🔍 Cassidy Copilot

AI-powered workflow debugging assistant built with Streamlit and Groq.

Features

• Workflow Analysis: Upload or paste JSON workflows for AI-powered diagnosis
• Issue Detection: Identifies structural, logical, and best practice issues
• Actionable Recommendations: Provides clear, prioritized fixes
• Clean Architecture: Separated UI, logic, and prompts for maintainability

Setup

1. Install dependencies:

   pip install -r requirements.txt

2. Configure API key:

   • Create a .env file in the project root
   • Add your Groq API key:

     GROQ_API_KEY=your_api_key_here
     

   • Get your API key from: https://console.groq.com/

3. Run the app:

   streamlit run app.py

Project Structure

cassidy-copilot/
│
├── app.py                  # Streamlit UI (UI only)
├── utils.py                # LLM logic + diagnosis engine (logic only)
├── prompts.py              # System + analysis prompts (prompt engineering)
├── requirements.txt        # Python dependencies
├── .env                    # API keys (never commit)
└── README.md               # This file

Design Principles

• app.py: UI only - handles user interaction and display
• utils.py: Logic only - core diagnosis functionality
• prompts.py: Prompt engineering isolated - easy to iterate on prompts

Usage

1. Open the app in your browser
2. Describe your workflow and the issue you're experiencing
3. Click "Diagnose Issue" to get AI-powered analysis
4. Review the root cause, category, recommended fix, and escalation status

Requirements

• Python 3.10+
• Groq API key (get one at https://console.groq.com/)
• Streamlit
• See requirements.txt for full dependencies

License

MIT