An AI-powered logic architect that transforms text into professional, stylized workflow diagrams instantly.
- 📂 Instant Visualization — Type your logic and see a real-time flowchart update.
- 🧠 Smart Categorization — Automatically assigns shapes (Diamonds for logic, Rounded for data).
- 🎨 Premium Dark UI — Sleek Glassmorphism design with a modern SaaS aesthetic.
↔️ Dual Orientation — Switch between Horizontal and Vertical layouts.
| Layer | Technology | Purpose |
|---|---|---|
| Frontend | Streamlit | Turning Python scripts into an interactive web interface. |
| Styling | CSS3 / HTML5 | Custom Glassmorphism UI and SaaS aesthetic. |
| Core Logic | Python 3 | String parsing and automated shape categorization. |
| Visualization | Mermaid.js | JavaScript-based diagramming and charting tool. |
AI-Workflow-Builder/
├── app.py # Main application logic & Streamlit entry point.
├── style.py # Modular CSS definitions for the Professional UI.
├── requirements.txt # List of external libraries required for execution.
├── .gitignore # Rules for Git to ignore sensitive/cache files.
├── README.md # Technical documentation and project overview.
└── workflow_log.txt # Local storage for generated workflow history.
- Python v3.9 or higher
- Streamlit for the web interface
# 1. Clone the repository
git clone [https://github.com/tanusha-19/AI-Workflow-Builder.git](https://github.com/tanusha-19/AI-Workflow-Builder.git)
cd AI-Workflow-Builder
# 2. Install dependencies
pip install -r requirements.txt
# 3. Set up your environment variables
# Create a .env file in the root directorystreamlit run app.py
- Environment Isolation: Uses a
.gitignorefile to ensure that local configuration files (like.env) are never pushed to the public repository. - Input Sanitization: The logic cleans and formats user text (replacing
/with<br/>) to prevent diagram breakage. - Modular Architecture: By separating the CSS (
style.py) from the logic (app.py), the project follows the Single Responsibility Principle (SRP). - Local Logic Fallback: The system is designed to work locally without requiring heavy cloud dependencies, ensuring data privacy for the user's logic.