A question-answer dataset for evaluating factual knowledge about the Nushell shell. Contains 151 Q&A pairs covering Nushell's features, commands, configuration, data types, closures, pipelines, and version-specific behavior.

Dataset Structure

Each example in examples/ is a YAML file with two fields:

yaml

question: What are closures in Nushell and how are they used?
answer: |
  Closures are anonymous functions...

Fields

Field	Type	Description
`question`	string	The question about Nushell
`answer`	string	The factual answer

Building

Run build.py to generate data.jsonl from the YAML examples:

uv run --with pyyaml python3 build.py

Usage

python

from datasets import load_dataset

ds = load_dataset("json", data_files={"train": "data.jsonl"})
example = ds["train"][0]
print(example["question"])
print(example["answer"])

Intended Use

This dataset is designed for:

Evaluating LLM factual knowledge about Nushell
Benchmarking model accuracy on shell-specific topics
Knowledge retrieval and RAG evaluation

License

MIT

knoopx/nushell-qa

Nushell Knowledge Evaluation Dataset