CMS in the post-AI Era: from MCP to Vibecoding
summary
Claims Drupal can both run agents and be run by agents with a practical path from prototype to production, backed by a working MCP server plus provider plugins for models like Claude and Gemini. Data shows a sharp uptick in vibecoding via Next.js since 2023 and hundreds of thousands of sites shifting to “none” this year. Demonstrations include Claude doing content modeling through MCP, vibe-coded prototypes that iterate fast but fail on editability and governance, and the recommended workflow: AI prototype, Drupal for scale.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 3:59AI research assistant in web consumption
- 7:49AI hype cycle expectations
- 10:15CMS role post-AI
- 13:54Vibecoding limitations
- 18:48Drupal’s AI opportunity
- 20:55AI agents in content workflows
- 23:24Operationalizing AI in CMS
- 24:27Summary
related talks
01h 28mAI Coding Agents and how to code them
33mFrontend after AI: The new UX
32mProgramming language evolution in the AI era
21mBuilding Agents (the hard parts!)
22mMCP UI: Extending the frontier
28mThe Agentic Developer: Orchestrating AI Workflows with MCPs
30mComposable AI: Building Next-Gen AI Agents with MCP
36mPutting AI Agents to Work: A New Era of Open Source Connectivity with MCP