Umbraco

Umbraco MCP Connector for Claude

A+

Automate content workflows via Umbraco — retrieve delivery content, execute backoffice CRUD, and browse media assets directly from your AI agent.

10 tools Official Updated Jun 28, 2026 Official Vinkius Partner

Connect your Umbraco CMS backend to any AI agent and take full autonomous control bridging the powerful Delivery and Management APIs purely through natural conversation.

What you can do

  • Delivery API Traversing — Instantly list public pages, query by content type, or securely fetch structured fields by their exact domain paths organically
  • Backoffice Document Control — Push new document permutations natively adhering to your configured schemas without opening a single GUI panel
  • Site Mutations — Command the targeted removal of any outdated published nodes or force updates to internal fields seamlessly via update_cms_document
  • Schema & Media Insight — Grab absolute lists tracking your stored binary media files alongside the global Document Types blueprints mapped out natively

How it works

  1. Subscribe manually to this connected MCP server
  2. Inject your active Umbraco Base URL mapping and your Management/Delivery Bearer Auth Token
  3. Engage directly your preferred AI model (Claude or Cursor) asking it to index pages natively

Eliminate messy IDE switching while structuring robust Headless projects. Let the AI prototype content inside Umbraco dynamically utilizing REST interfaces.

Who is this for?

  • CMS Platform Engineers — automate routine headless content extractions requesting schema structures from the LLM avoiding UI switch delays
  • Digital Editors — prompt the agent to bulk-view lists of raw content items skipping complicated table drill-downs natively
  • Frontend Architects — rapidly confirm internal content API stability extracting pure JSON payloads mapping to visual interfaces directly
cmscontent-deliverymedia-managementasp-netbackoffice-automationstructured-content

10 tools expose this connector's capabilities to your AI agent.

create_cms_document

Provide the document data as a JSON object adhering to the schema. Creates a new document in the Umbraco CMS

delete_cms_document

This action is irreversible. Permanently deletes a document from the Umbraco CMS

get_delivery_content_by_id

Retrieves a specific content item by its GUID or numeric ID via Delivery API

get_delivery_content_by_path

g., "/home/about"). Retrieves a specific content item by its URL path

get_management_document

Retrieves a specific document via the Umbraco Management API (Drafts/Backoffice)

list_delivery_content

Supports pagination via take and skip. Lists content available via the Umbraco Delivery API

list_document_types

Lists all document types (schemas) defined in Umbraco

list_media_assets

Lists media assets (images, files) from the Umbraco Media library

query_delivery_content

g., "contentType:blogPost"). Filters content items using the Umbraco Delivery API query syntax

update_cms_document

Provide the document ID and JSON updates. Updates fields of an existing document in Umbraco

See how to talk to your AI agent using Umbraco.

Use the delivery API to get the content from '/products/new-feature' and list out its properties.

I've pulled the content assigned to that path. It's a 'ProductPage' schema. Key properties include 'title' (New Feature Launch!) and strongly typed 'descriptionText'. Everything mapped perfectly fine over the Headless engine.

Look up our Document Types to see the exact schema required for a 'BlogPost'. Then create one JSON draft placeholder payload based on it.

A 'BlogPost' document requires 3 fields: `authorName`, `publishDate`, and `bodyText`. I have inherently crafted the JSON payload for this structure matching your schema accurately so we can push it cleanly up the line.

Delete the backoffice document holding ID d6ef43...

Done. The document d6ef43... has been successfully nuked from the active Umbraco Management context.

Absolutely. Instruct the agent to 'get_delivery_content_by_path' and supply something like '/home/about'. It traverses the Delivery API native routing map and returns perfectly all inner block grids and content variables attached.

Related Connectors