BeeAI

Overview

BeeAI is an open-source platform that enables discovering, running, and composing AI agents across different frameworks and programming languages. As part of the Linux Foundation AI & Data program, BeeAI aims to create a unified ecosystem for AI agent development and deployment.

Key Features

• Framework Agnostic: Seamlessly integrate agents built with different frameworks and languages
• Agent Composition: Create complex multi-agent workflows and systems
• Agent Discoverability: Search and discover agents through an integrated catalog
• First-Class Ecosystem Support: Built with Python and TypeScript as primary languages
• Open Standards: Promotes interoperability across the AI agent ecosystem

System Requirements

• CPU: 4+ cores recommended
• RAM: 8GB+ recommended
• GPU: Optional, but beneficial for running local models
• Storage: 2GB+ for the platform

Installation Guide

Prerequisites

• Basic knowledge of command line interfaces
• Git installed on your system
• Docker and Docker Compose (recommended for easy setup)

Option 1: Homebrew Installation (macOS/Linux)

1. Install BeeAI using Homebrew:

   brew install beeai
   

2. Initialize and start BeeAI:

   beeai init
   beeai start
   

3. Access the application:

   Open your browser and navigate to http://localhost:8080

Option 2: Docker Installation

1. Clone the repository:

   git clone https://github.com/i-am-bee/beeai-platform.git
   

2. Navigate to the project directory:

   cd beeai-platform
   

3. Start the Docker containers:

   docker-compose up -d
   

4. Access the application:

   Open your browser and navigate to http://localhost:8080

Option 3: Manual Installation

1. Clone the repository:

   git clone https://github.com/i-am-bee/beeai-platform.git
   

2. Navigate to the project directory:

   cd beeai-platform
   

3. Install dependencies:

   # For Python components
   pip install -e .
   
   # For TypeScript components
   npm install
   

4. Build and run the application:

   npm run build
   npm start
   

5. Access the application:

   Open your browser and navigate to http://localhost:8080

Note: For detailed installation instructions specific to your operating system and environment, please refer to the official documentation at docs.beeai.dev.

Practical Exercise: Creating a Multi-Agent Workflow

Now that you have BeeAI installed, let’s create a simple multi-agent workflow to demonstrate its capabilities.

Step 1: Initialize Your Project

Create a new project directory and initialize it:

mkdir my-beeai-project
cd my-beeai-project
beeai init project

Step 2: Define Your Agents

Create a configuration file that defines multiple agents and their interactions:

# agents.yaml
agents:
  - name: researcher
    type: llm
    model: gpt-4
    description: "Research agent that finds information"

  - name: writer
    type: llm
    model: claude-3-sonnet
    description: "Writing agent that creates content"

  - name: reviewer
    type: llm
    model: llama3
    description: "Reviewer agent that checks for accuracy"

workflow:
  - from: researcher
    to: writer
    condition: "research_complete"

  - from: writer
    to: reviewer
    condition: "draft_complete"

Step 3: Run Your Workflow

Start your multi-agent workflow:

beeai run workflow --config agents.yaml --input "Create a summary of quantum computing advances in 2023"

Step 4: Explore Advanced Features

Once you’re comfortable with the basics, explore some of BeeAI’s more advanced features:

• Create custom agent types
• Integrate external tools and APIs
• Develop cross-language agent communication
• Use the search catalog to discover community-built agents

Resources

Official Documentation

The official documentation is the best place to find detailed information about BeeAI.

Read the Documentation

Community Support

Join the community to get help, share your experiences, and contribute to the project.

• GitHub Discussions
• Discord Community

Contributing

BeeAI is an open-source project and welcomes contributions from the community:

• Contributing Guide
• Code of Conduct