In this session we’ll look at how the CAP MCP development server gives AI coding assistants like Claude Code or GitHub Copilot direct access to CAP documentation, enabling them to generate code that actually follows CAP conventions. We’ll also explore agent skills that infuse best practices, guide multi-step workflows, and provide troubleshooting context so your agent doesn’t have to rediscover every detail from scratch. We’ll show practical examples of what works well today and where the boundaries still are.
Daniel Schlachter
Johannes Vogt
Lars Plessing