Agent task: build a FLUX workflow in a local ComfyUI folder
What it tests: Coding-agent discipline in a real folder: explore first, follow existing file formats, no downloads, no moving or deleting files, validate the output and report assumptions.
Prompt length: 1,424 characters, roughly 410 tokens (estimate).
Du befindest dich in einem Ordner, der meine lokale ComfyUI-Installation enthält. Aufgabe: Erstelle einen neuen ComfyUI-Workflow, der ein lokal vorhandenes FLUX-Modell verwendet. Verwende bevorzugt ein kleines/schnelles …
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
The agent lists the folder structure before writing; picks a small FLUX model that is actually present; saves 'dein test.json' next to the existing workflows; the JSON is valid and loads in ComfyUI; every referenced model file exists; the optional ControlNet group is disabled and the workflow runs without it; the final summary answers all four questions honestly.
Scorecard tick what the model got right: 0/7
Runs in your browser; nothing is saved or sent.
Where models usually fail
Invented model filenames; downloading models; overwriting or moving existing files; claiming it validated the JSON without doing so.
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#11-comfyui-workflow-test
- 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?