Write the strategy. Prove it. Then let it trade.

A marketplace and execution engine for automated strategies on NSE, BSE and NFO. Subscribe to a strategy or build your own, put it through historical and live testing, then run it on the broker account you already have.

strategy.algoscriptwhat you write
strategy "RSI Mean Reversion" {  asset: NIFTY.NFO  timeframe: 15m  stop_loss: 1.5%  take_profit: 3%  sizing: percent_equity(5%)   when RSI(NIFTY, 14) < 30    and price(NIFTY) > EMA(NIFTY, 20) {    buy NIFTY.NFO  }   when RSI(NIFTY, 14) > 70 {    sell NIFTY.NFO  }}
Execution enginewhat runs

A signal, when the conditions hold

buy NIFTY strength 0.8
RSI(14) crossed below 30 while price held above EMA(20)

Sized against your equity, not the author's

  • 5% of your account, converted to whole lots
  • Stop at 1.5%, target at 3%, attached to the parent order

Checked before anything is placed

  • Within max position size and leverage
  • Daily loss limit and drawdown cap not breached
  • Market open for NFO, contract not expired
The same engine runs this in backtest, on paper against live ticks, and live on your broker. A strategy that behaves differently between them is a bug, not a surprise.

Nothing goes live until it has survived the two stages before it.

Backtest, paper and live are the same engine in three modes. That is the whole reason to trust the progression: results carry forward because the code that produced them does not change.

  1. 1

    Backtest

    Replay it over history

    Run against stored candles for the instruments and timeframes you care about. Fills are simulated with slippage and commission, so the result is not a curve drawn through perfect prices.

  2. 2

    Paper

    Then against live ticks, with nothing at stake

    The same strategy subscribes to real market data and trades a simulated account. This is where a strategy that only worked on tidy historical bars falls over.

  3. 3

    Live

    Only then, your broker

    Connect your own account and the engine places real orders through it. Positions, funds and history stay with your broker; nothing is pooled and no capital passes through us.

If you trade by a rule, build or sell it here—without writing code.

Start from one of eleven prebuilt strategy families and edit its numbers on a plain-English Rule Sheet, describe a new rule in chat, or write AlgoScript directly. Keep it private for your own broker, or publish it to the marketplace and set how it earns.

Publishing runs a quality check first. The strategy has to compile, resolve every NSE or NFO instrument it references, and clear a scored backtest review after brokerage before it can be listed.

Three views of the same strategy, plus one quality gate

Plain-English Rule Sheet
Start from any of 11 prebuilt strategy families across conservative, balanced, and aggressive presets. Every strategy reads back as plain sentences where every number, percentage, and time is editable in place.
Described in Chat
Explain the rule you want in plain words. The assistant restates the rules first, compiles them into a verified strategy, and hands you a Rule Sheet to refine.
AlgoScript & JS style
Prefer writing code? Our declarative DSL and JS-like dialect express multi-stock baskets, NIFTY regime switches, and option spreads (Iron Condor, Credit Spread, IV percentile) in a few lines.
Scored quality gate
Every edit is checked in real time for NSE/NFO symbol resolution, execution support, and a multi-year rehearsal with real brokerage and STT before it can be listed.

Your broker, your account

Connect the account you already trade with. Orders are placed through your broker’s own API, and you can revoke access from either side at any time.

  • Zerodha
  • Angel One
  • Dhan
  • Groww
  • Upstox

Start with a backtest. It costs nothing to be sure.

Pick one of our eleven strategy families, change any number on the Rule Sheet, and run a historical rehearsal with real NSE charges—or paper trade a verified winner on live ticks.