Martingale42/nautilus-dev / skills/nt-learn
nt-learn
FlaggedUse when learning NautilusTrader from scratch or deepening understanding. Provides a structured curriculum from installation to building custom NT components in Python and Rust.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 1.6 KB |
| curriculum/01-setup.md | markdown | 5.1 KB |
| curriculum/02-run-examples.md | markdown | 4.3 KB |
| curriculum/03-foundations.md | markdown | 5.7 KB |
| curriculum/04-first-strategy.md | markdown | 6.5 KB |
| curriculum/05-backtesting.md | markdown | 7.3 KB |
| curriculum/06-indicators-actors.md | markdown | 7.4 KB |
| curriculum/07-live-trading.md | markdown | 6.3 KB |
| curriculum/08-rust-internals.md | markdown | 6.9 KB |
| curriculum/09-full-rust-trading.md | markdown | 9.1 KB |
| curriculum/10-building-nt.md | markdown | 7.3 KB |
curriculum/04-first-strategy.md
View on GitHub# Stage 04: First Strategy
## Goal
Write a trading strategy from scratch that submits orders, manages positions, and uses indicators.
## Prerequisites
- Stage 03 complete (understand architecture, data model, cache, messaging)
## Concepts
### Strategy Lifecycle
A `Strategy` extends `Actor` with order management. The lifecycle is:
```
__init__() → on_start() → [on_bar / on_quote_tick / on_trade_tick / ...] → on_stop()
```
**Critical rule**: Do NOT use `self.clock`, `self.log`, or `self.cache` in `__init__()`. These are injected later during registration. Use `on_start()` for initialization.
### StrategyConfig
Strategies use a frozen config object for parameters:
```python
from nautilus_trader.trading import Strategy, StrategyConfig
class MyConfig(StrategyConfig, frozen=True):
instrument_id: InstrumentId
bar_type: BarType
ema_period: int = 20
class MyStrategy(Strategy):
def __init__(self, config: MyConfig) -> None:
super().__init__(config)
self.instrument_id = config.instrument_id
self.bar_type = config.bar_type
self.ema_period = config.ema_period
```
### Handler Hierarchy
When an order event occurs, handlers fire in sequence (most specific → most general):
1. `on_order_filled(event)` — specific handler
2. `on_order_event(event)` — catch-all order handler
3. `on_event(event)` — catch-all event handler
Same for positions: `on_position_opened` → `on_position_event` → `on_event`.
## Building a Simple EMA Cross Strategy
### Step 1: Define Config
```python
from nautilus_trader.model import InstrumentId
from nautilus_trader.model.data import BarType
from nautilus_trader.trading import StrategyConfig
class EMACrossConfig(StrategyConfig, frozen=True):
instrument_id: InstrumentId
bar_type: BarType
fast_period: int = 10
slow_period: int = 20
trade_size: float = 1.0
```
### Step 2: Write the Strategy
```python
from nautilus_trader.indicators.average.ema import ExponentialMovingAverage
from nautilus_trader.model.enums import OrderSide
from nautilus_trader.trading import Strategy
class EMACross(Strategy):
def __init__(self, config: EMACrossConfig) -> None:
super().__init__(config)
# Store config values (OK in __init__)
self.instrument_id = config.instrument_id
self.bar_type = config.bar_type
self.trade_size = config.trade_size
# Create indicators (OK in __init__ — they don't need system access)
self.fast_ema = ExponentialMovingAverage(config.fast_period)
self.slow_ema = ExponentialMovingAverage(config.slow_period)
def on_start(self) -> None:
# Load instrument from cache
self.instrument = self.cache.instrument(self.instrument_id)
# Register indicators for automatic bar updates
self.register_indicator_for_bars(self.bar_type, self.fast_ema)
self.register_indicator_for_bars(self.bar_type, self.slow_ema)
# Request historical data (goes to on_historical_data, NOT on_bar)
self.request_bars(self.bar_type)
# Subscribe to live/streaming data (goes to on_bar)
self.subscribe_bars(self.bar_type)
def on_bar(self, bar) -> None:
# Wait for both EMAs to be initialized
if not self.fast_ema.initialized or not self.slow_ema.initialized:
return
# Check if flat (no position)
if self.portfolio.is_flat(self.instrument_id):
if self.fast_ema.value > self.slow_ema.value:
self._buy()
elif self.fast_ema.value < self.slow_ema.value:
self._sell()
def _buy(self) -> None:
order = self.order_factory.market(
instrument_id=self.instrument_id,
order_side=OrderSide.BUY,
quantity=self.instrument.make_qty(self.trade_size),
)
self.submit_order(order)
def _sell(self) -> None:
order = self.order_factory.market(
instrument_id=self.instrument_id,
order_side=OrderSide.SELL,
quantity=self.instrument.make_qty(self.trade_size),
)
self.submit_order(order)
def on_stop(self) -> None:
# Flatten all positions on stop
self.close_all_positions(self.instrument_id)
```
### Step 3: Key Patterns
1. **`instrument.make_qty()` / `instrument.make_price()`** — Always use these. The RiskEngine rejects orders with wrong precision.
2. **`register_indicator_for_bars()`** — Indicators auto-update before `on_bar()` is called. No manual `indicator.update()` needed.
3. **`request_bars()` vs `subscribe_bars()`** — Request gets historical data (for warmup). Subscribe gets live/streaming data. Both are typically called in `on_start()`.
4. **`self.order_factory`** — Built-in factory for creating orders. Handles ID generation. Multiple strategy instances must use unique `order_id_tag` values.
5. **`self.portfolio.is_flat()`** — Check position state through Portfolio, not by tracking your own boolean.
## Order Types
```python
# Market order
order = self.order_factory.market(instrument_id, OrderSide.BUY, quantity)
# Limit order
order = self.order_factory.limit(instrument_id, OrderSide.BUY, quantity, price)
# Stop market
order = self.order_factory.stop_market(instrument_id, OrderSide.SELL, quantity, trigger_price)
# Bracket order (entry + take-profit + stop-loss)
bracket = self.order_factory.bracket(
instrument_id=instrument_id,
order_side=OrderSide.BUY,
quantity=quantity,
entry_trigger_price=entry_price, # or entry_price for limit entry
sl_trigger_price=stop_price,
tp_price=take_profit_price,
)
self.submit_order_list(bracket)
```
## Exercises
1. **Build the EMA cross strategy** above and run it with `example_01`'s engine setup pattern, using test data.
2. **Add position tracking**: Override `on_position_opened()` and `on_position_closed()` to log position details.
3. **Add a bracket order variant**: Instead of a plain market order, submit a bracket order with take-profit and stop-loss.
4. **Try multiple order_id_tags**: Run two instances of your strategy with different parameters and unique `order_id_tag` values.
## Checkpoint
You're ready for Stage 05 when:
- [ ] You can write a strategy that submits orders and reacts to fills
- [ ] You understand `on_start()` vs `__init__()` initialization
- [ ] You know why `instrument.make_price()` / `instrument.make_qty()` are required
- [ ] You can register indicators and check `indicator.initialized` before using values
- [ ] You understand the difference between `request_bars()` (historical) and `subscribe_bars()` (live)
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation). | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |