Integration · Websites & content
Connect Claude & ChatGPT to Umbraco
Ask your AI about the content on your Umbraco website, and it reads it through the Content Delivery API. It can fetch a page by its URL path, list the pages in a section like your blog, and find images and files in the media library. It only reads content, so nothing on your site changes.
Free plan, no credit card. Takes about a minute.
Try asking
“What does our /about-us/ page say about our history?”
The AI fetches the page by its path and reads the content.
“List our 10 newest blog posts.”
It queries the children of /blog/, sorted by update date, newest first, and takes 10.
“Which articles do we have about pricing?”
It queries published content filtered by the article content type and by name.
What your AI can do
Pages by path or ID
Fetch a page by its URL path or GUID, or several content items in one call.
Query the content tree
List the children or descendants of a section, filter by content type or name, sort by date and page through results.
Media library
Find images and files by path or ID, list folder contents and get their URLs and metadata.
“What is the URL of our logo in the media library?”
It looks up the media item by its path, such as /images/logo.png, and returns its URL and metadata.
“Check the draft of the new product page before we publish it.”
With preview on, it reads the unpublished draft of the content item by its ID. This needs an API key with preview access.
“Summarize every page under /services/ for our sales team.”
It fetches all descendants of the services section and writes a summary of each page.
8 tools for Umbraco Content Delivery API
These tools are switched on when you connect. You can switch any of them off, or require your approval before it runs.
get_content_by_idGet one published content item (page, article, etc.) by its GUID, passed as product_id. Set preview to true to read the unpublished draft.get_content_by_pathGet one published content item by its URL path, e.g. /about-us/. Use this when you know the page's address.get_content_itemsGet several content items at once by passing their GUIDs in id.get_media_by_idGet one media item (image, file) by its GUID, passed as product_id, including its URL and metadata.get_media_by_pathGet one media item by its media-library path, e.g. /images/logo.png.get_media_itemsGet several media items at once by passing their GUIDs in id.query_contentQuery published content. Use fetch for tree queries (e.g. children:/blog/ or descendants:{id}), filter for content type or name (e.g. contentType:article), sort (e.g. updateDate:desc), and skip/take via offset and take.query_mediaQuery the media library. Use fetch for folder queries (e.g. children:/images/), filter by media type or name, sort, and page with offset and take.
Set up in three steps
- 1
Pick the app
Create a free PipMCP account and choose Umbraco Content Delivery API from the app list.
- 2
Paste your key
Enter your website address and the Delivery API key. Your developer turns the Delivery API on and sets the key in appsettings (Umbraco:CMS:DeliveryApi:ApiKey); the same key also gives access to protected content and previews.
- 3
Add the link to your AI
You get a personal MCP link. Add it to Claude, ChatGPT or Cursor:
Click your name, then Settings › Connectors › Add custom connector. Paste your link as the Remote MCP server URL. Switch it on from the + menu in a chat.
Questions
What can the AI do in Umbraco?
It can read published content items and media from your site, by ID, by URL path or several at once. It can query the content tree and the media library with filters, sorting and paging. With a preview-enabled key it can also read unpublished drafts. It cannot create, edit or publish anything.
Does the AI see my Umbraco API key?
No. You paste your Umbraco Delivery API key into PipMCP once. It is encrypted at rest, never shown to the AI and never shown again after you save it. The AI only gets the results of the tools you have switched on.
Can I control what the AI is allowed to do?
Yes. You choose which tools are switched on, for example content only and no media. All Umbraco tools only read, so the AI cannot change your website. Every tool call is logged.
Does it work with ChatGPT?
Yes. In ChatGPT, open Settings › Apps & Connectors › Advanced, turn on Developer mode and add your PipMCP link. Developer mode needs a paid ChatGPT plan: Plus, Pro, Business or Enterprise. The same link also works in Claude (Settings › Connectors › Add custom connector) and in Cursor.
What does it cost?
PipMCP has a free plan, and you do not need a credit card to start. Paid plans are billed per completed task. You connect your own Umbraco website, and PipMCP only reads from it through the Content Delivery API.
Do I need a developer to set this up?
Someone with access to your Umbraco configuration has to enable the Delivery API in appsettings (Umbraco:CMS:DeliveryApi). For protected content or drafts, they also set an ApiKey there. After that, you enter your website address and key in PipMCP, and no code is needed.
Related integrations
Let your AI work in Umbraco Content Delivery API today.
Start free. Your key stays encrypted, and you decide what the AI may do.
Connect Umbraco Content Delivery API freePipMCP is not affiliated with Umbraco Content Delivery API. Product names are trademarks of their owners.