Skill collection for agents / Claude Code to play Kaggle competitions on tabular data using TabPFN, which beats classic ML (XGBoost, Catboost etc.) in >80% of cases.
dianaprior/kaggle-competition-agent-skill
Skill Claude Code
Run a TabPFN classification baseline, generate the first submission, rapidly probe features, then optimize with GBT ensembles, threshold tuning, and calibration. Use after tabpfn-explore has prepared the data and CV folds.
dianaprior/kaggle-competition-agent-skill
Skill Claude Code
Shared identity, behavior rules, workflow principles, and project conventions for TabPFN tabular competition skills. Referenced by tabpfn-classify, tabpfn-regress, and tabpfn-explore — not invoked directly.
dianaprior/kaggle-competition-agent-skill
Skill Claude Code
EDA, data profiling, adversarial validation, preprocessing checks, CV scheme setup, and API budget verification for tabular Kaggle competitions. Run at the start of every new competition before any modeling.
dianaprior/kaggle-competition-agent-skill
Skill Claude Code
Run a TabPFN regression baseline, generate the first submission, then optimize with GBT ensembles and regression-specific post-processing (clipping, target transforms, rank blending). Use after tabpfn-explore has prepared the data and CV folds.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: