Windsurf IDE

Overview

Windsurf is an AI-powered IDE by Codeium that features Cascade, an innovative agentic AI system. With Lemonado's MCP integration, Windsurf can access your business data while maintaining deep understanding of your codebase.

Prerequisites

• Windsurf IDE installed (latest version recommended)
• Active Lemonado AI Connector with MCP endpoint
• macOS, Windows, or Linux system

Setup Instructions

Step 1: Open Windsurf MCP Settings

1. Launch Windsurf IDE
2. Open the Command Palette:
   • macOS: Cmd+Shift+P
   • Windows/Linux: Ctrl+Shift+P
3. Type "MCP" and select "Windsurf: Configure MCP Servers"

Step 2: Add Lemonado MCP Server

In the MCP configuration file that opens, add your Lemonado server:

{
  "mcpServers": {
    "lemonado": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-fetch",
        "https://mcp.lemonado.io/mcp"
      ],
      "env": {
        "MCP_AUTH_TYPE": "oauth",
        "MCP_OAUTH_PROVIDER": "lemonado"
      },
      "description": "Lemonado data connector for business intelligence"
    }
  }
}

Step 3: Authenticate with OAuth

When you first use the Lemonado connector, you'll be prompted to authenticate via OAuth. This will open a browser window where you can log in to your Lemonado account and select which AI Connector to use.

Step 4: Activate the Connection

1. Save the configuration file
2. Reload Windsurf window:
   • Command Palette → "Developer: Reload Window"
3. The MCP connection will be established automatically

Step 5: Verify Connection

Open Cascade Chat (Cmd+L or Ctrl+L) and test:

@lemonado List available data sources

You should see your connected Lemonado data sources.

Using Lemonado with Cascade

Cascade Flows

Windsurf's Cascade can create complex workflows combining your code and data:

Create a dashboard component that displays our top 10 customers by revenue using @lemonado data

Cascade will:

1. Query your customer data via Lemonado
2. Analyze your existing component structure
3. Generate a type-safe dashboard component
4. Include proper error handling and loading states

Multi-Step Data Operations

@lemonado Analyze our user engagement metrics and suggest code improvements to increase retention

Real-time Data Validation

Validate this API endpoint against @lemonado production data schema

Advanced Features

Cascade + MCP Workflows

Windsurf's Cascade AI excels at multi-step operations:

1. @lemonado Fetch product catalog structure
2. Generate TypeScript types from the schema
3. Create CRUD operations with validation
4. Write unit tests using sample data
5. Generate API documentation

Intelligent Code Generation

Based on @lemonado customer data, create a recommendation engine service

Cascade will:

• Analyze your data patterns
• Suggest appropriate algorithms
• Implement the service with your tech stack
• Add monitoring and logging

Data-Driven Refactoring

@lemonado Show usage patterns, then refactor this service for better performance

Configuration Options

Multiple Environments

Configure separate connectors for different environments, each with its own OAuth authentication:

{
  "mcpServers": {
    "lemonado-dev": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-fetch",
        "https://mcp.lemonado.io/mcp"
      ],
      "env": {
        "MCP_AUTH_TYPE": "oauth",
        "MCP_OAUTH_PROVIDER": "lemonado"
      },
      "description": "Development data"
    },
    "lemonado-prod": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-fetch",
        "https://mcp.lemonado.io/mcp"
      ],
      "env": {
        "MCP_AUTH_TYPE": "oauth",
        "MCP_OAUTH_PROVIDER": "lemonado"
      },
      "description": "Production data"
    }
  }
}

Custom Commands

Create shortcuts for common queries:

{
  "mcpServers": {
    "lemonado": {
      "command": "npx",
      "args": ["@modelcontextprotocol/server-fetch", "--url", "https://mcp.lemonado.io/mcp"],
      "aliases": {
        "customers": "SELECT * FROM customers LIMIT 10",
        "revenue": "SELECT SUM(amount) FROM transactions WHERE date >= NOW() - INTERVAL '30 days'"
      }
    }
  }
}

Performance Optimization

For large datasets, configure caching:

{
  "mcpServers": {
    "lemonado": {
      "command": "npx",
      "args": [
        "@modelcontextprotocol/server-fetch",
        "--url",
        "https://mcp.lemonado.io/mcp",
        "--cache",
        "--cache-ttl",
        "300"
      ]
    }
  }
}

Cascade Superpowers

1. Contextual Understanding

Cascade understands both your code and data:

Why is this function slow? Check against @lemonado production query patterns

2. Automated Testing

Generate tests with real data patterns:

Create integration tests using @lemonado sample data for edge cases

3. Migration Planning

Plan database migration from @lemonado current schema to new microservices architecture

4. Performance Analysis

Analyze @lemonado query logs and optimize our most expensive operations

Best Practices

Use Cascade Flows

Leverage Cascade's ability to break down complex tasks:

Flow: Implement customer segmentation
1. Analyze @lemonado customer data distribution
2. Design segmentation algorithm
3. Implement service with caching
4. Add analytics tracking
5. Create admin interface

Combine Code and Data Context

Reference both your codebase and data:

Update the UserService to handle @lemonado new customer fields

Iterative Development

Start simple and enhance:

// First iteration
Create basic CRUD for @lemonado products

// Second iteration
Add validation based on @lemonado business rules

// Third iteration
Optimize queries using @lemonado performance metrics

Windsurf-Specific Features

Inline Data Preview

Hover over queries to see live data previews:

// Hover to see actual data
const customers = await lemonado.query('SELECT * FROM customers LIMIT 5');

Smart Completions

Windsurf autocompletes based on your actual schema:

// Autocompletes with real column names
const query = 'SELECT cust|
                       customer_id, customer_name, created_at

Cascade Commands

Built-in commands for common operations:

• /lemonado-schema - View complete data schema
• /lemonado-sample - Get sample data for testing
• /lemonado-analyze - Analyze query performance
• /lemonado-diff - Compare data structures

Troubleshooting

Cascade Not Recognizing MCP

1. Ensure MCP server is properly configured
2. Check Windsurf console for errors (View → Output → MCP)
3. Verify API key has correct permissions

Slow Queries

• Enable query caching in configuration
• Use specific field selections instead of SELECT *
• Check Lemonado dashboard for rate limiting

Connection Timeouts

Increase timeout in configuration:

{
  "mcpServers": {
    "lemonado": {
      "command": "npx",
      "args": ["..."],
      "timeout": 60000
    }
  }
}

Authentication Issues

Tips for Power Users

Quick Data Exploration

Use Cascade's split view to explore data while coding:

Split: Show @lemonado customers table | Edit CustomerService.ts

Generate Migrations

Compare @lemonado production schema with our models and generate migrations

Automated Documentation

Document all API endpoints with @lemonado example responses

Performance Profiling

Profile this function against @lemonado real workload patterns

Windsurf vs Other IDEs

Security Considerations

Next Steps

• Learn about Cascade workflows
• Explore MCP protocol details