This bridges R sessions with MCP-enabled AI tools like Claude Desktop and VS Code Copilot, letting models execute R code directly in your running sessions. You can query dataframes, inspect your global environment, read package documentation, and run arbitrary R functions without leaving your chat interface. It also works in reverse as an MCP client, so you can register third-party MCP servers with ellmer chats to pull in external context from GitHub, Confluence, or Google Drive. Essentially turns your R environment into a live workspace that AI assistants can interact with programmatically.
claude mcp add --transport stdio posit-dev-mcptools uvx mcptools