BYOAI with Swarm Integration

Introduction

This project integrates the BYOAI (Bring Your Own AI) framework with OpenAI's Swarm, creating a powerful, decentralized, and context-driven agent system. The integration provides a flexible and scalable solution for distributed task execution and AI-driven workflows.

Features

• Decentralized agent architecture
• Context-driven task distribution
• Specialization-based task assignment
• Priority queue for efficient task management
• RESTful API for task submission and status monitoring
• Dynamic agent scaling based on workload
• Detailed logging and monitoring
• Task redistribution for load balancing

Installation

1. Clone the repository:

   git clone https://github.com/your-repo/byoai-swarm.git
   cd byoai-swarm
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Set up environment variables:

   export CONTEXT_WORKFLOW_DIR=workflows
   export CONTEXT_AGENT_PORT=8099
   export CONTEXT_AGENT_HOST=0.0.0.0
   

Usage

1. Start the BYOAI-Swarm system:

   python byoai-script.py
   

2. Access the API endpoints:

   • Home: http://localhost:8099/
   • Status: http://localhost:8099/status
   • Swarm State: http://localhost:8099/swarm/state
   • Add Task: POST http://localhost:8099/swarm/add_task
   • Task Status: GET http://localhost:8099/swarm/task_status/<task_id>
   • Agent Load: GET http://localhost:8099/swarm/agent_load
   • Redistribute Tasks: POST http://localhost:8099/swarm/redistribute_tasks
   • Swarm Statistics: GET http://localhost:8099/swarm/statistics
   • Scale Agents: POST http://localhost:8099/agents/scale

3. Create and place workflow YAML files in the workflows directory. The system will automatically load and execute these workflows.

API Usage Examples

Adding a Task

curl -X POST http://localhost:8099/swarm/add_task \
  -H "Content-Type: application/json" \
  -d '{"description": "Complex math calculation", "priority": 8, "specialization": "math"}'

Getting Swarm State

curl http://localhost:8099/swarm/state

Checking Task Status

curl http://localhost:8099/swarm/task_status/1

Redistributing Tasks

curl -X POST http://localhost:8099/swarm/redistribute_tasks \
  -H "Content-Type: application/json" \
  -d '{"threshold": 5}'

Getting Swarm Statistics

curl http://localhost:8099/swarm/statistics

Scaling Agents

curl -X POST http://localhost:8099/agents/scale \
  -H "Content-Type: application/json" \
  -d '{"num_agents": 2}'

Architecture

The integrated BYOAI-Swarm system consists of the following main components:

1. BYOAI Script (byoai-script.py): The main entry point that initializes the system, loads workflows, and starts the Flask server.

2. Swarm Integration (swarm_integration.py): Implements the core Swarm functionality, including task management, agent specializations, and decentralized task execution.

3. Workflows: YAML files defining task sequences, priorities, and specializations.

4. Flask API: Provides endpoints for system interaction and monitoring.

5. Dynamic Agent Scaling: Automatically scales the number of agents based on the current workload.

Workflow Structure

Workflows are defined in YAML files with the following structure:

name: Sample Workflow
steps:
  - name: Task Name
    action: action_to_perform
    priority: 5
    specialization: math

Error Handling

The API provides clear error messages for various scenarios:

• Invalid JSON payload
• Missing required fields
• Task not found
• Internal server errors

Error responses include a JSON object with an "error" key containing a descriptive message.

Logging and Monitoring

The system provides detailed logging for all operations, including:

• Task addition and execution
• Agent registration and task assignment
• Workflow loading and execution
• API requests and responses
• Dynamic agent scaling

Logs can be found in the console output when running the application.

Contributing

1. Fork the repository
2. Create a feature branch: git checkout -b feature-name
3. Commit your changes: git commit -am 'Add some feature'
4. Push to the branch: git push origin feature-name
5. Submit a pull request

License

This project is licensed under the MIT License - see the LICENSE file for details.