> ## Documentation Index
> Fetch the complete documentation index at: https://docs.joinmobius.com/llms.txt
> Use this file to discover all available pages before exploring further.

# How the Mobius AI strategy builder interprets you

> The Mobius strategy builder turns plain-English descriptions into executable trading strategies using an AI that reads market data bar by bar.

The Mobius strategy builder lets you describe a trading idea the way you'd explain it to another trader — in plain English — and turns it into a fully executable strategy. No code required. The AI parses your intent, asks targeted clarifying questions, and extracts the exact parameters and indicators needed to backtest and deploy your strategy.

## How the AI understands your strategy

Describe your strategy in the Mobius chat interface. You can be as high-level or as specific as you like.

**Equity examples:**

* `"Buy AAPL when the 9-EMA crosses above the 21-EMA and price is above the 200-EMA. Sell when the 9-EMA crosses back below."`
* `"Go long SPY when RSI drops below 30 and close the position when RSI reaches 60."`
* `"Buy QQQ when the VIX falls more than 5% in a day."`

**Crypto examples:**

* `"Buy BTC/USD when the daily RSI is below 35. Sell when RSI exceeds 65."`
* `"Go long ETH when price crosses above its 50-day SMA. Exit when it drops 8% from entry."`
* `"Buy DOGE when WSB mentions spike above 500 per day and sentiment is positive."`

**Multi-asset and cross-asset examples:**

* `"Rotate between SPY and GLD based on the 200-day SMA of each. Hold whichever is above its moving average."`
* `"Buy NVDA when congressional trading activity shows a senator purchase in the last 30 days."`

After you describe your strategy, the AI asks one clarifying question at a time to fill in missing details — the asset to trade, the timeframe, stop-loss levels, take-profit targets, and position sizing. Once it has everything it needs, it locks in a structured strategy definition ready for backtesting.

<Tip>
  Include entry rules, exit rules, and a stop-loss in your first message. If you provide all details upfront, the AI skips clarifying questions entirely.
</Tip>

## Strategy components

Every Mobius strategy is built from four components that the AI populates from your description.

<AccordionGroup>
  <Accordion title="Entry conditions">
    The rules that trigger a buy order — indicator crossovers, threshold breaches, sentiment signals, or any combination. Example: *"Buy when 9-EMA crosses above 21-EMA and price is above 200-EMA."*
  </Accordion>

  <Accordion title="Exit conditions">
    The rules that close a position — take-profit targets, stop-loss levels, indicator reversals, or time-based exits. Example: *"Sell when RSI exceeds 70 or price drops 5% below entry."*
  </Accordion>

  <Accordion title="Assets">
    The ticker or tickers to trade — a single stock, a crypto pair, a multi-stock portfolio, or a pair for spread trading. See [Tradeable assets](/data/tradeable-assets) for the full list.
  </Accordion>

  <Accordion title="Position sizing">
    How much capital to allocate per trade — a fixed dollar amount, a percentage of portfolio, or a dynamic rule based on volatility. Example: *"Use 10% of portfolio per position, max 5 concurrent positions."*
  </Accordion>
</AccordionGroup>

## Supported strategy types

| Type                  | Description                                | Example                               |
| --------------------- | ------------------------------------------ | ------------------------------------- |
| Single asset          | Trade one ticker on its own signals        | Buy AAPL on RSI                       |
| Multi-asset portfolio | Apply rules across a basket of tickers     | Scan S\&P 500 for EMA crossovers      |
| Pairs trading         | Long one asset, short another              | Long MSFT / Short META on spread      |
| Conditional switching | Hold different assets based on conditions  | SPY when VIX \< 20, GLD when VIX > 30 |
| Cross-asset signals   | Use one ticker to signal trades on another | Buy QQQ when TLT is rising            |

## Visual block editor

After the AI parses your strategy, switch to the **visual block editor** to fine-tune individual parameters — thresholds, indicator periods, position sizes — without rewriting your description. Any changes you make in the block editor are reflected immediately in the strategy definition and carry through to the next backtest.

See [Visual editor](/guides/visual-editor) for details.

## Same logic in backtest and live

The same AI decision engine used during backtesting runs during live trading. There is no discrepancy between what you test and what executes on your brokerage account — every bar, the AI receives the same inputs and applies the same logic.