Skip to main content

Connecting Your MCP Server to AI Clients

After creating your MCP server in Cequence AI Gateway, connect it to your preferred AI client to start using it. This section provides detailed configuration instructions for each supported client.

Supported AI Clients​

Claude Desktop​

Configure your MCP server with Claude Desktop using Custom Connectors (recommended) or the configuration file. Anthropic's official desktop application with native MCP support.

Claude Code​

Connect your MCP server to Claude Code, Anthropic's AI-powered coding assistant that runs in your terminal. Uses CLI commands for simple configuration.

Cursor IDE​

Connect your MCP server to Cursor, the AI-powered code editor built for maximum productivity.

VS Code​

Set up your MCP server with Visual Studio Code through GitHub Copilot's MCP integration. Supports both workspace and global configurations.

Windsurf IDE​

Set up your MCP server with Windsurf, Codeium's AI IDE featuring the Cascade AI system.

Quick Start​

After creating your MCP server, you'll receive a server URL from Cequence AI Gateway.

Use Custom Connectors (UI-based):

  1. Open Claude Desktop → Settings → Connectors
  2. Click "Add custom connector"
  3. Enter your MCP server URL
  4. Click "Add" and authenticate

Claude Code​

Use CLI command:

claude mcp add --transport http your-server-name "your-mcp-server-url"

Or use NPX:

npx @cequenceai/mcp-cli@latest claudecode --url "your-mcp-server-url" --name "your-server-name"

Cursor IDE​

Configuration format:

{
  "mcpServers": {
    "your-server-name": {
      "url": "your-mcp-server-url"
    }
  }
}

VS Code​

Configuration format (in .vscode/mcp.json):

{
  "servers": {
    "your-server-name": {
      "type": "http",
      "url": "your-mcp-server-url"
    }
  }
}

Or use NPX:

npx @cequenceai/mcp-cli@latest vscode --url "your-mcp-server-url" --name "your-server-name"

Claude Desktop & Windsurf (Configuration File)​

Alternative configuration file format:

{
  "mcpServers": {
    "your-server-name": {
      "command": "npx",
      "args": [
        "-y",
        "@cequenceai/mcp-remote",
        "your-mcp-server-url"
      ]
    }
  }
}

Replace:

  • your-server-name - A descriptive name for this integration (e.g., "salesforce", "github", "databricks")
  • your-mcp-server-url - The URL provided by Cequence AI Gateway

Configuration File Locations​

Each client stores its MCP configuration in a different location:

ClientConfiguration Location
Claude Desktop (macOS)~/Library/Application Support/Claude/claude_desktop_config.json
Claude Desktop (Windows)%APPDATA%\Claude\claude_desktop_config.json
Claude CodeCLI-based (no config file) - use claude mcp add
Cursor (macOS/Linux)~/.cursor/config/mcp.json
Cursor (Windows)%USERPROFILE%\.cursor\config\mcp.json
VS Code (Workspace).vscode/mcp.json in project root
VS Code (macOS Global)~/Library/Application Support/Code/User/mcp.json
VS Code (Windows Global)%APPDATA%\Code\User\mcp.json
VS Code (Linux Global)~/.config/Code/User/mcp.json
WindsurfVia Settings UI: Advanced → Cascade

Next Steps​

  1. Choose your AI client from the list above
  2. Follow the detailed configuration guide for your client
  3. Start using your MCP server with natural language commands

Getting Help​

Each client configuration page includes:

  • Step-by-step setup instructions
  • Configuration examples
  • Troubleshooting common issues
  • Best practices

For additional support, visit Cequence AI Gateway Documentation or contact support@cequence.ai.