Configure MCP Server

Set up MCP server connections for Claude Code (WSL) and Claude Desktop (Windows).

When to Use

• Setting up Claude Code to connect to R via mcptools
• Configuring Claude Desktop with MCP servers
• Adding Hugging Face or other remote MCP servers
• Troubleshooting MCP connectivity between tools

Inputs

• Required: MCP server type (mcptools, Hugging Face, custom)
• Required: Client (Claude Code, Claude Desktop, or both)
• Optional: Authentication tokens
• Optional: Custom server implementation

Procedure

Step 1: Install MCP Server Packages

For R (mcptools):

install.packages("remotes")
remotes::install_github("posit-dev/mcptools")

For Hugging Face:

npm install -g mcp-remote

Got: mcptools installs from GitHub and loads in R without errors. mcp-remote is available globally via which mcp-remote or npm list -g mcp-remote.

If fail: For mcptools, ensure remotes is installed first. If GitHub rate-limits the install, set a GITHUB_PAT in ~/.Renviron. For mcp-remote, verify Node.js and npm are installed and on PATH.

Step 2: Configure Claude Code (WSL)

R mcptools server:

claude mcp add r-mcptools stdio \
  "/mnt/c/Program Files/R/R-4.5.0/bin/Rscript.exe" \
  -- -e "mcptools::mcp_server()"

Hugging Face server:

claude mcp add hf-mcp-server \
  -e HF_TOKEN=your_token_here \
  -- mcp-remote https://huggingface.co/mcp

Verify configuration:

claude mcp list
claude mcp get r-mcptools

Got: claude mcp list shows both r-mcptools and hf-mcp-server (or whichever servers were added). claude mcp get r-mcptools displays the correct command and arguments.

If fail: If the server does not appear in the list, verify ~/.claude.json contains the correct entry. If the claude command is not found, add it to PATH: export PATH="$HOME/.claude/local/node_modules/.bin:$PATH".

Step 3: Configure Claude Desktop (Windows)

Edit %APPDATA%\Claude\claude_desktop_config.json:

{
  "mcpServers": {
    "r-mcptools": {
      "command": "C:\\PROGRA~1\\R\\R-45~1.0\\bin\\x64\\Rscript.exe",
      "args": ["-e", "mcptools::mcp_server()"]
    },
    "hf-mcp-server": {
      "command": "mcp-remote",
      "args": ["https://huggingface.co/mcp"],
      "env": {
        "HF_TOKEN": "your_token_here"
      }
    }
  }
}

Important: Use 8.3 short paths for Windows directories with spaces (PROGRA~1 not Program Files). Use environment variables for tokens, not --header arguments.

Got: The JSON config file at %APPDATA%\Claude\claude_desktop_config.json is valid JSON with the correct server entries. Claude Desktop shows MCP server indicators after restart.

If fail: Validate the JSON with a linter (e.g., jq . < config.json). Use 8.3 short paths (PROGRA~1) if Windows path spaces cause parsing errors. Ensure Claude Desktop is fully restarted (not merely minimized).

Step 4: Configure R Session for MCP

Add to project .Rprofile:

if (requireNamespace("mcptools", quietly = TRUE)) {
  mcptools::mcp_session()
}

This starts the MCP session automatically when opening the project in RStudio.

Got: The .Rprofile file conditionally starts mcptools::mcp_session() when the project is opened in RStudio, making MCP tools available automatically.

If fail: If mcptools is not found at session start, verify it is installed in the library that RStudio uses (check .libPaths()). If using renv, ensure mcptools is in the renv library.

Step 5: Verify Connections

Test R MCP from WSL:

"/mnt/c/Program Files/R/R-4.5.0/bin/Rscript.exe" -e "mcptools::mcp_server()"

Test from within Claude Code:

Start Claude Code and use MCP tools — they should appear in the tool list.

Test Claude Desktop:

Restart Claude Desktop after configuration changes. Check for MCP server indicators in the UI.

Got: Running Rscript with mcptools::mcp_server() produces output without errors. MCP tools appear in the Claude Code tool list during an active session. Claude Desktop shows server status after restart.

If fail: If the Rscript command fails, check the full path is correct (ls "/mnt/c/Program Files/R/" to verify R version). If tools don't appear in Claude Code, restart the session. For Claude Desktop, check firewall settings.

Step 6: Multi-Server Configuration

Both Claude Code and Claude Desktop support multiple MCP servers simultaneously:

# Claude Code: add multiple servers
claude mcp add r-mcptools stdio "/path/to/Rscript.exe" -- -e "mcptools::mcp_server()"
claude mcp add hf-mcp-server -e HF_TOKEN=token -- mcp-remote https://huggingface.co/mcp
claude mcp add custom-server stdio "/path/to/server" -- --port 3001

Got: Multiple MCP servers configured and accessible simultaneously. claude mcp list shows all servers. Each server's tools are available in the same Claude Code session.

If fail: If servers conflict, check that each has a unique name in the configuration. If one server blocks others, verify servers use non-blocking I/O (stdio transport handles this automatically).

Validation

• claude mcp list shows all configured servers
• R MCP server responds to tool calls
• Hugging Face MCP server authenticates and responds
• Both Claude Code and Claude Desktop can connect (if both configured)
• MCP tools appear in the tool list during sessions

Pitfalls

• Windows path spaces: Use 8.3 short names or quote paths correctly. Different tools parse paths differently.
• Token in command args: On Windows, --header "Authorization: Bearer token" fails due to parsing. Use environment variables instead.
• Confusing Claude Code and Claude Desktop configs: These are separate tools with separate config files (~/.claude.json vs %APPDATA%\Claude\)
• npx vs global install: npx mcp-remote may fail in Claude Desktop context. Install globally with npm install -g mcp-remote.
• mcptools version: Ensure mcptools is up to date. It requires the ellmer package as a dependency.

Related Skills

• build-custom-mcp-server - creating your own MCP server
• troubleshoot-mcp-connection - debugging connection issues
• setup-wsl-dev-environment - WSL setup prerequisite