Phil Whittaker discusses Umbraco’s MCP platform and software factory, designed to enable partners and clients to create agentic AI tools for a more agent-ready CMS ecosystem. The Base MCP SDK and create-umbraco-mcp-server streamline development, allowing for local, hosted, and CLI MCP servers, enhancing integration and usability within the Umbraco environment.
Shannon Thompson Deminick discusses leveraging GitHub Next's Agentic Workflows to automate maintenance for Examine, a search library for Umbraco. The AI agents improved performance and efficiency, triaged issues, and facilitated releases, resulting in significant speed and memory allocation enhancements. This approach streamlined development, allowing for measurable, focused improvements.
Matt Brailsford discusses the integration of AI into content editing through Umbraco AI, emphasizing the need for a safe space for experimentation and a clear definition of quality. He argues that providing editors with the ability to branch content and check standards could enhance creativity and consistency, ultimately improving their workflow.
2026-07-16 from 14:00 to 20:00 (Europe/London) - Method4, 12-14 Trade Street, Cardiff, CF10 5DT, Cardiff, GB
Alan Ballard describes his experience setting up an MCP server to connect Claude with Umbraco, revealing it to be simpler than expected. He outlines the setup process, including configuring a JSON file and creating an API user. The integration allows Claude to assist in content creation and management, enhancing productivity and consistency.
Phil Whittaker and Bolette Kern from Umbraco HQ present the "Umbraco in AI" framework, detailing how the Umbraco Management API integrates with AI agents. They discuss Local and Hosted MCPs, a CLI tool, and the Base MCP SDK, enabling developers to streamline AI integration and enhance package readiness with minimal effort.
In this session, Umbraco HQ discusses integrating AI agents into Umbraco's framework, focusing on the development of a base MCP server and a chained MCP architecture. The aim is to enhance collaboration between AI and developers, streamline content management, and reduce friction for seamless interaction with Umbraco through shared best practices.
AI-powered schema mapping for SchemeWeaver using Umbraco.AI
Plug Azure AI Search into Umbraco Search as a fully-featured provider: filtering, faceting, sorting, culture/segment variants, and protected content out of the box.
Mats Persson discusses the recent rebranding in the enterprise digital experience platform (DXP) sector, emphasizing the risks of proprietary AI ecosystems. He advocates for a stable, open architecture that prioritizes true enterprise maturity, focusing on enhancing productivity, maintaining flexibility, and ensuring robust AI governance, rather than succumbing to legacy vendor hype.
Mark McDonald reflects on his virtual experience of Codegarden, highlighting Shannon Deminick's talk on AI in Umbraco site development. Inspired, he created a proof of concept for re-platforming existing sites using AI to automate workflows. The project demonstrated reusable processes, though it emphasized the need for human oversight and clear component frameworks.
Matt Brailsford, lead developer on the Umbraco AI project at Umbraco HQ, discusses integrating AI into the Umbraco back office. He showcases a podcasting demo, illustrating AI's capabilities in transcription, summarization, and guest validation. Brailsford emphasizes Umbraco AI's extensibility for custom solutions, encouraging community engagement and experimentation.
In the talk by Umbraco HQ, Polette discusses the integration of AI into Umbraco CMS, emphasizing a strategic approach that prioritizes choice, control, and governance. She explores how Umbraco is adapting to an unpredictable future, the evolving role of content management, and upcoming features aimed at enhancing editor collaboration.
Connects the Umbraco backoffice AI features (prompts, agents, Copilot, embeddings) to any OpenAI-compatible self-hosted server such as Ollama, LM Studio or vLLM — no cloud account required.
AI Agent Memory for Umbraco — persistent memory + learning layer for Umbraco's AI agent stack.