BaseHub MCP

This project was created with create-xmcp-app. The MCP (Model Context Protocol) integration enables AI agents to interact directly with your BaseHub repository through a comprehensive set of tools. From creating and updating content blocks to managing assets and automating workflows, you can build functional websites, migrate hardcoded content, and set up forms—all through natural language prompts with your favorite AI tools.

This was part of our AI Week venture. Learn more in our changelog

How to use it

You can check the user documentation in basehub.com/dcos/ai/mcp

Development

First, run the development server:

npm run dev
# or
yarn dev
# or
pnpm dev

This will start the MCP server with both SSE and STDIO transport methods.

Project Structure

This project uses the structured approach where tools are automatically discovered from the src/tools directory. Each tool is defined in its own file with the following structure:

import { z } from "zod";
import { type InferSchema } from "xmcp";

// Define the schema for tool parameters
export const schema = {
  a: z.number().describe("First number to add"),
  b: z.number().describe("Second number to add"),
};

// Define tool metadata
export const metadata = {
  name: "add",
  description: "Add two numbers together",
  annotations: {
    title: "Add Two Numbers",
    readOnlyHint: true,
    destructiveHint: false,
    idempotentHint: true,
  },
};

// Tool implementation
export default async function add({ a, b }: InferSchema<typeof schema>) {
  return {
    content: [{ type: "text", text: String(a + b) }],
  };
}

Adding New Tools

To add a new tool:

Create a new .ts file in the src/tools directory
Export a schema object defining the tool parameters using Zod
Export a metadata object with tool information
Export a default function that implements the tool logic

Building for Production

To build your project for production:

npm run build
# or
yarn build
# or
pnpm build

This will compile your TypeScript code and output it to the dist directory.

Running in Production

To run your bundled MCP server in production:

npm run start-sse
# or
npm run start-stdio

XMCP Application

README Documentation