# Umbraco in AI

Umbraco in AI depends on more than prompts. It depends on tools, context, and instructions working together.

This section introduces the core concepts behind AI-assisted Umbraco workflows. It covers how AI connects to Umbraco, how context shapes output quality, and how agent skills guide coding assistants toward better results.

### Core concepts

<table data-view="cards"><thead><tr><th></th><th></th><th data-hidden data-card-cover data-type="image">Cover image</th></tr></thead><tbody><tr><td><strong>Model Context Protocol (MCP)</strong></td><td>Learn how MCP connects LLMs to tools, APIs, and structured Umbraco capabilities.</td><td data-object-fit="contain"><a href="https://2684268505-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNnUDSAShVcdETvjguL9v%2Fuploads%2F7V6FWhbBHupt1ptFPj53%2Fmcp.svg?alt=media&#x26;token=c4174c76-324f-41a2-83f9-901edfc7cd71">mcp.svg</a></td></tr><tr><td><strong>Context Engineering</strong></td><td>Learn how to shape prompts, tools, and context so models produce better results.</td><td data-object-fit="contain"><a href="https://2684268505-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNnUDSAShVcdETvjguL9v%2Fuploads%2FQpUoTOhycemc05hiUGCL%2Fbrain.svg?alt=media&#x26;token=a3e263a6-7073-4bf2-846e-9c49846296be">brain.svg</a></td></tr><tr><td><strong>Agent Skills</strong></td><td>Learn how reusable instructions help coding agents apply Umbraco patterns correctly.</td><td data-object-fit="contain"><a href="https://2684268505-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNnUDSAShVcdETvjguL9v%2Fuploads%2F8gwBijqGMS0YwCmBb4JC%2Fhat-glasses.svg?alt=media&#x26;token=e086fd72-ab05-463f-a115-94a90b4eaa32">hat-glasses.svg</a></td></tr></tbody></table>

### Why these concepts matter

Strong AI output depends on the quality of the system around it. In Umbraco, that means giving the model the right tools, the right context, and the right guidance.

When these parts work together, AI can help with content operations, schema planning, technical workflows, and developer productivity. When they do not, output becomes vague, inefficient, or incorrect.

### Explore each topic

* [Model Context Protocol (MCP)](https://docs.umbraco.com/umbraco-ai-overview/model-context-protocol): explains how LLMs connect to Umbraco tools, APIs, and external systems.
* [Context Engineering](https://docs.umbraco.com/umbraco-ai-overview/context-enginerring): explains how to improve reliability by controlling what the model sees.
* [Agent Skills](https://docs.umbraco.com/umbraco-ai-overview/agent-skills): explains how agent skills provide reusable instructions for AI coding tools.

### What you will learn

* How AI connects to Umbraco through MCP
* How context engineering improves relevance and reduces hallucinations
* How agent skills help AI assistants follow current Umbraco patterns