Chatbot is a Python-based agentic AI framework that combines large language models with practical agent implementations for real-world scenarios.
The project addresses the need for intelligent automation in tasks like order processing and data extraction. It provides a Shopping Agent that uses browser automation, screenshot recognition, and large model analysis to extract and summarize order information from web interfaces that lack direct export functionality. The system combines FastAPI for backend services with Streamlit for the frontend, allowing users to control workflows through natural language instructions while the agent handles screenshot capture, information extraction, and Excel export.
The Shopping Agent suits teams managing orders across multiple systems without native export capabilities. It works best for scenarios where users need to process order screenshots in batch, extract structured data, and generate reports. The tool requires Python 3.8 or later, Google Chrome version 140 or higher, and appropriate API credentials for the language model being used. Configuration is handled through environment variables and a settings panel, covering server address, model parameters, and browser paths.
Development activity shows ongoing expansion of the project's scope. The codebase has grown to include both the original chatbot functionality and the new Shopping Agent feature set, indicating active feature development. The project maintains documentation in Chinese with detailed setup instructions, troubleshooting guidance, and configuration options, suggesting attention to user onboarding. The inclusion of specific version requirements and browser compatibility notes reflects practical experience addressing real deployment issues.
