LLM Test Prompts

All prompts › Simulations & physics

Bee hive simulation with self-checked invariants

Simulations & physicsvery hardbuild testPrompt in Germansingle-file HTMLcanvasinvariantsGerman

What it tests: Coupled subsystems (cells, three resource pools, brood cycle, bee roles), a fixed 10 ticks/s logic loop decoupled from rendering, a priority order for features, and a selfCheck() that must test invariants rather than existence.

Prompt length: 6,637 characters, roughly 1,900 tokens (estimate). Set a large context in your runner so the prompt and a long answer both fit.

ROLLE Du baust interaktive Simulationen als Single-File-Web-Apps. Deine Stärke ist, dass die Systemlogik unter der Oberfläche wirklich stimmt — nicht nur die Optik. AUFGABE Erstelle eine interaktive Simulation eines Bienenstocks …

The source has no licence that allows copying, so only a short excerpt is shown here. The full prompt is one click away at the source.

How to grade it

Open the file from file:// and read the selfCheck table. Full marks only if honey can only appear via stored nectar dried below 20 % water; the resource balance holds exactly (collected = stored + used + in transit); the brood counter matches the cells; logic runs at 10 ticks/s independent of FPS, and 2× doubles ticks, not frames; the stress-test button applies the stated settings for 30 sim days and prints a report; colony collapse stops the sim and shows the cause chain. The 'Design decisions' block must name five real trade-offs.

Scorecard tick what the model got right: 0/6

Runs in your browser; nothing is saved or sent.

0/6

Where models usually fail

Honey cells that skip the nectar phase; a selfCheck that only tests whether the canvas exists; one DOM element per bee (lags at a few hundred agents); half-built features with no note in the status area, which the prompt explicitly forbids.

Running it

Run it one-shot and judge the first answer. Open the result exactly as delivered (double-click for HTML). Give the model a large output budget: long specs get cut off on small context settings.

Source

Author
Arnie (GitHub @Arnie936)
From
Arnie936 / llm-prompting-tests
Exact file or post
https://github.com/Arnie936/llm-prompting-tests/blob/03445cf074852ec5535ece9d123bdff442e63ef5/README.md#1-bienenstock-simulation
Licence
No license file (all rights reserved)
How we use it
Excerpt only: open the full prompt at the source

Testing a local model? Check first whether it fits your hardware with Can I run this LLM?

More in Simulations & physics