mstr-advisor

MSTR Monitor

Small, transparent stock monitor for MSTR. One page, live price marker, explainable scoring, and clear cadences.

What it does

• Shows price with MA20/MA60 and a 1‑minute live marker
• Calculates daily/weekly/monthly scores with per‑term contributions
• Displays trade plan (entry zone, stop, targets)
• News list with lightweight sentiment
• Update cadence card so users know what’s fresh

How data updates

• Price: data/public/hot.json every minute (GitHub Actions → hotdata branch). Frontend polls raw file.
• Snapshot/Features: built from daily OHLCV + cross‑asset; by default refreshed EOD (can run hourly if wanted).
• Weekly ML + backtests: heavier jobs; kept to weekends.

Workflows

• .github/workflows/pages.yml – builds frontend/ + data/public/ and deploys to Pages on pushes to main.
• .github/workflows/hot-data.yml – writes data/public/hot.json every minute to the hotdata branch.

Branches:

• main – static site (Pages).
• hotdata – tiny live artifacts (only hot.json).

Local preview

python -m http.server 8080
# open http://localhost:8080/frontend/index.html

Deploy

1) Push to main → Pages workflow publishes. 2) In repo Settings → Pages → Source: GitHub Actions. 3) Run the “hot-data” workflow once (workflow_dispatch) to seed hotdata.

Stack

• Frontend: static HTML/CSS/Chart.js (no framework)
• Backend scripts: Python + yfinance
• Hosting: GitHub Pages; Live data: raw file from hotdata