Connect Your AI Tools to LayerZero Docs

Step-by-step guide to connect Cursor, VS Code, Claude, ChatGPT, Windsurf, JetBrains, Zed, and other AI tools to LayerZero documentation via MCP.

Connect your AI tools to the LayerZero documentation MCP server so they can search the docs in real-time while generating responses.

MCP server URL:

https://docs.layerzero.network/mcp

IDEs

Cursor

Option 1: One-click install

Visit any page on docs.layerzero.network and click the contextual menu button (top-right). Select Connect to Cursor to install the MCP server automatically.

Option 2: Manual configuration

Add the server to your Cursor MCP config file:

Global (all projects): ~/.cursor/mcp.json

Project-level: <project-root>/.cursor/mcp.json

{
  "mcpServers": {
    "layerzero-docs": {
      "url": "https://docs.layerzero.network/mcp"
    }
  }
}

MCP tools are only available in Cursor’s Agent mode, not in Ask or normal chat mode.

VS Code

Option 1: One-click install

Visit any page on docs.layerzero.network and click the contextual menu button (top-right). Select Connect to VS Code to install the MCP server automatically.

Option 2: Manual configuration

Create or edit .vscode/mcp.json in your project root:

{
  "servers": {
    "layerzero-docs": {
      "type": "http",
      "url": "https://docs.layerzero.network/mcp"
    }
  }
}

VS Code uses "servers" as the top-level key (not "mcpServers"). Requires VS Code 1.99+ with GitHub Copilot.

MCP tools only appear in Copilot Agent Mode (not inline or standard chat).

Windsurf

Add the server to the Windsurf MCP config:

Config file: ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "layerzero-docs": {
      "serverUrl": "https://docs.layerzero.network/mcp"
    }
  }
}

You can also add it through the UI: open Windsurf Settings and navigate to Cascade > MCP Servers.

Windsurf uses "serverUrl" for remote MCP servers, not "url".

JetBrains

Supported in IntelliJ IDEA, WebStorm, PyCharm, and other JetBrains IDEs (version 2025.2+).

1. Open Settings > Tools > AI Assistant > Model Context Protocol (MCP).
2. Click the + button.
3. Select HTTP as the connection type.
4. Paste the configuration:

{
  "mcpServers": {
    "layerzero-docs": {
      "url": "https://docs.layerzero.network/mcp"
    }
  }
}

5. Click OK, then Apply.

Requires the AI Assistant plugin to be installed and active.

AI assistants

Claude Code

Run a single command to add the server:

claude mcp add --transport http layerzero-docs https://docs.layerzero.network/mcp

Verify it was added:

claude mcp list

Scope options:

Flag	Scope
--scope user	Available across all your projects
--scope project	Shared via .mcp.json at project root (committable)

Use /mcp inside a Claude Code session to check server status.

Claude Desktop

Option 1: Connectors UI

1. Open Claude Desktop and go to Settings > Connectors.
2. Click Add custom connector.
3. Enter LayerZero Docs as the name and paste the URL:
   https://docs.layerzero.network/mcp
   
4. Click Add.
5. In a conversation, click the attachments button (plus icon) and select the LayerZero Docs connector.

Requires a Claude Pro, Max, Team, or Enterprise plan.

Option 2: Config file with mcp-remote

Edit your Claude Desktop config file:

Platform	Path
macOS	~/Library/Application Support/Claude/claude_desktop_config.json
Windows	%APPDATA%\Claude\claude_desktop_config.json
Linux	~/.config/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "layerzero-docs": {
      "command": "npx",
      "args": [
        "mcp-remote@latest",
        "https://docs.layerzero.network/mcp"
      ]
    }
  }
}

The config file method requires Node.js to be installed since it uses the mcp-remote proxy.

ChatGPT

1. Open chatgpt.com and go to Settings > Apps & Connectors > Advanced settings.
2. Toggle on Developer Mode.
3. In the Connectors section, click to create a new connector.
4. Paste the MCP server URL:
   https://docs.layerzero.network/mcp
   
5. Save and start a new chat.

Requires a ChatGPT Pro, Team, Business, Enterprise, or Edu plan. Only available on the web version.

Codex CLI

Add the server to your Codex config:

Global: ~/.codex/config.toml

Project-level: <project-root>/.codex/config.toml

[mcp_servers.layerzero-docs]
url = "https://docs.layerzero.network/mcp"

Or use the CLI:

codex mcp add layerzero-docs --url https://docs.layerzero.network/mcp

Use /mcp inside a Codex session to check server status.

Codex CLI supports Streamable HTTP only, not SSE. If you encounter connection issues, confirm the server transport is compatible.

Other MCP clients

Zed

Edit your Zed settings file (open via Zed > Settings > Open Settings or Cmd+,):

{
  "context_servers": {
    "layerzero-docs": {
      "command": {
        "path": "npx",
        "args": ["-y", "mcp-remote", "https://docs.layerzero.network/mcp"]
      }
    }
  }
}

Zed uses the mcp-remote wrapper for remote MCP servers. Requires Node.js to be installed. A green indicator dot in the Agent Panel confirms the server is connected.

Any MCP client

Add the server URL to your MCP client’s configuration:

https://docs.layerzero.network/mcp

The server uses HTTP transport (Streamable HTTP). Most MCP clients auto-detect the transport type from the URL.

Verifying the connection

After setup, test that your AI tool can reach the LayerZero docs:

1. Open your AI tool and start a new conversation or agent session.
2. Ask a question about LayerZero, for example:
   • “What is the OFT standard in LayerZero?”
   • “How do I configure DVN security for my OApp?”
   • “What chains does LayerZero support?”
3. The AI should pull answers directly from the docs and include links to relevant pages.

If the AI does not search the docs, check that:

• The MCP server is enabled and connected (check your tool’s MCP settings panel).
• You are in Agent mode (Cursor, VS Code) rather than standard chat.
• The server URL is exactly https://docs.layerzero.network/mcp.