riskkit

A framework-agnostic risk-management toolkit for systematic traders.

Most open-source trading tools focus on the fun part — signals, indicators, backtesting engines. They leave the part that decides whether you survive thin: how big a position to take, where stops live, when to cut size, and when to stop trading altogether. That's what blows up retail algo traders, not a bad entry signal.

riskkit is that missing layer. Every component is pure Python with no dependency on any exchange, data provider, or backtesting framework. You feed it numbers; it hands back auditable decisions you can drop into backtesting.py, vectorbt, freqtrade, or your own loop.

Not financial advice

riskkit helps you implement a risk policy you have chosen. It cannot make a losing strategy profitable. Test everything on paper first.

Install

pip install riskkit-quant           # imports as `riskkit`
# with dynamic correlation (adds pandas):
pip install "riskkit-quant[pandas]"

Note

The PyPI distribution is named riskkit-quant (the bare name was too close to an existing package), but you import riskkit as usual.

The six components

Component	What it does
PositionSizer	Volatility-adjusted sizing with a half-Kelly ceiling and a reduction ladder.
DrawdownManager	Drawdown tier ladder: cut size → raise the bar → halt, with a recovery ramp.
StopEngine	Composable per-position stop stack; tightest stop wins, stops only tighten.
CorrelationGuard	One open position per correlation group (static + dynamic).
SessionManager	Daily caps, cooldowns, and tilt detection.
PreTradeValidator	The composable final gate that vetoes a trade if any rule fails.

Want all six working together? The RiskManager façade wires them from one config and turns a single TradeIntent into a sized, validated decision — see Quickstart.