Modal MCP Toolbox

🚀 Modal MCP Toolbox 🛠️

A collection of Model Context Protocol (MCP) tools running on Modal. This allows you to extend the capabilities of LLMs in applications such as Goose or Claude Desktop App.

✨ Features

• run_python_code_in_sandbox: Enables you to run Python code in a sandbox environment.
• generate_flux_image: Generates images using the FLUX model.

🎥 Demos

Flux Image Generation

Python Code Execution

📋 Prerequisites

• A Modal account and a configured Modal CLI.
• UV
• An MCP-compatible client. For example, Claude Desktop App or Goose

This tool runs on your Modal account, so you need to have a Modal account and be logged in.

📦 Installation

Installation depends on the MCP client you are using. Instructions are provided here for Claude and Goose.

Claude

Go to Settings > Developer in the Claude Desktop App. Then click on Edit Configuration.

Add the configuration for the MCP server. My configuration looks like this:

{
  "mcpServers": {
    "modal-toolbox": {
      "command": "uvx",
      "args": ["modal-mcp-toolbox"]
    }
  }
}

Goose

Go to Settings and then click on Add.

Then add an extension as shown in the following figure. The important part is to set the command to:

uvx modal-mcp-toolbox

You can fill in the rest as you like.

Installation via Smithery (Currently Unavailable)

To automatically install the Modal MCP Toolbox for Claude Desktop via Smithery:

npx -y @smithery/cli install @philipp-eisen/modal-mcp-toolbox --client claude