feature-engineering-helper

Feature Engineering Helper - Auto-activating skill for ML Training. Triggers on: feature engineering helper, feature engineering helper Part of the ML Training skill category.

1.07x

Quality

Does it follow best practices?

Impact

86%

1.07x

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/feature-engineering-helper/SKILL.md

Evaluation results

90%

13%

Customer Churn Prediction: Data Preparation Pipeline

Data preparation pipeline with sklearn

Criteria

Without context

With context

Uses sklearn

100%

pip install present

Handles missing values

100%

Handles categorical encoding

70%

100%

Numeric scaling

100%

Pipeline object

100%

Validation report produced

100%

No nulls in output

100%

Step-by-step structure

100%

Output file produced

100%

Without context: $0.3433 · 1m 41s · 18 turns · 19 in / 5,662 out tokens

With context: $0.5198 · 2m 1s · 30 turns · 62 in / 6,268 out tokens

90%

12%

Model Training Experiment: Hyperparameter Search

Hyperparameter tuning and experiment tracking

Criteria

Without context

With context

Uses sklearn or pytorch

100%

pip install present

Multiple configurations tried

100%

Experiment log produced

100%

Metric recorded per run

100%

Best config identified

100%

Reproducibility seed

100%

Validation step present

100%

Step-by-step structure

100%

Production-ready config format

100%

Without context: $0.3611 · 1m 22s · 21 turns · 22 in / 5,105 out tokens

With context: $0.5151 · 2m 5s · 30 turns · 110 in / 6,531 out tokens

80%

-5%

End-to-End ML Feature Engineering Workflow

Production-ready feature engineering workflow

Criteria

Without context

With context

Uses sklearn or pandas

100%

pip install present

Date feature decomposition

100%

Derived ratio feature

100%

Transformation log produced

100%

Rationale in log

100%

Validation summary produced

100%

No nulls in output

100%

Step-by-step structure

100%

Engineered features file produced

100%

Production-ready code quality

37%

100%

Without context: $0.4129 · 1m 58s · 19 turns · 19 in / 7,005 out tokens

With context: $0.6843 · 2m 41s · 31 turns · 291 in / 9,906 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

Customer Churn Prediction: Data Preparation Pipeline Model Training Experiment: Hyperparameter Search End-to-End ML Feature Engineering Workflow

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.