EDDI

View on GitHub

Config-driven engine that turns JSON into production-grade AI agents. Multi-agent orchestration, 12+ LLM providers, MCP/A2A protocols, RAG, persistent memory, and enterprise compliance (EU AI Act, GDPR, HIPAA). Built on Quarkus.

★ 377128 forksJavaApache-2.0labsai

Java/Quarkus middleware that turns versioned JSON configs into multi-agent conversational AI systems. Adds routing, persistent memory, RAG, MCP/A2A protocols, audit ledger and GDPR/HIPAA/EU AI Act controls. Self-hostable via Docker with 12+ LLM providers.

Use Cases

Deploy config-driven AI agents from JSON without codeMulti-agent orchestration with intelligent intent routingRAG over enterprise knowledge basesCross-session persistent user memoryMulti-agent debate and peer review workflowsHuman-in-the-loop approval flowsSlack and chat channel agent deploymentConnect agents to tools via MCP client/server and A2AScheduled and cron-triggered agent executionEU AI Act, GDPR and HIPAA compliant AI deploymentsImmutable audit ledger for agent actionsSelf-hosted local LLM agents via OllamaTenant quotas, cost budgets and rate limitingGroup conversations between multiple agents

Built With

Language: Java
Frameworks: Quarkus · LangChain4j · Docker · Kubernetes · MongoDB · PostgreSQL · Keycloak · Prometheus · Grafana · NATS JetStream · Chroma · Ollama · OpenAPI · OAuth 2.0 · Slack

EDDI

Use Cases

Built With

Tags