SKILL.md
Generate a quick inline backtest and print stats. Do NOT create a file - output code directly for the user to run or execute in a notebook.
Arguments
$0= symbol (e.g., SBIN, RELIANCE). Default: SBIN$1= exchange. Default: NSE$2= interval. Default: D
Instructions
Generate a single code block the user can paste into a Jupyter cell or run as a script. The code must:
- Fetch data from OpenAlgo (or DuckDB if user provides a DB path, or yfinance as fallback)
- Use OpenAlgo ta for EMA 10/20 crossover by default (never VectorBT built-in); only use TA-Lib if the user explicitly says "talib"/"TA-Lib"
- Clean signals with
ta.exrem()(always.fillna(False)before exrem) - Use Indian delivery fees:
fees=0.00111, fixed_fees=20 - Fetch NIFTY benchmark via OpenAlgo (
symbol="NIFTY", exchange="NSE_INDEX") - Print a compact results summary:
Symbol: SBIN | Exchange: NSE | Interval: D
Strategy: EMA 10/20 Crossover
Period: 2023-01-01 to 2026-02-27
Fees: Delivery Equity (0.111% + Rs 20/order)
-------------------------------------------
Total Return: 45.23%
Sharpe Ratio: 1.45
Sortino Ratio: 2.01
Max Drawdown: -12.34%
Win Rate: 42.5%
Profit Factor: 1.67
Total Trades: 28
-------------------------------------------
Benchmark (NIFTY): 32.10%
Alpha: +13.13%- Explain key metrics in plain language for normal traders
- Show equity curve plot using Plotly (
template="plotly_dark")
Example Usage
/quick-stats RELIANCE /quick-stats HDFCBANK NSE 1h