Stock Scanner Bot

A daily post-market technical analysis scanner for US stocks and ETFs. Connects to Interactive Brokers (IB Gateway / TWS), evaluates a watchlist against technical filters with confirmation signals, scores each symbol, and exports results to CSV and Excel.

Project Structure

stock_scanner/
├── src/
│   └── scanner.py          # Main scanner logic
├── tests/
│   └── test_scanner.py     # Unit tests (49 tests)
├── config/
│   └── watchlist.csv       # Symbol watchlist
├── docs/
│   └── Scan_Bot_Spec_Detailed_18-03-2026.pdf
├── output/                  # Generated scan results (gitignored)
├── requirements.txt
└── README.md

Features

• Loads watchlist from CSV, JSON, or TXT
• Fetches daily and weekly OHLCV data via IBKR
• Pivot-point support/resistance detection with clustering (not simple min/max)
• 6 mandatory filters including at least one confirmation signal
• Score from 0–11 based on 7 technical criteria (including volatility)
• RSI(14), 52-week high/low distances, ATR, volatility flag
• Timestamped CSV + XLSX + log on every run

Filters (all must pass)

#	Filter	Condition
1	Weekly Trend	Close > Weekly MA200 and MA200 trending upward
2	Liquidity	Avg Dollar Volume (20d) >= $20M
3	Amplitude	(Resistance - Support) / Support >= 10%
4	Support Proximity	(Close - Support) / Support <= 5%
5	Resistance Distance	(Resistance - Close) / Close >= 5%
6	Confirmation Signal	At least one of: rebound pattern, volume >= 1.5x avg, or two strong green candles

Scoring Model (0–11)

Criterion	Points
Weekly trend confirmed	0 or 2
Support proximity <= 2% / <= 5%	2 / 1 / 0
Amplitude >= 20% / >= 10%	2 / 1 / 0
Rebound / bullish rejection pattern	0 or 2
Volume confirmation (>= 1.5x 20d avg)	0 or 1
Liquidity (avg dollar vol >= $20M)	0 or 1
Volatility flag (ATR% >= 2.0)	0 or 1

Output Columns

Column	Description
Symbol	Ticker
Close	Last daily close
Support	Nearest pivot-low support zone
Resistance	Nearest pivot-high resistance zone
SupportDistance	(Close - Support) / Support
ResistanceDistance	(Resistance - Close) / Close
Amplitude	(Resistance - Support) / Support
Score	0–11
CandidateFlag	TRUE / FALSE
WeeklyMA200	200-week moving average
PriceAboveMA200	Boolean
WeeksAboveMA200	Consecutive weeks above MA200
MA200Slope	UP / DOWN
RSI14	14-day RSI (Wilder's)
Dist52WkHigh	% distance from 52-week high
Dist52WkLow	% distance from 52-week low
TwoStrongGreenCandles	Last 2 candles green with body >= median
ReboundPattern	Hammer / bullish rejection detected
LatestVolume	Most recent session volume
AvgDailyVolume	20-day average volume
VolumeConfirmed	Latest volume >= 1.5x 20d average
ATR14	14-day Average True Range
ATR_Pct	ATR as % of close
VolatilityFlag	ATR% >= 2.0
AvgDollarVolume	20-day average dollar volume
LiquidityFlag	AvgDollarVolume >= $20M
Comment	Summary of pass/fail reasons

Setup

git clone https://github.com/NadirAliOfficial/stock-scanner-bot.git
cd stock-scanner-bot

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Usage

1. Edit config/watchlist.csv with your symbols (one per row)
2. Make sure IB Gateway or TWS is running with API enabled
3. Run the scanner:

python src/scanner.py

4. Results are saved to output/:

output/
  scan_20260320_2155.csv
  scan_20260320_2155.xlsx
  scan_20260320_2155.log

CLI Options

python src/scanner.py --host 127.0.0.1 --port 4002 --client-id 3 --watchlist config/watchlist.csv --output output

Configuration

All thresholds are defined at the top of src/scanner.py:

SR_PERIOD             = 252          # Support/resistance lookback (trading days)
MIN_AVG_DOLLAR_VOLUME = 20_000_000   # $20M
MIN_AMPLITUDE         = 0.10         # 10%
MAX_SUPPORT_DISTANCE  = 0.05         # 5%
MIN_RESISTANCE_DIST   = 0.05         # 5%
VOLUME_CONFIRM_MULT   = 1.5          # 1.5x average volume
VOLATILITY_ATR_PCT_THRESH = 2.0      # 2% ATR threshold

IBKR Connection

Setting	Default	Notes
Host	127.0.0.1	localhost
Port	4002	IB Gateway paper (4001 live, 7496 TWS live, 7497 TWS paper)
Client ID	3	Must be unique across connected bots

Testing

pip install pytest
python -m pytest tests/ -v

Notes

• Only fully closed candles are used (daily and weekly)
• Symbols with insufficient history are skipped and logged
• The bot performs analysis only — no trade execution
• Foreign symbols require corresponding IBKR market data subscriptions