A comprehensive AI-powered suite built for developers, leveraging Google Gemini Pro & Flash.
This application serves as an "All-in-One" AI workspace. It combines advanced prompt engineering, web scraping, and computer vision into a single Streamlit interface, allowing users to generate full-stack code, analyze live websites, and replicate UI designs instantly.
- Vibe Coder: Transforms abstract ideas into production-ready code structures.
- System Protocol Builder: Generates complex "God-Mode" system prompts for other AI agents.
- Strategies Included: CO-STAR, Chain of Thought, and Custom Persona generation.
- Recursive Crawling: Scans up to 50 pages deep to map site architecture.
- Asset Extraction: Automatically detects and lists fonts, icons, and image assets.
- RAG Analysis: Chat directly with the scraped content to extract insights or summary data.
- UI-to-Code: Upload any interface screenshot and get the exact frontend code (React, Tailwind, HTML/CSS).
- Style Matching: Detects the "vibe" and color palette of the uploaded image.
- Frontend: Streamlit
- AI Model: Google Gemini 1.5 Pro & Flash (via
google-generativeai) - Web Scraping: BeautifulSoup4, Requests
- Image Processing: PIL (Pillow)
- Data Handling: JSON (Local History)
Follow these steps to run the tool locally:
git clone [https://github.com/iammagdy/wise-prompt-ai.git](https://github.com/iammagdy/wise-prompt-ai.git)
cd wise-prompt-ai