iStorm PoC — Retail Agent (Streamlit)

This repository contains a minimal PoC for an iStorm retail agent using local mock data.

Files

• products.json, accessories.json, inventory.json — mock data
• agent.py — PoC agent logic (recommend, format_output)
• demo_runner.py — demo scenarios runner
• streamlit_app.py — Streamlit UI to demo the agent
• requirements.txt — required packages

Run locally (Windows / PowerShell)

1. Open PowerShell and go to the folder with these files, e.g.:

   cd C:\Users\polyzoidoud\Downloads\istorm_mock_data

2. (Optional but recommended) Create a virtual environment:

   python -m venv venv
   .\\venv\\Scripts\\Activate.ps1

3. Install requirements:

   pip install -r requirements.txt

4. Run Streamlit:

   streamlit run streamlit_app.py

5. The app will open in your browser (usually at http://localhost:8501).

Deploy to Streamlit Community Cloud

1. Push this folder to GitHub (create a repo).
2. In Streamlit Cloud, click "New app" and connect your GitHub repo & branch.
3. Set the main file to streamlit_app.py and deploy.

Notes

• This PoC uses mock data. Replace stub functions in agent.py with real API calls when ready.
• For production, add logging, authentication, rate limiting, and safety checks.