For AI agents: the complete documentation index is available at https://viteval.dev/llms.txt, the full documentation bundle is available at https://viteval.dev/llms-full.txt, and this page is available as Markdown at https://viteval.dev/cli/data.md.
V1 release candidate available.Migration guide →
GitHub

data

Discover named datasets and run their generators using the selected Viteval configuration and selected destination providers. Evaluation tasks and scorers do not run during generation.

Usage

viteval data [pattern] [options]

Arguments

ArgumentDefaultDescription
pattern**/*.{dataset,eval}.{js,ts,mts,mjs}Glob pattern matching dataset or evaluation modules

Quote patterns to prevent your shell from expanding them before Viteval runs.

Options

OptionDefaultDescription
--overwritefalseRegenerate and replace existing datasets
--root <path>Current directorySelect the project root
--config <path>, -cViteval discoverySelect a Viteval configuration file
--verbose, -VfalseEnable verbose output

Discovery and persistence

Modules can use default exports, named exports, or multiple defineDataset() calls. Datasets inside evaluation files are discovered even when not exported. Aliases and re-exports of the same dataset generate it once. Distinct dataset objects with the same name fail instead of overwriting each other.

Existing datasets are preserved unless --overwrite is passed. Datasets with providers: [] run their generators without writing persisted files. Modules that fail during import do not publish their datasets.

Module-level code still runs on import. Keep expensive calls inside dataset generators or evaluation tasks.

Examples

viteval data
viteval data "src/user-questions.dataset.ts"
viteval data --overwrite
viteval data --root ./app --config viteval.config.ts

See datasets for storage modes and defineDataset() for the generator API.