Skip to main content
This page collects example configurations for connecting popular MCP clients to Lexful. In each example, replace {POD} with us or eu for your account’s pod (see MCP endpoint) and {YOUR_ACCOUNT_ID} with your account ID.

Example: Claude (claude.ai)

Claude’s chat interface supports MCP via Connectors.
  1. Go to Settings → Connectors.
  2. Click Add custom connector.
  3. Enter a name (for example, Lexful) and paste your MCP URL:
  1. Click Add and complete the OAuth flow when prompted.
  2. Lexful will then be available as a connector in your chats.

Example: Claude Desktop

Claude Desktop supports remote MCP servers through the same Connectors UI as claude.ai.
  1. Go to Settings → Connectors.
  2. Click Add custom connector.
  3. Enter a name (for example, Lexful) and paste your MCP URL:
  1. Click Add and complete the OAuth flow when prompted.
If you prefer to configure via JSON, you can use the mcp-remote wrapper in claude_desktop_config.json, then restart Claude Desktop so it picks up the change:
Keep the version pinned. Versions of mcp-remote before 0.1.16 let a malicious MCP server run commands on your machine, and an unpinned npx mcp-remote runs whatever version npm resolves.

Example: Claude Code

Claude Code has built-in MCP support. Add Lexful as a server from the CLI:
By default this applies to the current project. Use --scope user to make it available globally, or --scope project to share it with your team via .mcp.json. To complete the OAuth flow, start a Claude Code session, type /mcp, select lexful, and choose Authenticate. You can also use /mcp to manage your MCP servers interactively.

Example: Cursor

  1. Open Cursor Settings (Cmd/Ctrl + ,).
  2. Navigate to Tools & MCP.
  3. Click New MCP Server and add the following configuration:
This can also be added directly to ~/.cursor/mcp.json (global) or .cursor/mcp.json (project-level).

Example: ChatGPT

ChatGPT supports MCP via Apps & Connectors on eligible plans (Plus, Pro, Team, Enterprise, Edu).
  1. Ensure Developer Mode is enabled under Settings → Apps & Connectors → Advanced settings.
  2. Click Add new connector.
  3. Enter a name (for example, Lexful) and paste your MCP URL:
  1. Set the auth mode to OAuth and complete the authorization when prompted.
  2. In a new chat, enable the connector via the apps menu.

Example: Microsoft Copilot Studio

Copilot Studio supports MCP as a tool type for agents.
1

Navigate to the Agents page

In Copilot Studio, go to the Agents page. Create a new agent or select an existing one.
2

Open the Tools tab

Inside your agent, click the Tools tab, then click + Add a tool.
3

Select Model Context Protocol

In the “Add tool” dialog, select Model Context Protocol under “Create new.”
4

Configure the Lexful MCP server

Fill in the following details:
  • Server name: Lexful MCP
  • Server description: Lexful is our IT documentation and asset management platform
  • Server URL:
  • Authentication: Select OAuth 2.0, then choose Dynamic discovery as the type.
5

Create and enable the tool

Click Create. The tool will appear in your agent’s Tools list. Ensure the Enabled toggle is on, then test in the “Test your agent” panel.

Example: VS Code

VS Code supports MCP servers through GitHub Copilot’s Agent mode. Add a .vscode/mcp.json file to your workspace:
Reload VS Code so the change takes effect. Lexful tools will be available when using Copilot in Agent mode.

Example: Generic MCP client

For other MCP-aware tools that accept a JSON configuration, the standard format is:
Refer to your client’s documentation for the exact config file location and key names, as these vary between tools.