MCP Catalogs
← Homemcptools screenshot

mcptools

by posit-dev·★ 174·Score 48

mcptools implements the Model Context Protocol in R, enabling R to act as both MCP server and client for AI integration.

developer-toolsai-llm
15
Forks
11
Open issues
2 mo ago
Last commit
2d ago
Indexed

Overview

mcptools is a comprehensive implementation of the Model Context Protocol for the R programming language. It enables R to function as both an MCP server, allowing AI tools like Claude Desktop to execute R code in active R sessions, and as an MCP client to register third-party MCP servers within R environments. The package provides two main functionalities: R as an MCP server exposes tools for reading package documentation, running R code, and inspecting global environment objects; while R as an MCP client allows integration of third-party servers with ellmer chats. This bidirectional capability creates a powerful bridge between R and AI assistants.

Try asking AI

After installing, here are 3 things you can ask your AI assistant:

you:Allowing Claude Desktop to analyze data in active R sessions by executing R code
you:Integrating third-party MCP servers like GitHub with R-based applications through ellmer
you:Enabling AI assistants to access and manipulate R environments and package documentation

When to choose this

Choose mcptools if you're working with R and want AI assistants to interact with your R sessions, access your environment variables, or execute code directly within your R workflow.

When NOT to choose this

Avoid mcptools if you're not working with R, as it's specifically designed for the R ecosystem and won't be useful for other programming languages.

Tools this server exposes

3 tools extracted from the README
  • mcp_servermcp_server()

    Initializes the mcptools MCP server for R

  • mcp_sessionmcp_session()

    Registers an R session to be accessed by MCP clients

  • mcp_toolsmcp_tools()

    Returns a list of MCP tools from configuration

Note: The documentation describes MCP server functionality but doesn't provide a complete list of individual tools. The tools mentioned are primarily initialization functions rather than specific data processing tools.

Comparable tools

acquaintellmerbtw

Installation

Install mcptools from CRAN:

install.packages("mcptools")

Or install the development version:

pak::pak("posit-dev/mcptools")

To use R as an MCP server with Claude Desktop, add this to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "r-mcptools": {
      "command": "Rscript",
      "args": ["-e", "mcptools::mcp_server()"]
    }
  }
}

To use R as an MCP client, create a configuration file at ~/.config/mcptools/config.json with your desired MCP servers.

Compare mcptools with

mcptools vs ultimate_mcp_servermcptools vs mcp-server-chartmcptools vs everythingmcptools vs filesystemmcptools vs time
GitHub →

Last updated · Auto-generated from public README + GitHub signals.