Model Context Protocol

Userplane’s MCP server gives AI tools like Claude, Cursor, Goose, and ChatGPT direct access to your documentation — so they can search it inline while generating responses, rather than relying on web search.

Connect to your AI tool

Your Userplane MCP server is available at:

https://docs.userplane.io/mcp

Claude
Claude Code
Cursor
VS Code

Open Connectors

Navigate to the Connectors page in your Claude settings.

Add a custom connector

Select Add custom connector.

Enter server details

Enter a name (e.g., userplane) and the MCP server URL:

https://docs.userplane.io/mcp

Select Add.

Use it in a conversation

Open a new conversation, select the attachments button (the + icon), then select your Userplane connector.

Run the following command in your terminal:

claude mcp add --transport http userplane https://docs.userplane.io/mcp

Verify the connection:

claude mcp list

The userplane server should appear in the list.

See the Claude Code documentation for more details.

Open the command palette

Press Cmd+Shift+P (or Ctrl+Shift+P on Windows).

Open MCP settings

Search for Open MCP settings and select it. This opens mcp.json.

Configure the server

Add Userplane to the mcpServers object:

{
  "mcpServers": {
    "userplane": {
      "url": "https://docs.userplane.io/mcp"
    }
  }
}

Test the connection

In Cursor’s chat, ask “What tools do you have available?” — Cursor should list the Userplane MCP server.

See Installing MCP servers in the Cursor documentation for more details.

Create the config file

Create a .vscode/mcp.json file in your project root.

Configure the server

Add Userplane as a server:

{
  "servers": {
    "userplane": {
      "type": "http",
      "url": "https://docs.userplane.io/mcp"
    }
  }
}

See the VS Code documentation for more details.

MCP vs web search

AI tools can search the web, but MCP provides distinct advantages for documentation access.

	Web search	MCP
Source	Search engine index — may be stale or incomplete	Your current indexed documentation, directly
Workflow	Separate search step before or after a response	Searches inline during response generation
Signal quality	SEO and ranking algorithms influence results	Straight to your documentation content, no noise

Rate limits

To protect availability, Userplane applies the following rate limits to MCP requests.

Scope	Limit
Per user (IP address)	5,000 requests / hour
Per documentation site	1,000 requests / hour

Getting Started

AI Agents

SDK Reference

Framework Guides

Privacy & Data

Connect to your AI tool

MCP vs web search

Rate limits

Getting Started

AI Agents

SDK Reference

Framework Guides

Privacy & Data

​Connect to your AI tool

​MCP vs web search

​Rate limits

Connect to your AI tool

MCP vs web search

Rate limits