Eddi
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.
What is Eddi?
Eddi is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to 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, gdp...
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.
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Config-driven engine that turns JSON into production-grade A
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx eddiConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Eddi
EDDI (Execution-Driven Dialogue Interface) is a config-driven, enterprise-grade multi-agent orchestration middleware that turns JSON configuration files into production AI agents without writing code. Built on Quarkus and Java, it supports 12+ LLM providers, the MCP and A2A protocols, retrieval-augmented generation, persistent cross-session memory, and compliance infrastructure for GDPR, HIPAA, and the EU AI Act. Teams use EDDI to deploy customer-facing chatbots, internal knowledge assistants, and complex multi-agent pipelines that need an immutable audit trail and enterprise security controls.
Prerequisites
- Docker and Docker Compose (required for the recommended deployment path)
- Java 21+ and Maven (for building from source)
- A supported LLM API key (OpenAI, Anthropic, Google, or any OpenAI-compatible endpoint)
- At least 4 GB RAM for the default Docker stack; 8 GB recommended for the full stack with PostgreSQL and Keycloak
Run the one-command installer
The installer script detects your platform, generates encryption keys, and starts EDDI with MongoDB via Docker Compose. On Windows, use the PowerShell script instead.
# Linux / macOS / WSL2
curl -fsSL https://raw.githubusercontent.com/labsai/EDDI/main/install.sh | bash
# Windows (PowerShell)
Invoke-WebRequest -UseBasicParsing -Uri "https://raw.githubusercontent.com/labsai/EDDI/main/install.ps1" -OutFile "install.ps1"
.\install.ps1Verify EDDI is running
EDDI starts on port 7070 by default. Open the Swagger UI to confirm the API is live and explore available endpoints.
# Check health
curl http://localhost:7070/q/health
# Open Swagger UI
open http://localhost:7070/q/swagger-uiDeploy a full stack with PostgreSQL and authentication (optional)
For production deployments, use the extended Docker Compose profiles to add PostgreSQL and Keycloak authentication.
# PostgreSQL backend
EDDI_DATASTORE_TYPE=postgres docker compose -f docker-compose.yml -f docker-compose.postgres.yml up
# Add Keycloak OAuth2 auth
docker compose -f docker-compose.yml -f docker-compose.auth.yml upCreate an agent via JSON configuration
EDDI agents are defined entirely in JSON. POST a configuration to the REST API to instantiate a new agent with behavior, memory, and tool access defined declaratively.
curl -X POST http://localhost:7070/agentstore/agents \
-H "Content-Type: application/json" \
-d '{
"name": "support-agent",
"llmProvider": "openai",
"model": "gpt-4o",
"systemPrompt": "You are a helpful support agent.",
"tools": ["web_search", "user_memory"]
}'Integrate EDDI as an MCP server
EDDI exposes an MCP-compatible endpoint that AI clients can connect to, allowing external agents to delegate tasks to EDDI-managed sub-agents over the MCP protocol.
{
"mcpServers": {
"eddi": {
"command": "npx",
"args": ["eddi"],
"env": {
"EDDI_HOST": "http://localhost:7070"
}
}
}
}Eddi Examples
Client configuration (Claude Desktop)
Connect Claude Desktop to a locally running EDDI instance as an MCP server.
{
"mcpServers": {
"eddi": {
"command": "npx",
"args": ["eddi"],
"env": {
"EDDI_HOST": "http://localhost:7070"
}
}
}
}Prompts to try
Example prompts once EDDI is running and connected.
- "Create a new EDDI agent that searches the web and summarizes results with persistent memory"
- "List all agents currently deployed in EDDI"
- "Start a Round Table debate between three agents on the best database for a SaaS product"
- "Query the user memory store for facts about user ID 42"
- "Show me the GDPR audit trail for the last 24 hours of agent activity"Troubleshooting Eddi
Docker Compose fails to start EDDI — port 7070 already in use
Change the host port mapping in docker-compose.yml (e.g., '8080:7070') or stop the conflicting process. Set the quarkus.http.port environment variable if you need to change the internal port as well.
Agent API calls return 401 Unauthorized after enabling Keycloak
With the auth profile enabled, all API requests require a Bearer token obtained from Keycloak. Use the Keycloak admin console (default http://localhost:8180) to create a client and user, then obtain a token via the /protocol/openid-connect/token endpoint. Pass it as Authorization: Bearer <token> in API calls.
MongoDB connection errors on startup
EDDI requires MongoDB to be ready before the application starts. If you see connection errors immediately after docker compose up, MongoDB may still be initializing. Add a health-check depends_on condition in docker-compose.yml or simply wait 20-30 seconds and restart the EDDI container.
Frequently Asked Questions about Eddi
What is Eddi?
Eddi is a Model Context Protocol (MCP) server that 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. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Eddi?
Follow the installation instructions on the Eddi GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Eddi?
Eddi works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Eddi free to use?
Yes, Eddi is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Eddi Alternatives — Similar Coding Agents Servers
Looking for alternatives to Eddi? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Eddi in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Eddi?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.