Skip to content

About

AI-powered stock signals with Buy, Hold, Sell probabilities and severity scores using Olostep and TypeSafe Jev.

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Latest commit

 

History

9 Commits

Folders and files

Repository files navigation

StockSignal

AI stock signals, right in your terminal.

Analyze up to 5 stocks with yfinance, the Olostep Python SDK and the TypeSafe Python SDK (Jev). Get color-coded BUY / HOLD / SELL decisions, probability breakdowns, market impact and news links for a 5-trading-day horizon.

Warning

For educational and experimental use only. This project does not provide financial advice. AI signals can be wrong, and confidence scores do not predict investment returns. Do not use its BUY / HOLD / SELL signals to make real trading or investment decisions.

StockSignal terminal dashboard — fictional demo values

Quick start

Requires uv and Python 3.11–3.13 (Olostep SDK compatibility).

uv sync

Copy .env.example to .env and add your keys (see Get API keys below):

OLOSTEP_API_KEY=your_key_here
JEV_API_KEY=your_key_here
uv run app

Enter 1–5 tickers, then press R. A add · D delete · ↑/↓ select · Esc focus stocks · R refresh · Q quit.

Try the dashboard without keys: uv run app --demo.

Get API keys

  • Olostep: Sign up with Google or email, then generate a key under API Keys. Save it as OLOSTEP_API_KEY in .env. The free trial includes 500 successful requests, with no credit card required.
  • TypeSafe Jev: Sign up with Google or email, then create a key under API Keys. Save it as JEV_API_KEY in .env. Check trial credits and billing in the console; API usage may require payment.

yfinance needs no key. Keep .env private and restart the app after changing keys.

How it works

flowchart LR
    T[Tickers] --> M[yfinance · market data]
    T --> N[Olostep · recent news]
    M --> J[TypeSafe Jev · choice + score]
    N --> J
    J --> D[Terminal dashboard]
Loading

Market impact (1–5) rates the significance of recent developments, from Very Low to Critical. Confidence is the selected signal's model-estimated preference. These preferences are not calibrated probabilities of future returns.

About

AI-powered stock signals with Buy, Hold, Sell probabilities and severity scores using Olostep and TypeSafe Jev.

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Contributors

Languages