hyperparameter-tuner

Hyperparameter Tuner - Auto-activating skill for ML Training. Triggers on: hyperparameter tuner, hyperparameter tuner Part of the ML Training skill category.

0.88x

Quality

Does it follow best practices?

Impact

77%

0.88x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./planned-skills/generated/07-ml-training/hyperparameter-tuner/SKILL.md

Evaluation results

95%

Churn Prediction Model Tuning

Sklearn hyperparameter search with cross-validation

Criteria

Without context

With context

Uses sklearn

100%

Uses pip for install

100%

Hyperparameter search method

100%

Cross-validation applied

100%

Best params reported

100%

Best score reported

100%

Model saved to disk

100%

results.json written

40%

50%

Script is self-contained

100%

Production-ready structure

100%

Without context: $0.2749 · 1m 20s · 17 turns · 17 in / 3,952 out tokens

With context: $0.5227 · 2m 21s · 30 turns · 111 in / 6,257 out tokens

56%

-26%

Neural Network Learning Rate and Architecture Search

PyTorch training loop with hyperparameter tuning and experiment tracking

Criteria

Without context

With context

Uses PyTorch

100%

Uses pip for install

100%

75%

Multiple hyperparameters searched

100%

Training loop present

100%

Validation accuracy tracked

100%

experiments_log.json written

33%

best_config.json written

30%

40%

No large checkpoint files

100%

Production-ready structure

100%

60%

Script runs end-to-end

70%

60%

Without context: $0.2663 · 5m 6s · 18 turns · 17 in / 4,523 out tokens

With context: $1.2387 · 1s · 1 turns · 3 in / 24 out tokens

81%

-4%

House Price Regression Pipeline

End-to-end ML training lifecycle with data prep and hyperparameter tuning

Criteria

Without context

With context

Uses sklearn or supported framework

100%

Uses pip for install

100%

Data preparation step

100%

Hyperparameter search over multiple params

100%

Cross-validation used

100%

Score metric reported

100%

best_model.pkl saved

100%

50%

pipeline_results.json written

30%

No large temp files

100%

Production-ready structure

20%

Script is self-contained

100%

Without context: $0.4494 · 1m 44s · 26 turns · 26 in / 5,372 out tokens

With context: $0.5045 · 2m 2s · 29 turns · 291 in / 5,795 out tokens

Repository: jeremylongshore/claude-code-plugins-plus-skills
Commit: f17dd51

Evaluated: 17 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Churn Prediction Model Tuning Neural Network Learning Rate and Architecture Search House Price Regression Pipeline

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.