Repolode
IMNMV

IMNMV/ClaudeR

R and Python bridge that connects a live RStudio or Positron session to Claude Code, Codex, and other MCP-based LLM agents.

Stars
354
Forks
38
Open issues
17
Last commit
22 days ago

What is it?

ClaudeR lets MCP-configured LLM agents such as Claude Code or Codex work inside the R session you already have open. The agent runs code against your in-memory data and fitted models, sees the plots it draws, reads objects in your environment, and can edit the file open in your editor. A companion Python package provides the same for a live Python session, and both work in RStudio and Positron.

Multiple agents can share one session and divide the work, or run in separate sessions so each one works on its own dataset. Each execution is attributed to the agent that ran it and written to a session log, so there is a record of every line an agent executed. The project also bundles research-oriented workflows: a manuscript audit protocol called Reviewer Zero that reconciles reported numbers against code output, reference verification through CrossRef, systematic review screening, and response-to-reviewers drafting.

Installation uses R's devtools and a uvx-launched Python bridge (clauder-mcp), with installer helpers for Claude Desktop, Cursor, Claude Code, and Codex CLI.

Key features

  • 01Connects a live RStudio or Positron R or Python session to MCP-compatible agents.
  • 02Agents can see plots they draw and read objects in the session environment.
  • 03Multiple agents can share one session or run in separate sessions.
  • 04Every execution is attributed to its agent and written to a session log.
  • 05Includes Reviewer Zero, a protocol that reconciles manuscript values against code output.

Use cases

  • Asking an AI agent to analyze a dataset already loaded in an RStudio session.
  • Auditing a manuscript so that reported p-values and coefficients match the analysis code.
  • Running long model fits asynchronously while other agents keep working in the same session.
  • Screening systematic review abstracts with two independent model-based reviewers.

Quick start

  1. 1

    Install the ClaudeR R package from GitHub

    devtools::install_github("IMNMV/ClaudeR")
  2. 2

    Set up the CLI for Claude Code

    install_cli(tools = "claude")
  3. 3

    Start the server in RStudio

    claudeAddin()

Things to know

  • Installation is from GitHub via devtools, not CRAN.
  • Works only inside RStudio or Positron with an MCP-capable agent such as Claude Code or Codex.
  • The license is not clearly identified in the project metadata, although the README shows an MIT badge.

Topics

FAQ

What is ClaudeR?
R and Python bridge that connects a live RStudio or Positron session to Claude Code, Codex, and other MCP-based LLM agents. ClaudeR lets MCP-configured LLM agents such as Claude Code or Codex work inside the R session you already have open. The agent runs code against your in-memory data and fitted models, sees the plots it draws, reads objects in your environment, and can edit the file open in your editor. A companion Python package provides the same for a live Python session, and both work in RStudio and Positron.
Is ClaudeR free to use? What license does it have?
ClaudeR is open source under the Other license. Check the license terms before using it in commercial products.
Is ClaudeR still maintained?
Actively maintained. The last commit to ClaudeR was on Sep 30, 2026.
What can I use ClaudeR for?
Asking an AI agent to analyze a dataset already loaded in an RStudio session. Auditing a manuscript so that reported p-values and coefficients match the analysis code. Running long model fits asynchronously while other agents keep working in the same session. Screening systematic review abstracts with two independent model-based reviewers.
How do I install ClaudeR?
Install the ClaudeR R package from GitHub: devtools::install_github("IMNMV/ClaudeR") Set up the CLI for Claude Code: install_cli(tools = "claude") Start the server in RStudio: claudeAddin()
Where can I get ClaudeR?
The source code and installation instructions are on GitHub: https://github.com/IMNMV/ClaudeR

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects