This is an MCP server for interacting with the VRChat API. You can retrieve information about friends, worlds, avatars, and more in VRChat.
Unlock the potential of VRChat data within your AI applications using the VRChat MCP server. This open-source project provides a standardized interface for accessing a wealth of VRChat information, empowering you to build intelligent and context-aware experiences.
The VRChat MCP server acts as a bridge between the VRChat API and the Model Context Protocol. This allows you to retrieve and utilize VRChat data, such as user profiles, friend lists, avatar details, and world information, in a structured and consistent manner within your LLM-powered applications.
Before launching the server, ensure you have the following environment variables configured:
export VRCHAT_USERNAME=your_username export VRCHAT_PASSWORD=your_password export VRCHAT_TOTP_SECRET=your_totp_secret export VRCHAT_EMAIL=[email protected]
Important Security Note:
[!WARNING]
TOTP Secret Acquisition
Obtaining your TOTP secret involves extracting it from the QR code displayed when enabling Two-Factor Authentication on the VRChat website. This method carries inherent security risks. Exercise extreme caution and consider the implications before proceeding.
With the environment variables set, initiate the MCP server using the following command:
npx vrchat-mcp
This command starts the server, enabling you to interact with the VRChat API through the MCP interface.
For users of Claude Desktop, manual server execution is unnecessary. Integrate the VRChat MCP server directly by adding the following configuration to your Claude Desktop config file:
~/Library/Application Support/Claude/claude_desktop_config.json
%APPDATA%\Claude\claude_desktop_config.json
{ "mcpServers": { "vrchat-mcp": { "command": "npx", "args": ["vrchat-mcp"], "env": { "VRCHAT_USERNAME": "your-username", "VRCHAT_PASSWORD": "your-password", "VRCHAT_TOTP_SECRET": "your-totp-secret", "VRCHAT_EMAIL": "[email protected]" } } } }
After adding the configuration, launch Claude Desktop as usual. If you are using nodenv
or nvm
, ensure that the full path to the npx
command is specified.
The following table outlines the VRChat API endpoints currently supported or planned for implementation within the VRChat MCP server. Both GET and POST methods are considered to provide a complete feature set.
Debugging MCP servers that operate over stdio can be complex. For an optimal debugging experience, the MCP Inspector is highly recommended.
Build the Project:
npm install npm run build
Launch the MCP Inspector:
npx @modelcontextprotocol/inspector "./dist/main.js"
Ensure that all necessary environment variables are correctly configured before launching the inspector.
The Inspector will provide a URL to access a browser-based debugging interface.
We encourage contributions to enhance the VRChat MCP server. Fork the repository and submit pull requests to contribute improvements and bug fixes.
๐ VOYP Voice Over Your Phone MCP Server for making calls.
๐ An MCP server with openAPI specs for using the WhatsApp unnoficial API (https://waha.devlike.pro/ also open source: https://github.com/devlikeapro/waha
๐ ๐ JMeter MCP Server for performance testing
๐ โ๏ธ Slack workspace integration for channel management and messaging