In this article
synth-data-generate
| Field | Value |
|---|---|
| Kind | skill |
| Source | .github/skills/data-science-engineering/synth-data-generate |
| Invocation | Invoked directly as /synth-data-generate; model invocation is disabled, so agents do not load it |
| Interactive | No |
What it does
Generate synthetic datasets in a Jupyter notebook after the dataops synthetic-data preflight passes. Use when creating new synthetic data or preparing a guarded local replacement candidate from an approved source.
When to use it
Use this skill when you need a notebook that produces realistic fictional data for development, demonstrations, or data-science experiments. Use approved source data instead when the work requires actual observed records.
How to use it
Describe the subject and optionally provide an example schema or sample structure. Confirm before generating PII-like fields, keep values fictional, and review any proposed update to an existing data source.
Example usage
/synth-data-generate subject="retail inventory demand" example_data="inventory.csv schema"
The skill creates a notebook that generates fictional inventory and demand records with realistic relationships.