R Language MCP

Overview

R Language MCP is a secure Docker-based MCP server built with Python and FastMCP. It enables conversational R data visualization and analysis. All code runs in isolated containers for security.

Features

• Natural Language Analysis: R operations through Claude conversations
• Docker Security: Isolated container execution for full security
• File Integration: Local directory mounting for direct data access
• Publication-Quality Visualizations: Create ggplot2 charts through conversation
• Package Management: On-demand R package installation
• Smart Caching: Instant results for repeated operations
• Persistent Containers: Reuse the same environment (no startup delay)

Installation

macOS

brew install r uv python@3.12 --cask docker
uvx --from git+https://github.com/saidsurucu/rlang-mcp-python rlang-mcp-python

Linux (Ubuntu/Debian)

sudo apt install r-base python3.12 docker.io
pip install uv
uvx --from git+https://github.com/saidsurucu/rlang-mcp-python rlang-mcp-python

Windows

Download R and Python 3.12+, then:

pip install uv
uvx --from git+https://github.com/saidsurucu/rlang-mcp-python rlang-mcp-python

Usage Examples

Data Mounting

"Mount ~/Documents/analysis directory as R working directory."

File Discovery

"List all Excel files and show available data."

Analysis

"Load sales_data.xlsx, show summary statistics, calculate monthly totals."

Visualization

"Create a publication-quality bar chart of monthly sales by category."

Security

• All R code runs in isolated Docker containers
• No access to host system
• Volume mounting for secure file sharing
• Resource usage limits

Supported R Packages

• Data Processing: dplyr, tidyr, data.table
• Visualization: ggplot2, plotly, lattice
• Statistics: stats, MASS, car
• File Handling: readxl, readr, haven

License

MIT License