Strategy Lab

Write a strategy as code in ISL (the Intelo Strategy Language) and check it instantly.

How this page works

Author strategies in ISL, the typed deterministic Intelo Strategy Language. Compilation runs in your browser with the same compiler the engine uses.

ISL document

Typed, declarative, deterministic. The compiler validates references, rejects cycles and unbounded periods, and derives the exact lookback.

{
  "isl": "1.0",
  "name": "EMA crossover with ATR stop",
  "description": "Enter when the fast EMA crosses above the slow EMA and RSI is not overbought; exit on the reverse cross or risk overlay.",
  "timeframe": "1h",
  "universe": [
    {
      "asset": "0x0000000000000000000000000000000000000000",
      "symbol": "ASSET"
    }
  ],
  "features": {
    "close": {
      "kind": "price",
      "field": "close"
    },
    "ema_fast": {
      "kind": "ema",
      "input": "close",
      "period": 12
    },
    "ema_slow": {
      "kind": "ema",
      "input": "close",
      "period": 48
    },
    "rsi_14": {
      "kind": "rsi",
      "input": "close",
      "period": 14
    },
    "vol_20": {
      "kind": "realised_vol",
      "input": "close",
      "period": 20
    }
  },
  "signals": {
    "golden_cross": {
      "op": "cross_above",
      "a": {
        "op": "ref",
        "feature": "ema_fast"
      },
      "b": {
        "op": "ref",
        "feature": "ema_slow"
      }
    },
    "death_cross": {
      "op": "cross_below",
      "a": {
        "op": "ref",
        "feature": "ema_fast"
      },
      "b": {
        "op": "ref",
        "feature": "ema_slow"
      }
    },
    "not_overbought": {
      "op": "<",
      "left": {
        "op": "ref",
        "feature": "rsi_14"
      },
      "right": {
        "op": "const",
        "value": 70
      }
    },
    "entry": {
      "op": "and",
      "args": [
        {
          "op": "signal",
          "name": "golden_cross"
        },
        {
          "op": "signal",
          "name": "not_overbought"
        }
      ]
    }
  },
  "rules": [
    {
      "id": "enter",
      "when": "entry",
      "action": "enter_long",
      "sizing": {
        "type": "vol_target",
        "targetVolBps": 1500,
        "volFeature": "vol_20",
        "maxFractionBps": 5000,
        "kellyCapBps": 2500
      },
      "cooldownBars": 4
    },
    {
      "id": "exit",
      "when": "death_cross",
      "action": "exit_long",
      "cooldownBars": 0
    }
  ],
  "risk": {
    "stopLossBps": 400,
    "trailingStopBps": 600,
    "maxBarsInTrade": 240
  }
}

Compilation

valid

Plan hash: 8f0c972e08255cd8309ae1d18e1db6062fac279411b7d1eb2fc63d1ffa07338f
Timeframe · lookback: 1h · 145 bars
Feature order: close → ema_fast → ema_slow → rsi_14 → vol_20
Signal order: golden_cross → death_cross → not_overbought → entry
Rules: enter:enter_long, exit:exit_long

Save to your library

Stored with its compile hash. Saving a document with the same name creates a new version; earlier versions stay as they were.

Strategies run in the Backtesting Studio (BACKTEST) and its paper trader (PAPER). In a backtest, the zero asset address means every Pons launch in the fetched history; put a token address there to test one token. The live agent trades on its own rule set and limits, not on ISL documents.