Enterprise Multi-Agent Team

Overview

The Enterprise Multi-Agent Team is a collaborative AI agent framework that coordinates multiple specialized agents to handle complex content creation tasks. Built on LangGraph for workflow orchestration and A2A Protocol for agent-to-agent communication.

LLM Agnostic Key Features

Getting Started

Installation

git clone https://github.com/AGenNext/AGenNext-Registry.git
cd AGenNext-Registry
pip install -r requirements.txt

Environment Variables

# Create .env file
cp .env.example .env

# Edit with your API keys
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...
GOOGLE_API_KEY=...

Quick Start

# View all agent cards
python main.py --mode cards

# Single agent demo
python main.py --mode demo --agent openai --topic "AI Writing"

# Multi-agent team demo
python main.py --mode team --topic "Cloud Computing"

# A2A protocol demo
python main.py --mode a2a

Run the Server

# Start A2A server
python main.py --mode server --port 8000

API Reference

Full OpenAPI specification available in openapi.yaml

Base URL

Local:     http://localhost:8000
Production: https://your-domain.com

Endpoints

Health Check

curl http://localhost:8000/health

Run Agent Task

curl -X POST http://localhost:8000/agents/openai/run \
  -H "Content-Type: application/json" \
  -d '{"task": "Write a blog post", "content_type": "blog"}'

Run Team Workflow

curl -X POST http://localhost:8000/team/run \
  -d '{"task": "Create a project", "strategy": "parallel"}'

Agents

Each agent is specialized for different content types and integrates with specific SDKs.

LLM Agnostic Providers

Docker

Build Image

docker build -t content-writers .

Run Container

docker run -d -p 8000:8000 \
  -e OPENAI_API_KEY=sk-... \
  content-writers

Use Pre-built

docker run -p 8000:8000 ghcr.io/agennext/agennext-registry:latest

Source Code

For detailed implementation, see the source files.