clauder-mcp 0.4.0

MCP server connecting AI assistants to RStudio for interactive R coding and data analysis

clauder-mcp

MCP server that connects AI assistants to RStudio for interactive R coding and data analysis.

This is the Python MCP bridge component of ClaudeR. It works with Claude Desktop, Claude Code, Codex CLI, Qwen Code CLI, Gemini CLI, Cursor, and any other MCP-compatible client.

Prerequisites

You need the ClaudeR R package installed and running in RStudio:

if (!require("devtools")) install.packages("devtools")
devtools::install_github("IMNMV/ClaudeR")
library(ClaudeR)
claudeAddin()

Start the server from the RStudio Viewer pane before connecting your AI tool.

Usage

Claude Code

claude mcp add --transport stdio --scope user r-studio -- uvx clauder-mcp

Codex CLI

codex mcp add r-studio -- uvx clauder-mcp

Qwen Code CLI

qwen mcp add --scope user --transport stdio r-studio uvx clauder-mcp

Gemini CLI

Add to your ~/.gemini/settings.json:

{
  "mcpServers": {
    "r-studio": {
      "command": "uvx",
      "args": ["clauder-mcp"]
    }
  }
}

Claude Desktop / Cursor

Add to your MCP config file:

{
  "mcpServers": {
    "r-studio": {
      "command": "uvx",
      "args": ["clauder-mcp"]
    }
  }
}

Tools

• execute_r - Execute R code and return output
• execute_r_with_plot - Execute R code that generates a plot
• execute_r_async - Run long-running code in a background process
• get_async_result - Poll for async job results
• read_file - Read any text file from disk
• get_active_document - Get the active document in RStudio
• modify_code_section - Edit code in the active document
• get_r_info - Get R environment information
• list_sessions - List available RStudio sessions
• connect_session - Connect to a specific session
• get_session_history - View execution history by agent
• create_task_list - Create a task list for analysis
• update_task_status - Update task progress

Links

• ClaudeR R Package - Full documentation, installation guide, and features
• License - MIT