feature-importance-analyzer

Feature Importance Analyzer - Auto-activating skill for ML Training. Triggers on: feature importance analyzer, feature importance analyzer Part of the ML Training skill category.

1.11x

Quality

Does it follow best practices?

Impact

90%

1.11x

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-importance-analyzer/SKILL.md

Evaluation results

92%

10%

Customer Churn Feature Analysis

Sklearn feature importance pipeline

Criteria

Without context

With context

Uses sklearn

100%

Feature importance computed

100%

Features ranked

100%

Step-by-step structure

80%

100%

Output written to file

100%

Production-ready code style

25%

100%

Result validation

80%

100%

Handles data preparation

100%

Uses pip for dependencies

Visualization or summary

100%

Without context: $0.4102 · 1m 37s · 22 turns · 23 in / 5,511 out tokens

With context: $0.5264 · 1m 49s · 31 turns · 138 in / 6,281 out tokens

90%

20%

Neural Network Feature Attribution for Sensor Data

PyTorch gradient-based feature importance

Criteria

Without context

With context

Uses PyTorch

100%

Gradient-based importance

66%

100%

Model trained on data

100%

Feature scores output

100%

Step-by-step structure

100%

Production-ready code style

75%

100%

Output validation

60%

70%

Data preparation step

100%

Ranked output

100%

pip install used

Without context: $0.3656 · 1m 41s · 18 turns · 17 in / 6,104 out tokens

With context: $0.7082 · 6m 20s · 35 turns · 33 in / 10,248 out tokens

90%

-3%

Loan Default Prediction: Feature Analysis and Model Optimization

Hyperparameter tuning with experiment tracking

Criteria

Without context

With context

Uses sklearn or pytorch/tensorflow

100%

Hyperparameter tuning

100%

Experiment tracking log

100%

Feature importance in output

100%

Step-by-step code structure

100%

Production-ready style

100%

Output validation

100%

80%

Data preparation step

100%

87%

Best config identified

100%

pip install used

Without context: $0.3842 · 1m 44s · 19 turns · 20 in / 5,837 out tokens

With context: $0.5731 · 2m 26s · 26 turns · 356 in / 7,656 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 Feature Analysis Neural Network Feature Attribution for Sensor Data Loan Default Prediction: Feature Analysis and Model Optimization

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.