This project demonstrates an interactive chatbot built using Chainlit for the user interface and the OpenAI Agents SDK for the conversational logic. The chatbot responds to user queries and automatically saves the complete chat history in a structured chat_history.json file when the session ends.
- 🧱 Chainlit – UI and session management
- 🤖 OpenAI Agents SDK – Agent logic and completions
- 🐍 Python 3.10+
- ⚡ uv – Ultra-fast Python dependency manager
✅ Browser-based chatbot interface
✅ Handles multi-turn conversations
✅ Maintains full user + assistant history
✅ Saves history as chat_history.json on session end
📁 chatbot_project/ ├── agents.py ├── my_secrets.py ├── main.py ├── chat_history.json └── README.md
git clone https://github.com/MuhammadUsmanGM/Chainlit-bot cd chatbot_project 2. Install Dependencies bash Copy code uv add chainlit openai 3. Configure API Keys Create a my_secrets.py file with your OpenAI or Gemini-compatible credentials:
class Secrets: gemini_api_key = "your-api-key" gemini_base_url = "https://generativelanguage.googleapis.com/v1beta/openai/" gemini_api_model = "gemini-2.0-flash" # or another model 📝 Adjust variable names if you're using other Models instead of Gemini.
- Run the Chatbot chainlit run main.py Then open the local URL (e.g., http://localhost:8000) in your browser.
💾 Chat History At the end of every session, the complete chat is saved as:
[ { "role": "user", "content": "Hello, who are you?" }, { "role": "assistant", "content": "I'm your AI assistant. How can I help you today?" } ] The file is saved as chat_history.json in the root directory.
