wandb-experiment-logger

Wandb Experiment Logger - Auto-activating skill for ML Training. Triggers on: wandb experiment logger, wandb experiment logger Part of the ML Training skill category.

1.02x

Quality

Does it follow best practices?

Impact

97%

1.02x

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/wandb-experiment-logger/SKILL.md

Evaluation results

95%

Adding Experiment Tracking to an ML Classification Pipeline

Wandb experiment logging with sklearn

Criteria

Without context

With context

wandb import

100%

wandb.init called

100%

Project name set

100%

Hyperparameters in config

100%

Metrics logged

100%

wandb.finish called

100%

Offline mode configured

100%

pip install wandb

25%

37%

Script executes

100%

Production-ready structure

100%

Without context: $0.2511 · 1m 14s · 18 turns · 17 in / 3,162 out tokens

With context: $0.4621 · 1m 52s · 30 turns · 27 in / 5,012 out tokens

100%

Neural Network Training Experiment Tracker

PyTorch training with wandb metric tracking

Criteria

Without context

With context

wandb.init with config

100%

Project name provided

100%

Per-epoch metric logging

100%

wandb.watch used

100%

wandb.finish called

100%

Offline mode set

100%

pip install invoked

62%

100%

Script executed successfully

100%

Step tracking in log calls

100%

experiment_notes.md created

100%

Without context: $0.4337 · 5m 38s · 23 turns · 23 in / 6,935 out tokens

With context: $0.5480 · 4m 12s · 31 turns · 292 in / 6,816 out tokens

96%

Comparing Model Configurations with Experiment Tracking

Hyperparameter comparison experiment tracking

Criteria

Without context

With context

Multiple wandb.init calls

100%

wandb.finish per run

100%

Config per run

100%

Run names set

100%

Metrics logged per run

100%

Project name consistent

100%

Offline mode used

100%

pip install present

50%

Script executed

100%

results_summary.md created

100%

Without context: $0.4699 · 2m 4s · 24 turns · 24 in / 6,501 out tokens

With context: $0.3441 · 1m 36s · 21 turns · 55 in / 4,828 out tokens

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

Evaluated: 18 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Adding Experiment Tracking to an ML Classification Pipeline Neural Network Training Experiment Tracker Comparing Model Configurations with Experiment Tracking

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.