Guide

No-Code Crypto Trading Bot: From One Sentence to Paper

Build a crypto trading bot with no coding: say the idea in one sentence, let Claude or ChatGPT write and backtest it, read the grade, then run it on paper.

Updated

You can build a crypto trading bot with no coding by describing the strategy in one sentence to Claude or ChatGPT. The assistant writes the code, OptAlgo backtests it on real exchange data and grades how honest the result is, and you run it as a paper bot. This guide follows one sentence all the way to paper, and says plainly what no-code does not do for you.

Everything below runs on paper, with no real money. Trading is risky, and a backtest is never proof that a strategy will make money.

What no-code means here

No-code does not mean there is no code. It means you do not write it. Your assistant turns your sentence into a strategy.py written against the OptAlgo SDK: plain Python you can read, download and keep. You never open an editor, but the rules are not hidden in a black box.

That matters for two reasons. You can ask the assistant to explain any line in plain words. And the bot that runs later uses the same code the backtest used, not a re-typed copy of your idea.

There is a second no-code path on OptAlgo: sending TradingView alerts to a webhook. This guide covers the first path, describing the strategy to an AI assistant. Both are on the no-code trading bot page.

What you need

You do not need an exchange account. To connect, add https://mcp.optalgo.com/mcp as a custom connector in Claude.ai or ChatGPT, sign in to OptAlgo and approve. It uses OAuth, so no key is pasted into the chat. The connector setup guide has the screens. The connector is paper-only by design: it can backtest and run paper bots, and it can never place a live order or change a live bot.

Choose an idea you can say in one sentence

A good first idea has four parts: a coin, a bar size, an entry rule and an exit rule. If you cannot fit it in one sentence, it is probably several ideas, and you will not know which one did the work.

Avoid sentences that describe a wish. “Buy low and sell high” and “find me a profitable strategy” are not rules. An assistant will still produce something for them, and that something will be its guess, not your idea.

Five example sentences

The first is trend following, the second a breakout, the fourth mean reversion. We will follow the first sentence through every step.

What the assistant does at each step

  1. Reads the SDK. The assistant asks OptAlgo what a strategy may use and gets a working example. Strategy code may import only numpy, math and the OptAlgo SDK, and it runs in a sandbox.
  2. Writes strategy.py. Your sentence becomes entry and exit rules in Python. Ask to see the code, or ask for it line by line in plain words.
  3. Validates for free. Validation queues nothing, loads no market data and costs nothing. It runs the code once on synthetic bars and lists every problem at once: code, symbols, data for the window, plan limits and quota, plus lookahead and non-deterministic code. Each finding comes with its fix, and the assistant applies them and validates again.
  4. Runs the backtest. This is the step that uses quota. A backtest is charged by run time: one unit is up to 10 seconds.
  5. Reports the result. First the honesty grade, then the search verdict, then the numbers and a chart of every entry and exit.

You only have to do two things in this loop: say the idea, and hold the assistant to the order of step 5. A useful line to add to your first message is “report the honesty grade and the verdict before any return”.

Reading the grade and the verdict

Our example sentence, run on OptAlgo on 2026-11-22 with Binance data: long when EMA 20 crosses above EMA 80, 3 ATR trailing stop, BTCUSDT perpetual, 4h bars, 2022-01-01 to 2026-01-01.

Start with the honesty grade. It runs from A to D. A means every fill settled on real data with fees, funding and exchange rules. B relies on a few standard assumptions. C is usable, with caveats the findings explain. D is not realistic: fix it and rerun, and never quote its numbers.

Then the search verdict. OptAlgo counts how many variants of the strategy you ran on overlapping data in the last 30 days and checks whether the Sharpe ratio survives that many tries. The verdict is ok, caution, likely overfit, or no edge when the run lost money. No edge means change the idea, not the numbers.

Only then read the numbers, and read them as a description of what you would have lived through. A win rate of 40% means six trades in ten lost. A max drawdown of -15.0% means the account was at some point 15% below its earlier peak. If that would have made you switch the bot off, the +70.8% was never yours to earn.

And one more thing a no-code builder needs to hear early: this is one backtest, in-sample. Grade A says the fills were honest. It does not say the rule has an edge. When we ran a walk-forward research run of this same idea, the verdict was likely overfit. The AI backtest checklist shows that run and the seven checks to make before trusting any result.

Deploying to paper

Deploying is one request to the assistant, or one click in the app. OptAlgo then runs your code at every bar close in a sandbox, on paper, as a bot in My Bots.

A deploy gate sits in front of this step. Only backtests graded A to C can be deployed. The gate always refuses a losing backtest, and it refuses one judged likely overfit unless an out-of-sample research run of the same code passed. If the gate refuses, that is the product doing its job: the answer is a better test, not a way around it.

Once the bot runs, you can watch every decision it makes, pause it or wind it down. Compare with backtest matches the bot's real paper trades against a backtest of the same period, trade by trade, so you see whether live behaviour matches the test. Paper trading is where you find out, at no cost, whether the idea holds on prices it has never seen.

Limits of the Free plan

Hosted strategies cannot go live on the Free plan. Taking a hosted strategy live needs Plus or Pro, and your confirmation in the OptAlgo app. Current plans are on the pricing page.

What no-code does not remove

No-code removes the typing. It does not remove the judgement. Used that way it is a real help: you can test an idea in minutes, read an honest result, and throw away the ones that do not hold up before they cost anything.

For the limits that apply once real money is involved, see security.

Frequently asked questions

Can I build a crypto trading bot with no coding?

Yes. Describe the strategy to Claude or ChatGPT through the OptAlgo connector. The assistant writes the Python, validates it for free and backtests it, and OptAlgo runs it as a paper bot. You can also use TradingView alerts with a webhook. Neither path needs you to write code.

Is a no-code AI trading bot free?

On OptAlgo the Free plan includes 50 backtest units a month, 3 coins per run, one hosted strategy on paper, 200 paper trades a month, 2 bots and 10 live trades a month, with no card. Taking a hosted strategy live needs Plus or Pro.

Can the AI trade real money on its own?

No. The chat connector is paper-only and can never place a live order or change a live bot. A strategy goes live only when you confirm it in the OptAlgo app, on a Plus or Pro plan.

Do I get to see the code the AI wrote?

Yes. The strategy is a plain Python file, strategy.py. You can read it, ask the assistant to explain it, and download the exact file a backtest or a hosted strategy ran.

Does a good backtest mean the bot will make money?

No. One backtest is in-sample and is never proof. In our example a run graded A with +70.8% was judged likely overfit by a walk-forward of the same idea. Check out-of-sample results and run on paper first.

Your next strategy is one sentence away.
Test it honestly before it touches money.

Start free, no card

50 backtest units a month on the Free plan.