Modern AI tools often fail on complex Drupal architectures due to context loss and copy-paste friction. This live demonstration showcases how terminal-native AI agents operate directly inside local development containers to plan, build, test, and refactor Drupal 11 features seamlessly.

Attendees will witness real-time agentic execution using ddev-claude-code inside a live DDEV environment. The session covers setting up governance files (CLAUDE.md), delegating Drush commands to agents, leveraging Model Context Protocol (MCP) servers, and enforcing security and PSR-4 coding standards automatically in agent loops.

Run-of-Show (45 Minutes):

• 00:00–00:10 | Governance & Setup: Boot ddev-claude-code and review an enterprise CLAUDE.md specification with strict security and architectural boundaries.
• 00:10–00:25 | Agentic Module Construction: Execute terminal agent commands (claude -p) to scaffold a custom module, perform entity inspections, and run Drush configuration exports directly inside DDEV.
• 00:25–00:35 | TDD & Self-Correction Loops: Direct Claude Code to generate PHPUnit tests, execute them via terminal, and observe a self-healing loop as the agent parses failures and refactors code.
• 00:35–00:45 | Architecture Best Practices & Q&A: Discuss token/cost optimization, API security, and access to the open-source demo repository.

Key Takeaways:

1. How to configure CLAUDE.md and ddev-claude-code for safe, context-aware local development.
2. Techniques for automating module scaffolding, Drush operations, and PHPUnit testing via CLI agents.
3. Operational frameworks for keeping human-in-the-loop oversight on agent-generated code.