// Set up the Python backtesting environment. Detects OS, creates virtual environment, installs dependencies (openalgo, ta-lib, vectorbt, plotly), and creates the backtesting folder structure.
| name | setup |
| description | Set up the Python backtesting environment. Detects OS, creates virtual environment, installs dependencies (openalgo, ta-lib, vectorbt, plotly), and creates the backtesting folder structure. |
| argument-hint | [python-version] |
| allowed-tools | Bash, Read, Write, Glob, AskUserQuestion |
Set up the complete Python backtesting environment for VectorBT + OpenAlgo.
$0 = Python version (optional, default: python3). Examples: python3.12, python3.13Run the following to detect the OS:
uname -s 2>/dev/null || echo "Windows"
Map the result:
Darwin = macOSLinux = LinuxMINGW* or CYGWIN* or Windows = WindowsPrint the detected OS to the user.
Create a Python virtual environment in the current working directory:
macOS / Linux:
python3 -m venv venv
source venv/bin/activate
pip install --upgrade pip
Windows:
python -m venv venv
venv\Scripts\activate
pip install --upgrade pip
If the user specified a Python version argument, use that instead of python3:
$PYTHON_VERSION -m venv venv
TA-Lib requires a C library installed at the OS level BEFORE pip install ta-lib.
macOS:
brew install ta-lib
Linux (Debian/Ubuntu):
sudo apt-get update
sudo apt-get install -y build-essential wget
wget http://prdownloads.sourceforge.net/ta-lib/ta-lib-0.4.0-src.tar.gz
tar -xzf ta-lib-0.4.0-src.tar.gz
cd ta-lib/
./configure --prefix=/usr
make
sudo make install
cd ..
rm -rf ta-lib ta-lib-0.4.0-src.tar.gz
Linux (RHEL/CentOS/Fedora):
sudo yum groupinstall -y "Development Tools"
wget http://prdownloads.sourceforge.net/ta-lib/ta-lib-0.4.0-src.tar.gz
tar -xzf ta-lib-0.4.0-src.tar.gz
cd ta-lib/
./configure --prefix=/usr
make
sudo make install
cd ..
rm -rf ta-lib ta-lib-0.4.0-src.tar.gz
Windows:
pip install ta-lib
If that fails, download the appropriate .whl file from https://github.com/cgohlke/talib-build/releases and install with:
pip install TA_Lib-0.4.32-cp312-cp312-win_amd64.whl
Install all required packages (latest versions):
pip install openalgo vectorbt plotly anywidget nbformat ta-lib pandas numpy yfinance python-dotenv tqdm scipy numba nbformat ipywidgets quantstats ccxt duckdb psutil
Create only the top-level backtesting directory. Strategy subfolders are created on-demand when a backtest script is generated (by the /backtest skill).
mkdir -p backtesting
Do NOT pre-create strategy subfolders.
6a. Check if .env.sample exists at the project root. If it does, use it as a template.
6b. Ask the user which markets they will be backtesting using AskUserQuestion:
6c. If the user selected Indian Markets, ask for their OpenAlgo API key:
.env6d. If the user selected Indian Markets (DuckDB), ask for the DuckDB database path:
market_data table with symbol, exchange, interval, timestamp columns, it is OpenAlgo Historify format (store as HISTORIFY_DB_PATH). Otherwise store as DUCKDB_PATH.6e. If the user selected Crypto Markets, ask if they want to configure exchange API keys:
.env.env6f. Write the .env file in the project root directory. Use this template, filling in any keys/paths the user provided:
# Indian Markets (OpenAlgo)
OPENALGO_API_KEY={user_provided_key or "your_openalgo_api_key_here"}
OPENALGO_HOST=http://127.0.0.1:5000
# DuckDB Data Sources (direct database loading - fastest)
# Custom DuckDB (user-created with OHLCV table)
DUCKDB_PATH={user_provided_path or ""}
# OpenAlgo Historify DuckDB (market_data table with epoch timestamps)
HISTORIFY_DB_PATH={user_provided_path or ""}
# Crypto Markets (CCXT) - Optional
CRYPTO_API_KEY={user_provided_key or ""}
CRYPTO_SECRET_KEY={user_provided_key or ""}
6g. Add .env to .gitignore if it exists (never commit secrets):
Scripts use find_dotenv() to automatically walk up and find the single root .env, so no copies are needed in subdirectories.
grep -qxF '.env' .gitignore 2>/dev/null || echo '.env' >> .gitignore
Run a quick verification:
python -c "
import vectorbt as vbt
import openalgo
import plotly
import talib
import duckdb
import anywidget
import nbformat
import quantstats as qs
from dotenv import load_dotenv
print('All packages installed successfully')
print(f' vectorbt: {vbt.__version__}')
print(f' plotly: {plotly.__version__}')
print(f' duckdb: {duckdb.__version__}')
print(f' nbformat: {nbformat.__version__}')
print(f' quantstats: {qs.__version__}')
print(f' TA-Lib: available')
print(f' python-dotenv: available')
"
If TA-Lib import fails, inform the user that the C library needs to be installed first (see Step 3).
Print a summary showing:
/backtest).env file status (configured with keys / placeholder) — single file at project rootcp .env.sample .env and fill in API keys if you skipped configuration"brew install ta-lib.env files — they contain secrets. Always use .gitignore..env — do not ask them to edit the file manuallypython-dotenv is included in the pip install and must be used by all scripts to load .envQuick backtest a strategy on a symbol. Creates a complete .py script with data fetch, signals, backtest, stats, and plots.
Optimize strategy parameters using VectorBT. Tests parameter combinations and generates heatmaps.
Quickly fetch data and print key backtest stats for a symbol with a default EMA crossover strategy. No file creation needed - runs inline in a notebook cell or prints to console.
Compare multiple strategies or directions (long vs short vs both) on the same symbol. Generates side-by-side stats table.
VectorBT backtesting expert. Use when user asks to backtest strategies, create entry/exit signals, analyze portfolio performance, optimize parameters, fetch historical data, use VectorBT/vectorbt, compare strategies, position sizing, equity curves, drawdown charts, or trade analysis. Also triggers for openalgo.ta helpers (exrem, crossover, crossunder, flip, donchian, supertrend).