MCP Server Installation

Connect Lambda Bench's MCP server to Cursor or Claude Code to get AI-tuned implementation instructions for common software components. This significantly improves correctness and reduces debugging loops.

Cursor Configuration

Configure Cursor to connect to the MCP server

Open Cursor Settings

Open the command palette (Cmd/Ctrl + Shift + P) and search for 'Open User Settings (JSON)'

Add MCP Server Configuration

Add the following configuration to your settings.json file:

{
  "mcpServers": {
    "code-examples": {
      "url": "https://api.lambdabench.com/api/mcp",
      "transport": "http"
    }
  }
}

Reload Cursor

Reload Cursor from the command palette by searching for 'Reload Window'

Verify Connection

You should see requests being made to https://lambdabench.com in the backend server's console logs when using MCP features

Claude Code Configuration

Configure Claude Code to connect to the MCP server

Locate Configuration File

Find your Claude Code configuration file based on your operating system:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Linux: ~/.claude.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

Create or Edit Configuration

Open the configuration file and add the MCP server configuration:

{
  "mcpServers": {
    "code-examples": {
      "url": "https://api.lambdabench.com/api/mcp",
      "type": "http"
    }
  }
}

Restart Claude Code

Completely quit and restart Claude Code for the changes to take effect

Verify Connection

Start a new conversation and ask Claude to 'list available MCP resources' or 'fetch the secret phrase resource'

Troubleshooting

Common Issues:

Connection errors: Verify that the URL in the configuration file is correct (`https://lambdabench.com/api/mcp`)
Configuration not loading: Ensure the JSON syntax is valid (no trailing commas, proper quotes, etc.)
No resources showing: Try restarting your AI coding assistant, then wait a few seconds for the connection to establish
Network issues: Check your internet connection and ensure you can access https://lambdabench.com

Testing the Connection:

Once configured, you can test the connection by asking your AI assistant to "list available MCP resources" or "fetch the secret phrase resource". The AI should be able to access resources from the Lambda Bench MCP server.