IMNMV/ClaudeR

Connect RStudio to Claude Code, Codex, Gemini, and other LLM agents via MCP. Multi-agent orchestration, automated manuscript auditing, and zero-config setup with uvx

What it solves

ClaudeR connects RStudio to LLM agents (such as Claude Code, Codex, Qwen Code, and Gemini CLI) via the Model Context Protocol (MCP). It solves the problem of AI agents being unable to interact directly with a live R environment, allowing them to execute code, view plots, and perform complex data analysis tasks in real-time within the active RStudio session.

How it works

ClaudeR acts as an MCP server that exposes a suite of R-specific tools to an LLM agent. The agent can use these tools to execute R code (synchronously or asynchronously), read and modify files, capture viewer content from interactive widgets, and access the R environment. It supports multi-agent orchestration, where multiple agents can share a single session or operate in siloed windows, coordinating through a shared message board and execution logs.

Who it’s for

It is designed for researchers, data scientists, and statisticians who use RStudio and want to integrate AI agents into their coding, data analysis, and academic auditing workflows.

Highlights

  • Interactive Coding: Agents can write and execute code directly in RStudio, seeing results and plots in real-time.
  • Multi-Agent Orchestration: Support for multiple agents (e.g., Claude, Codex, Qwen) on one session with unique IDs and shared history.
  • Automated Academic Auditing: The "Reviewer Zero" protocol allows agents to extract statistical claims from manuscripts and verify them against R code.
  • AI-Driven Data Annotation: Tools for labeling CSV datasets row-by-row, either interactively or via isolated subprocesses (including local Ollama support).
  • Async Execution: Handles long-running R processes via callr to prevent session timeouts.
  • Reference Verification: Integrates with the CrossRef API to verify DOIs and metadata in academic bibliographies.