Overview
What is R-Server MCP?
R-Server MCP is a specialized Model Context Protocol server that enables AI models to generate data visualizations using R's ggplot2 library and execute arbitrary R scripts. It exposes two tools (render_ggplot and execute_r_script) and is intended for developers integrating R capabilities into MCP-powered applications.
How to use R-Server MCP?
Configure the server in an MCP client's settings file, specifying either the local binary path or using the Docker startup script with the --docker flag. The server communicates via stdio transport and requires Go 1.22+, R 4.0+ with ggplot2, and optionally Docker for containerized execution.
Key features of R-Server MCP
- ggplot2 Rendering: Generate visualizations from R code containing ggplot2 commands.
- R Script Execution: Run any R script and return its text output.
- Output Formats: Supports PNG, JPEG, PDF, and SVG.
- Customization: Control image dimensions and resolution.
- Error Handling: Clear messages for invalid R code or rendering failures.
- Docker Integration: Isolated container execution for security and dependency management.
Use cases of R-Server MCP
- Generate statistical charts (scatter plots, bar charts, etc.) from AI-generated R ggplot2 code.
- Execute R scripts for data cleaning, analysis, or modeling and return textual results.
- Embed R-based data visualization in MCP-powered chat or assistant workflows.
- Run R code in a secure, containerized environment without exposing the host system.
FAQ from R-Server MCP
What exactly does R-Server MCP do?
It provides two MCP tools: render_ggplot to create plots from ggplot2 code and return them as image files, and execute_r_script to run any R code and capture text output. Both are accessed through the stdio MCP transport.
What are the runtime requirements?
The server requires Go 1.22 or later and R 4.0 or later with the ggplot2 package installed. Docker is optional but recommended for isolated execution.
Can I run R-Server MCP without Docker?
Yes. You can run the pre-built binary directly on a system that has R installed. Docker is only needed if you want containerized execution. The README shows configuration for both local and Docker execution.
How does the server handle R code errors?
If R code fails (invalid syntax, missing packages, etc.), the server returns clear error messages indicating the nature of the failure, without crashing the MCP server.
What output formats are supported for render_ggplot?
The tool supports PNG, JPEG, PDF, and SVG. Users can also control image dimensions and resolution.
