TSX Canadian Stock Screener + Auto Entry Pipeline

Python tools for screening Canadian (TSX) stocks and running a daily swing-trading “entry pipeline” that tracks candidates across days and emits actionable alerts when setups move through a signal state machine.

This repo contains two primary scripts:

• canadian_stock_screener.py — multi-factor momentum screener that ranks a predefined TSX universe and saves a top-N CSV.
• auto_pipeline.py — consumes daily screener outputs, tracks tickers over time, detects technical entry patterns, and produces alerts + a persistent signal database.

Data source: Yahoo Finance via yfinance.

---

Repo layout

.
├── canadian_stock_screener.py
├── auto_pipeline.py
├── config.py
├── requirements.txt
└── data/
    └── can_tickers        # TSX universe list (one ticker per line)

---

1) Canadian stock screener

What it does

canadian_stock_screener.py downloads ~2 years of daily data and scores each ticker with a weighted stack (see CONFIG["weights"] in the script):

• Weinstein Stage II alignment
• Relative Strength vs benchmark (XIU.TO)
• MACD momentum
• OBV slope (volume accumulation)
• ADX trend strength
• Volatility-adjusted momentum (VAM)
• 52-week high proximity / breakout pressure

It also applies basic filters such as minimum price and minimum average volume.

Output

• Prints a table to the console.
• Saves the top picks to:
  • out/screener_outputs/YYYYMMDD_HHMM.csv

Run

python canadian_stock_screener.py

Configure

Edit CONFIG inside canadian_stock_screener.py:

• top_n (default 10)
• min_price (default 2.0 CAD)
• min_avg_volume (default 100,000)
• weights (factor weighting)
• lookback_days (default 504 trading days)

Universe list:

• data/can_tickers

---

2) Automated entry pipeline

What it does

auto_pipeline.py reads daily screener outputs, tracks tickers across days, detects entry patterns, and alerts on signal state transitions.

Accepted inputs (flexible):

• Any CSV with a Ticker or symbol column
• Example names (from the script header):
  • top10_tsx_20260218_1430.csv
  • my_tickers_2026-02-18.csv
• JSON is also supported for some upstream tools (see auto_pipeline.py header).

Patterns detected

The pipeline currently runs three detectors:

1. VCP (Volatility Contraction Pattern)
2. EMA pullback reclaim (EMA21 and EMA50 variants)
3. Base breakout (tight range + volume confirmation)

Signal state machine

FORMING  →  AT_PIVOT  →  CONFIRMED  →  ACTIVE / FAILED

Alert priority

• 🔴 URGENT — state jumped to CONFIRMED today (often “enter next session open” style)
• 🟡 WATCH — state reached AT_PIVOT today (often “place buy-stop” / watch volume)
• 🟢 FORMING — setup building, check again
• ⚫ EXPIRED — invalidated/expired

Directory layout (auto-created)

On first run, the pipeline creates:

<base_dir>/
  out/
    screener_outputs/      # drop daily top-N CSVs here
    signal_db/
      signal_history.csv   # persistent signal state across days (auto-managed)
    alerts/
      alerts_YYYYMMDD.csv  # actionable alerts for each run
      report_YYYYMMDD.txt  # human-readable daily report

3) Operational services and schedule

This repository is run as three operational services:

• main.py

  • Builds swing universe, runs screener, runs pipeline, and sends report.
  • Typical schedule: once daily after market close.

• virtual_buy.py

  • Executes virtual buys from the candidate queue and updates funds/positions files.
  • Typical schedule: once daily after the pipeline/report step.

• position_monitor.py

  • Monitors open positions and can execute virtual sells in pre-close mode.
  • Typical schedule:
    • Pre-close run for actionable intraday exits.
    • Post-close run for informational end-of-day monitoring.

---

Installation

Requirements

• Python 3.10+ recommended
• Internet access (Yahoo Finance via yfinance)

Install deps

python -m venv .venv
# Linux/macOS:
source .venv/bin/activate
# Windows:
# .venv\Scripts\activate

pip install -r requirements.txt

---

Notes / limitations

• yfinance depends on Yahoo endpoints; intermittent rate limits or missing data can happen.
• The screener is designed around TSX tickers and uses XIU.TO as the benchmark.
• The pipeline caps tracked tickers (--max-tickers, default 40) to avoid excessive API calls.

---

Disclaimer

This repository is for research/education and does not constitute financial advice. Trading involves risk.

---

License

• MIT