Furi
CLI & API for MCP management
What is Furi?
Furi is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to cli & api for mcp management
CLI & API for MCP management
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- CLI & API for MCP management
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx furiConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Furi
Furi is a local CLI and HTTP API for managing MCP (Model Context Protocol) servers, allowing developers to install, start, stop, and interact with MCPs programmatically. It acts as an MCP aggregator that exposes tools from all running MCPs to AI clients through either an SSE endpoint or stdio mode. Furi is ideal for developers who want to orchestrate multiple MCP servers, automate MCP lifecycle management, or build HTTP-accessible MCP infrastructure without manual configuration.
Prerequisites
- Bun runtime (automatically installed by Furi's install script if not present)
- A Unix-like operating system (macOS or Linux)
- GitHub access token (optional but recommended to avoid API rate limits)
- An MCP client such as Claude Desktop or any MCP-compatible AI assistant
- curl available in your terminal for the install script
Install Furi
Run the official install script which downloads Furi and installs Bun automatically if it is not already present on your system.
curl -fsSL https://furi.so/install | bashAdd an MCP from GitHub
Install any MCP server hosted on GitHub by providing the author/repo path. Furi downloads and registers it in its .furikake config directory.
furi add <author/repo>Start an MCP server with environment variables
Launch a registered MCP with the required environment variables. Use the -e flag with a JSON object containing the key-value pairs the MCP needs.
furi start <mcpName> -e '{"API_KEY":"your-key"}'Start the MCP aggregator (meta server)
Run Furi's meta server to expose all running MCPs through a single SSE endpoint. Clients that support SSE can connect to this to access every registered tool.
furi meta start -p 9338Start the HTTP API server
Enable the REST API to manage and call MCPs via HTTP. Add --sudo to unlock install and lifecycle management routes, and set HTTP_AUTH_TOKEN to secure the API.
HTTP_AUTH_TOKEN=your-secret-token furi http start -p 9339 --sudoCall an MCP tool via CLI
Execute any tool exposed by a running MCP directly from the terminal using furi call, passing arguments as a JSON string.
furi call <mcpName> <toolName> '{"param":"value"}'Furi Examples
Client configuration (stdio mode)
Connect Claude Desktop to all Furi-managed MCPs through stdio using furi connect, which acts as a stdio bridge to the aggregated tool set.
{
"mcpServers": {
"furi": {
"command": "furi",
"args": ["connect"],
"env": {
"GITHUB_KEY": "your-github-token"
}
}
}
}Prompts to try
Example natural language prompts you can use once Furi is connected to your AI client.
- "List all the MCP servers currently running through Furi"
- "What tools are available from the mcp-fetch server?"
- "Install the modelcontextprotocol/servers MCP from GitHub and start it"
- "Show me the logs for the last 20 lines of the filesystem MCP"Troubleshooting Furi
furi command not found after installation
Restart your shell or source your profile (source ~/.bashrc or source ~/.zshrc) so the PATH update from the install script takes effect.
GitHub API rate limit errors when adding MCPs
Set the GITHUB_KEY environment variable to a personal access token. This raises your rate limit from 60 to 5000 requests per hour.
HTTP API returns unauthorized errors
Ensure you set HTTP_AUTH_TOKEN before starting the HTTP server and include the same token as a Bearer token in the Authorization header of your requests.
Frequently Asked Questions about Furi
What is Furi?
Furi is a Model Context Protocol (MCP) server that cli & api for mcp management It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Furi?
Follow the installation instructions on the Furi GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Furi?
Furi works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Furi free to use?
Yes, Furi is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.
Furi Alternatives — Similar Developer Tools Servers
Looking for alternatives to Furi? 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 Furi 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 Furi?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.