Rlang MCP Server

🚀 R-Server MCP

A professional Model Context Protocol (MCP) server that enables AI models to generate data visualizations using the ggplot2 library in R and execute R scripts.

🚀 Quick Start

The R-Server MCP is a powerful tool that simplifies the process of generating statistical charts and executing R scripts without direct access to the R environment. It exposes two MCP tools for seamless integration with AI models.

✨ Features

• ggplot2 Plotting: Execute R code containing ggplot2 commands and return the resulting charts.
• R Script Execution: Run any R script and receive text output.
• Format Options: Support for PNG, JPEG, PDF, and SVG output formats.
• Customization: Control image dimensions and resolution.
• Error Handling: Provide clear error messages for invalid R code or rendering failures.
• MCP Protocol Compliance: Full implementation of the Model Context Protocol.
• Docker Integration: Securely execute R code in isolated containers.

📦 Installation

Prerequisites

• Go 1.22 or higher.
• R 4.0 or higher, with the ggplot2 package installed.
• Docker (for containerized execution).

Build

# Build the Docker image
task docker:build

# Run the server in Docker
task docker:run

Using Docker with Standard Input/Output

The server can run in Docker while preserving standard input and output communication, which is crucial for MCP:

# Build and run using docker-compose
./start_server.sh --docker

Alternatively, set an environment variable:

USE_DOCKER=true ./start_server.sh

This method ensures proper connection of standard input/output between the host and the container, enabling seamless MCP communication.

💻 Usage Examples

MCP Integration

To use this server in an MCP client, configure it in the MCP configuration file:

Local Execution

{
  "mcpServers": {
    "r-server": {
      "command": "/path/to/r-server",
      "disabled": false,
      "autoApprove": []
    }
  }
}

Docker Execution

{
  "mcpServers": {
    "r-server": {
      "command": "/path/to/start_server.sh",
      "args": ["--docker"],
      "disabled": false,
      "autoApprove": []
    }
  }
}

The MCP client will automatically communicate with the server via stdio, which is the recommended method for stability and reliability. The Docker version provides isolation and dependency management while maintaining this communication mode.

📄 License

This work is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License.

For more details, see the LICENSE file.