ClaudeR

Overview

ClaudeR is an R package that creates a direct connection between RStudio and MCP-configured LLM agents like Claude Code, Codex, or Qwen Code. It enables interactive coding sessions where AI agents can execute R code in real-time within the RStudio environment, view outputs, and analyze plots. The package supports multi-agent coordination, allowing multiple AI agents to work on one script or operate independently across different RStudio windows with siloed environments.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• execute_r

  Execute R code and return the output.

  Try:Run a linear regression on this dataTry:Calculate mean and standard deviation of this vector
• execute_r_with_plot

  Execute R code that generates a plot that the model can see.

  Try:Create a scatter plot of x vs yTry:Generate a histogram of this distribution
• execute_r_async

  Execute long-running R code asynchronously and returns a job ID.

  Try:Run a complex simulation that might take minutesTry:Process this large dataset in the background
• get_async_result

  Poll for the result of an asynchronous job.

  Try:Check if my simulation has completedTry:Get the results of my background data processing
• list_sessions

  List all active RStudio sessions the agent can connect to.

  Try:Show me all available R sessionsTry:What sessions are currently active
• connect_session

  Connect to a specific RStudio session by name.

  Try:Connect to my 'analysis' sessionTry:Switch to the 'modeling' session
• get_session_history

  View execution history filtered by agent ID.

  Try:Show me what other agents have done in this sessionTry:Get the execution history for agent 123
• read_file

  Read any text file from disk without needing it open in RStudio.

  Try:Read the contents of my analysis.R scriptTry:Load the parameters from config.csv
• get_active_document

  Get the content of the active document in RStudio.

  Try:Show me the current script in the editorTry:What's in the active document
• modify_code_section

  Modify a specific section of code in the active document.

  Try:Fix the function in lines 15-25Try:Update the data cleaning section
• insert_text

  Insert text at the current cursor position or specific line/column.

  Try:Add this import at the top of the fileTry:Insert this comment before line 10
• get_viewer_content

  Read HTML content from the viewer pane (plotly, DT, leaflet widgets).

  Try:Extract the data table from the viewerTry:Get the plot details from the interactive plot

Comparable tools

Installation

# Install using devtools
if (!require("devtools")) install.packages("devtools")
devtools::install_github("IMNMV/ClaudeR")

# Set up your AI tool
library(ClaudeR)
install_clauder()          # For Claude Desktop / Cursor
install_cli(tools = "claude")  # For Claude Code CLI

# Start the server in RStudio
claudeAddin()

{
  "mcpServers": {
    "clauder": {
      "command": "Rscript",
      "args": ["-e", "library(ClaudeR); claudeServe()"],
      "env": {
        "R_PROFILE_USER": ""
      }
    }
  }
}

FAQ

Which LLM agents are supported?

ClaudeR supports Claude Code, Codex, Gemini CLI, Qwen Code, and any other MCP-based LLM agents. It can also work with Ollama for private, offline labeling.

Can multiple AI agents work on the same R session?

Yes, ClaudeR has built-in multi-agent coordination protocol that allows multiple agents to share one RStudio session through a shared message board in the R environment.

Compare ClaudeR with