Practical overview
When it fits
Use Eddi when a conversational application needs configurable agent behavior, multi-agent orchestration, or a middleware layer connecting agents, model providers and business systems.
Who it is for
- Developers building configurable conversational AI applications.
- Platform operators deploying and managing AI agents with Docker or Kubernetes.
- Teams coordinating multiple agents and connecting conversational workflows to business systems.
Documented uses
- Define agent behavior through versioned JSON configurations.
- Compose and orchestrate multi-agent conversational workflows.
- Connect agents to supported language model providers and APIs.
- Add retrieval-augmented generation and persistent conversation memory.
- Integrate agent workflows with MCP and A2A.
Requirements & limits
The repository is public and lists an Apache-2.0 license. The quick-start installer requires Docker.
- Agent behavior is defined in JSON rather than application code, and the lifecycle pipeline is configurable.
- The project documents integrations with multiple LLM providers, MCP, A2A, the OpenAI Chat Completions API, Slack, OpenAPI and OAuth 2.0.
- The quick start sets up Eddi and a database through Docker Compose; Kubernetes and OpenShift deployment paths are also documented.
- The repository identifies Java and Quarkus as the implementation stack.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Install Eddi with Docker Compose
Set up Eddi and a database using the documented Linux, macOS or WSL2 installer.
Before you start
- Docker
- Linux, macOS or WSL2
Steps
Run the installer command from the Eddi repository: curl -fsSL https://raw.githubusercontent.com/labsai/EDDI/main/install.sh | bash.
Follow the installer wizard to set up Eddi and a database.
Open http://localhost:7070 to reach the Manager and Workforce chooser, or use /chat for the chat interface.
Use the Manager or form-based agent wizard to create an initial AI agent.
Expected outcome
Eddi and its database are set up through Docker Compose, with the Manager and chat interfaces available locally.
Guide limitations
- The guide documents installation and initial agent creation, not configuration or execution of a particular agent workflow.
- Docker is required.
Compare options for Agent orchestration
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| Eddi | Define agent behavior through versioned JSON configurations. | Docker, Self-hosted, Server | The repository is public and lists an Apache-2.0 license. The quick-start installer requires Docker. |
| Agenite | Define typed tools with JSON Schema validation and error handling. | CLI | |
| Veldra | Create an agent with a policy, tools, skills, knowledge base and reasoning method. | Self-hosted | The documented deployment is self-hosted and uses a sign-in gate after a first-run administrator account is created. |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 13 October 2026
- Original website title
- labsai/EDDI · Official homepage title · retained in its source language