How the random selection works
Enter one option per line. Blank lines are removed, surrounding spaces are trimmed, and at most fifty choices are included. When Pick a slip is pressed, the browser requests a cryptographically generated random integer and maps it to one eligible option. The tool does not rank, rewrite, understand, or transmit the list.
Every eligible line has the same chance under the simple modulo selection used here, subject to the tiny implementation bias that can occur when a 32-bit range is not evenly divisible by the list length. For everyday low-stakes choices that bias is negligible. The selection is not seeded by a name, date, horoscope, or prior result.
Decide which options are eligible first
Randomness can choose among acceptable alternatives; it cannot make an unsafe alternative acceptable. Before using the button, remove anything illegal, dangerous, unaffordable, non-consensual, or inconsistent with an obligation. If one choice has materially different consequences, compare those consequences directly instead of giving every line equal weight.
For a reversible decision, noticing your reaction to the selected slip can also reveal a preference. You remain free to reject the result. For medical treatment, contracts, investments, travel safety, employment, or relationship boundaries, use evidence and qualified advice. A fair random pick is not an assessment of what is best.
Local by design
Choices stay inside the page’s component state. They are not saved to an account, placed in the URL, or sent to an analysis service. Reloading the page clears the custom list. That makes the picker suitable for ordinary private prompts, though users should still avoid pasting secrets when a generic label would work.
The interface reports the number of eligible choices and states that it cannot evaluate them. This separates a transparent utility from divination or automated decision making. Use the picker when you genuinely accept any outcome, and keep responsibility for the choice with the people affected rather than assigning it to the tool. If options need different weights, dependencies, deadlines, or vetoes, this equal-choice picker is the wrong model; write those constraints down and compare them directly first, carefully.