Kumo Data Science

Use for any work with KumoAI predictive models on Snowflake data: train a model, score data with a trained one, or answer questions about an existing model or run — explain its predictions, diagnose its training, publish, reload or retain it, train or score on custom labels or rows, or tune it. Covers churn, propensity, fraud, recommendation, forecasting and link prediction; feature attribution and column cohorts; metrics, trials and learning curves; registry publishing and artifact retention; caller-supplied labels, weights and splits; encoders, run mode and experiments. For a new model it checks that the request names the data (a schema, connector, or semantic view), the outcome and the time window, confirms the inferred tables and the task, then generates one notebook or script; it runs the job only when asked, in the notebook or as a Snowflake ML Job on a compute pool. Triggers: kumo, predict, churn, recommend, forecast, fraud, train a model, run inference, explain a prediction, feature importance, model metrics, publish or load a model, tune a model, semantic view, ML Job, compute pool.

mkdir -p ~/.claude/skills/kumo-data-science
curl -L https://claudskills.com/skills/kumo-data-science/SKILL.md \
  -o ~/.claude/skills/kumo-data-science/SKILL.md

@misc{kumo-data-science-2026,
  author    = {NVIDIA},
  title     = {Kumo Data Science [Claude Code skill]},
  year      = {2026},
  publisher = {ClaudSkills},
  url       = {https://claudskills.com/skills/kumo-data-science/}
}