Show HN: EDDI – Multi-agent AI engine where agent logic lives in JSON, not code (github.com)

E.D.D.I (Enhanced Dialog Driven Interface) has been announced as a versatile multi-agent orchestration middleware designed for conversational AI, enabling the configuration of agent logic through JSON rather than traditional coding. Built with Java 25 and Quarkus, EDDI streamlines the deployment process with a one-command installer and features such as intelligent routing and persistent memory. Its integration with APIs like OAuth 2.0 and Slack, along with its support for protocols like MCP (Model Context Protocol) and A2A (Agent-to-Agent), positions it as an innovative solution that safeguards operational governance in production environments. This announcement is significant for the AI/ML community as it challenges the norm of using Python and Node.js for similar frameworks, offering a more robust, deterministic engine capable of handling concurrent processes with true OS-level parallelism. EDDI emphasizes security, with no dynamic code execution and built-in compliance features for regulations like GDPR and HIPAA. The use of versioned JSON configurations allows for seamless updates without downtime, making EDDI a promising tool for developers seeking efficiency and regulatory compliance in deploying complex AI systems.