorCode is a Python-based code repository indexing tool that builds a local vector database (powered by ChromaDB and sentence-transformer embeddings) from your codebase, then exposes that index to AI assistants via an MCP server. This lets AI clients like Claude retrieve semantically relevant code snippets from large or proprietary repositories — reducing hallucination and dramatically improving code-generation quality for projects the model has never seen before. It is especially valuable for teams working on internal codebases, emerging libraries, or monorepos that exceed a model's context window.
Prerequisites
- Python 3.10+ and the uv package manager installed (https://astral.sh/uv)
- A local code repository you want to index (any language supported by tree-sitter)
- ChromaDB accessible — either started locally (`chroma run`) or as a remote server
- An MCP-compatible client such as Claude Desktop, Cursor, or a Neovim plugin with MCP support
Install VectorCode with MCP support
Install the vectorcode package including the optional MCP extra using uv, the recommended package manager.
uv tool install 'vectorcode[mcp]<1.0.0'Initialize a project and index your repository
Navigate to your project root and initialize VectorCode, then vectorize your source files. VectorCode respects .gitignore automatically.
cd /path/to/your/project
vectorcode init
vectorcode vectorise .Verify the index is populated
List the indexed files to confirm the vectorization completed successfully.
vectorcode files lsStart the VectorCode MCP server
Launch the MCP server process. It exposes five tools: ls, query, vectorise, files_ls, and files_rm.
vectorcode-mcp-serverAdd the server to your MCP client configuration
Register vectorcode-mcp-server in claude_desktop_config.json so your AI client can query the indexed codebase.
{
"mcpServers": {
"vectorcode": {
"command": "vectorcode-mcp-server",
"args": [],
"env": {
"VECTORCODE_LOG_LEVEL": "INFO"
}
}
}
}Vectorcode Examples
Client configuration
claude_desktop_config.json entry for the VectorCode MCP server with optional logging level.
{
"mcpServers": {
"vectorcode": {
"command": "vectorcode-mcp-server",
"args": [],
"env": {
"VECTORCODE_LOG_LEVEL": "INFO"
}
}
}
}Prompts to try
Prompts that exercise VectorCode's semantic code search and indexing tools.
- "Search my indexed codebase for functions that handle authentication"
- "List all collections currently indexed in VectorCode"
- "Find code related to database connection pooling in this project"
- "Show me the files indexed under the 'my-app' collection"
- "Index the new files I added under src/modules and then search for error handling patterns"Troubleshooting Vectorcode
vectorcode-mcp-server not found after installation
Ensure uv tool installs are on your PATH. Run `uv tool update-shell` or add `~/.local/bin` to your PATH in your shell profile. Verify with `which vectorcode-mcp-server`.
ChromaDB connection errors when vectorizing or querying
VectorCode defaults to a local ChromaDB at http://127.0.0.1:8000. Start ChromaDB with `chroma run` in a separate terminal, or configure `db_url` in ~/.config/vectorcode/config.json to point to your ChromaDB instance.
Query returns empty results despite files being indexed
Confirm the correct project root is active — VectorCode uses .git or .vectorcode.json to identify projects. Run `vectorcode ls` and `vectorcode files ls` to verify your collection exists and contains files.
Frequently Asked Questions about Vectorcode
What is Vectorcode?
Vectorcode is a Model Context Protocol (MCP) server that code repository indexing tool to supercharge your llm experience. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Vectorcode?
Follow the installation instructions on the Vectorcode GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Vectorcode?
Vectorcode works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Vectorcode free to use?
Yes, Vectorcode is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Vectorcode Alternatives — Similar Developer Tools Servers
Looking for alternatives to Vectorcode? 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 Vectorcode 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 Vectorcode?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.