How do I start researching a strategy in AI?

Last updated · 2026/11/02

Open a strategy and copy its prompt to discuss it with your AI. Connect AlgoRover tools to continue with adjustments and backtests.

Choose a strategy in the library

Browse the strategy examples, trading logic and supported markets in the library. When you find a strategy you want to explore, click View details on its card.

English strategy library with strategy examples, equity curves, comparison and detail controls

Browse strategies

Open the strategy details

Review the strategy overview and backtest evidence. The Validate in AI panel on the right contains the AI workspace selector and the button for continuing your research.

English strategy detail panel with AI workspace selection and Continue in ChatGPT button

Select the AI you use

Select Codex, ChatGPT or Claude, then click the corresponding Continue button to open the prompt dialog. You can also change the AI workspace in the dialog. Choose the AI you actually use.

English strategy prompt dialog showing Codex, ChatGPT and Claude in the AI workspace menu

Copy the prompt and paste it into your AI

Click Copy prompt. When the button shows Copied, open your selected AI, paste the prompt and send it. The prompt includes this strategy and AlgoRover Live installation instructions.

Install the plugin in Codex or Claude Code. You can discuss the strategy in ChatGPT; running a backtest requires an AI environment connected to AlgoRover tools and an AlgoRover product login. If those tools are not connected, follow the installation guide and continue in Codex or Claude Code.

English strategy prompt dialog with Codex selected and the copy button showing Copied

Explain what you want to test

Tell the AI which market, time frame, rules or parameters you want to examine. For an existing strategy, ask for a specific change and compare the backtests before and after it.

English Research product demo requesting BTC strategies and specifying timeframe, test period and cost assumptions — Website product demo: describe your strategy goal and specify the test conditions.

Keep the evidence in view

Compare the test period, data source and assumptions for fees and slippage alongside each version. A historical backtest describes the selected data and assumptions; it does not predict future results.

English Research product demo checking strategy source, code version, parameters, data window and costs — Website product demo: check sources, versions and reproduction records.