• JSON
  • MEDIA
  • QUERY
  • HUMAN

How to fetch delivery api data with your AI

Connect your Umbraco instance once. From then on, your agent pulls the specific content and media you need for your frontend.

Ask AI about this page

Short answer

How can I get Umbraco content into my AI?

Your agent uses the Umbraco MCP to grab structured JSON and media assets via direct paths or IDs. You'll get the raw data needed to feed your frontend components.

Data outcomes

Where your data lands.

The AI pulls information from Umbraco and organizes it for your specific needs.

JSON

Structured content

The AI fetches raw JSON data. You get the exact properties needed for your components.

MEDIA

Asset retrieval

Your agent finds media files in the CMS. You get the links and assets for your site.

QUERY

Filtered results

The AI searches the content tree. You get a specific subset of data that matches your filters.

HUMAN

Manual selection

The AI presents the data it found. You decide which content is actually ready to go live.

The workflow

What your AI does when content is requested.

The AI handles the communication between your agent and the Umbraco delivery API.

  1. Find content

    The AI locates specific nodes using their unique paths or IDs.

    get_delivery_content_by_pathget_delivery_content_by_id

  2. Scan the tree

    Your agent looks through the available content to see what exists.

    list_delivery_content

  3. Search for data

    The AI runs queries to find content that meets your specific criteria.

    query_delivery_content

  4. Collect assets

    The agent pulls the media files associated with your content nodes.

    list_media_assets

Try it

Copy these to start.

Use these prompts to start pulling data from Umbraco.

Starting points

Replace the bracketed text with your actual Umbraco paths or content IDs.

Umbraco Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_Q8Bf5msx9nS9Avd1WticWndyaU4yMUEsTqaFluBx/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Umbraco capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "umbraco-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_Q8Bf5msx9nS9Avd1WticWndyaU4yMUEsTqaFluBx/mcp"
    }
  }
}
Copy into chat04

  • Get the JSON content at /content/home/about-us.

  • List all media assets in the Umbraco media library.

  • Find the content item with ID 550e8400.

  • Query for all content items where the category is 'news'.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Umbraco once, then ask.

You connect your instance once. Your credentials stay encrypted, and you can immediately start asking your agent for data.

Connect Umbraco to your AI

FAQ

How this task behaves.

  • 01

    Can the AI edit content in Umbraco?

    No. This MCP is for fetching and listing data only. It cannot change or delete anything in your CMS.

  • 02

    Can I get images through the AI?

    Yes. The AI can list and retrieve media assets from your Umbraco instance.

  • 03

    Can the AI create new content nodes?

    No. The available tools are strictly for reading and querying existing content and media.

  • 04

    Does the AI need the exact path to find content?

    It's the most direct way, but the AI can also list or query the content tree to find what you need.

  • 05

    Can the AI delete media files?

    No. There are no tools in this MCP that allow for the deletion of assets or content.

  • More questions about Umbraco? See everything the Umbraco Connector can do

Connect Umbraco to Claude, Cursor, ChatGPT & more