> For the complete documentation index, see [llms.txt](https://docs.umbraco.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.umbraco.com/umbraco-in-ai/17.latest/mcp/base-mcp/create-umbraco-mcp-server/project-template.md).

# Project Template

When you create a project with `npx @umbraco-cms/create-umbraco-mcp-server`, you get a complete Umbraco MCP server project using the [`@umbraco-cms/mcp-server-sdk`](/umbraco-in-ai/17.latest/mcp/base-mcp/sdk.md) and [`@umbraco-cms/mcp-hosted`](/umbraco-in-ai/17.latest/mcp/base-mcp/hosted-mcp.md) packages.

## Project Structure

The project includes:

* **Server entry point** — A configured MCP server at `src/index.ts` that loads tool collections, applies filtering, and registers tools
* **API client** — [Orval](https://orval.dev/) configuration for generating a typed API client and Zod schemas from your Umbraco instance's OpenAPI spec
* **Tool collections** — A directory structure under `src/umbraco-api/tools/` where each collection groups related tools by HTTP method (`get/`, `post/`, `put/`, `delete/`)
* **Configuration** — Mode and slice registries in `src/config/` that control tool filtering. See [Configuration](/umbraco-in-ai/17.latest/mcp/base-mcp/sdk/configuration.md) and [Tool Filtering](/umbraco-in-ai/17.latest/mcp/base-mcp/sdk/tool-filtering.md) in the SDK documentation
* **Testing infrastructure** — Jest setup for both integration tests (against a real Umbraco instance) and LLM eval tests (using the Claude Agent SDK). See [Testing and Evals](/umbraco-in-ai/17.latest/mcp/base-mcp/sdk/testing.md) in the SDK documentation
* **Hosted Worker** — A Cloudflare Worker entry point at `src/worker.ts` using the [`@umbraco-cms/mcp-hosted`](/umbraco-in-ai/17.latest/mcp/base-mcp/hosted-mcp.md) package for remote deployment with OAuth authentication and multi-site support

## Optional Features

The [init phase](/umbraco-in-ai/17.latest/mcp/base-mcp/create-umbraco-mcp-server/development-workflow.md#phase-2-init) lets you keep or remove these optional features:

### Mock Infrastructure (MSW)

The `src/mocks/` directory contains Mock Service Worker handlers for testing without a running Umbraco instance. The mock server setup, API handlers, and in-memory data store provide a lightweight alternative for development.

### MCP Chaining

The `src/umbraco-api/mcp-client.ts` file configures connections to other MCP servers (such as `@umbraco-cms/mcp-dev`). MCP chaining lets you proxy tools from existing servers alongside your custom tools. See [MCP Chaining](/umbraco-in-ai/17.latest/mcp/base-mcp/sdk/mcp-chaining.md) in the SDK documentation.

### Example Tools

An example tool collection demonstrates the patterns for building your own tools. The init phase gives you the choice to remove these examples if you prefer a clean starting point.

### Agent Evaluation Tests

The `tests/evals/` directory provides an LLM evaluation framework with eval setup, helpers, and example test files. These tests verify that an LLM can use your tools to complete real workflows.

## Scripts

The project includes the following npm scripts:

| Script                  | Purpose                                                    |
| ----------------------- | ---------------------------------------------------------- |
| `npm run build`         | Build the MCP server with `tsup`                           |
| `npm run compile`       | Type-check without emitting files                          |
| `npm run watch`         | Rebuild on file changes                                    |
| `npm run generate`      | Regenerate the Orval API client from the OpenAPI spec      |
| `npm run start:umbraco` | Start the local Umbraco instance                           |
| `npm run inspect`       | Launch the MCP Inspector to test your server interactively |
| `npm run test`          | Run integration tests                                      |
| `npm run test:evals`    | Build and run LLM eval tests                               |
| `npm run test:all`      | Run both integration and eval tests                        |
| `npm run dev:worker`    | Start the Cloudflare Worker locally                        |
| `npm run deploy:worker` | Deploy the Cloudflare Worker                               |


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.umbraco.com/umbraco-in-ai/17.latest/mcp/base-mcp/create-umbraco-mcp-server/project-template.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.