This repository contains examples of using OpenAI's AI models to create agent-based systems for the construction industry. These examples demonstrate how to build AI assistants that can understand and address construction-specific queries.

• Python 3.9 or higher
• An OpenAI API key

1. Clone this repository:

git clone https://github.com/yourusername/openai-construction-agents.git
cd openai-construction-agents

2. Install the required dependencies:

pip install requests python-dotenv

3. Create a .env file in the root directory and add your OpenAI API key:

OPENAI_API_KEY=your_api_key_here

This example demonstrates a basic agent architecture that uses domain specialization to route construction queries to the appropriate expert agent:

• Planning Specialist
• Materials Specialist
• Safety Specialist

To run this example:

python simple_agent.py

This example demonstrates a more advanced multi-agent system with explicit handoffs between agents, conversation memory, and specialized construction knowledge domains.

To run this example:

python multi_agent.py

This is a simple "hello world" example that demonstrates the basic usage of OpenAI's API for a construction-focused assistant.

To run this example:

python hello_agent.py

• Domain-specific expertise in construction planning, materials, and safety
• Agent-to-agent handoffs for specialized knowledge
• Conversation context maintenance
• Construction industry-focused instruction sets

• For best results, be specific in your construction-related queries
• The agents maintain conversation history, so you can refer to previous questions
• Type 'exit' to quit any of the example applications
• In the multi-agent system, type 'reset' to start a new conversation

These examples implement patterns similar to those in OpenAI's Agents SDK but using the direct OpenAI API. They demonstrate:

1. Agent specialization (different domains of expertise)
2. Agent orchestration (routing queries to the right specialist)
3. Agent handoffs (transferring control between agents)
4. System message customization (construction-specific instructions)

For production use, consider exploring OpenAI's official Agents SDK for more advanced features.

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