Someone promoting a trading tool usually shows you their best result. The author of day-trader-skill put an example in the README that came out "close to a coin flip". That's why this project caught our attention.
It's the sequel to trader-skill, and in 16 days it collected more than 5,000 stars. We read the docs. The numbers are the author's and we didn't recalculate them.
TeslaLet's say it first, as the author does: most people who day trade lose money. This article isn't encouraging you to day trade.
One trade: which of the 11 skills runs when
Finding and entering
/trader-scannersearches thousands of charts for setups, such as opening-range breakouts./trader-newsshows the news on that stock./trader-entrywatches the setup bar by bar and tells you when to get in./trader-exitsets the stop and target before you enter.
Checking and sizing
/trader-timemachinemore on this below./trader-budgetworks out the share count. The README accepts a $50 maximum loss per trade with a stop $5.86 away, so 8 shares./trader-riskchecks your limits and stops anything over them./trader-brokernothing happens unless you type yes. A paper account is the default.
Watching
/trader-alertsrings on the stop, the target or closing time./trader-analystreads earnings reports, and/trader-pelosireads Congress's trades.
What the Time Machine does
The plan is set: entry, stop, target. Before any money goes in, the skill rebuilds the rest of the day 10,000 times from the stock's own 5-minute bars over the last 60 days. Then it counts: how often the target came first, how often the stop, how often neither.
The README example: Tesla, October 19, entry at $364.18.
- Neither by the close45.1%
- Target first (profit)29.4%
- Stop first (loss)25.5%
Source: day-trader-skill README (TSLA, 2026-10-19, the author's calculation). A simulation, not a prediction.
Profit is ahead by a small margin. But don't get excited yet. The skill also tested the same rule on the previous 59 real sessions. It fired 33 times, and the result went the other way:
- Stop first (loss)15
- Target first (profit)13
- Neither5
Source: day-trader-skill README. A small sample: one stock, under 60 days of data.
The author's verdict is short: close to a coin flip. And they add that this is the whole point of looking before you risk a dollar.
What we think
The Time Machine's best quality is that it doesn't hand out optimism. Many people see a pretty pattern on a chart and tell themselves "this will work". Here the answer comes out as a number.
Keep its limits in mind too:
- The last 60 days aren't the future. An earnings report, a trading halt or surprise news may not be in that window.
- 33 times is very few. You can't draw big conclusions from a sample like that.
- A stop isn't a guarantee. In a fast move the order can fill well below the stop. The author says this too.
- Costs. More trades mean more commission.
If you want to try this, start on a paper account and size your positions properly. As for the religious side of rapid trading, please ask scholars; we don't issue rulings here.
For information only, not investment advice. This is a third-party project and we're not affiliated with it.
Stocks mentioned in this article
Terms in this article
Check halal stocks yourself
Verdicts, ratios and last check dates for every NASDAQ and NYSE stock.
Related articles
- AI Toolstrader-skill: what do these 11 Claude skills actually do?A free GitHub project that reads fund filings, splits your budget and won't place an order until you say yes. Each skill in plain words, with its main limitation.
- GuideWhy you should trade on a demo account firstA demo account is a free practice ground: test your plan, find mistakes cheaply and build statistics. How to use it and where it differs from the real market.
- GuidePosition size and risk/reward: the 2 questions before every tradeHow many shares to buy and whether the trade is worth it — work out both before you enter. Formulas, a worked example and the most common mistakes.