building-classification-models

Build and evaluate classification models for supervised learning tasks with labeled data. Use when requesting "build a classifier", "create classification model", or "train classifier". Trigger with relevant phrases based on skill purpose.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

12%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body reads as a marketing-style overview: verbose, abstract, and devoid of executable guidance or links to the bundled scripts and assets that actually do the work. Workflow sequencing is present but lacks validation checkpoints, and progressive disclosure fails because the real bundle is never referenced.

Suggestions

Replace the Overview/How It Works prose with concrete, executable steps, e.g. 'Run `python scripts/model_builder.py data.csv assets/model_config_template.json`' and a copy-paste code snippet.

Cut generic filler Claude already knows and link explicitly to the bundled files (scripts/model_builder.py, scripts/data_validator.py, assets/model_config_template.json).

Add explicit validation checkpoints to the workflow, such as 'Run data_validator.py first; only train when validation passes, then re-evaluate and report metrics'.

Dimension	Reasoning	Score
Conciseness	The body is padded with concept-level filler Claude already knows ('This skill empowers Claude...', 'data preprocessing, feature selection...') and restates the description in the Overview, wasting the context budget on generic explanations rather than skill-specific instructions.	1 / 3
Actionability	There is no executable code, no concrete commands, and no real tool invocations; it only describes actions in the abstract ('Generate Python code using the classification-model-builder plugin') and references scripts that the body never links to or shows how to run.	1 / 3
Workflow Clarity	How It Works lists a sequenced three-step process and there is a separate Error Handling section, but there are no explicit validation checkpoints or feedback loops for the model-training workflow despite it involving evaluation and batch operations.	2 / 3
Progressive Disclosure	The body is essentially a monolithic overview that never points to the bundled references/scripts/assets; the Resources section lists only 'Project documentation' and 'Related skills and commands' generically, so the actual bundle (model_builder.py, data_validator.py, model_config_template.json) is not signaled or navigated.	1 / 3
	Total	5 / 12 Passed

Description

50%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description covers the what and offers an explicit Use-when trigger clause, but the trigger guidance is partly boilerplate and the capability list is broad rather than concretely enumerated. It sits at a solid level 2 across all dimensions.

Suggestions

Replace 'Trigger with relevant phrases based on skill purpose' with specific natural phrasings a user would say, such as 'detect spam', 'predict churn', or 'classify labels'.

Add a couple more concrete capabilities (e.g., preprocessing, feature selection, hyperparameter tuning, metric reporting) to lift specificity to level 3.

Dimension	Reasoning	Score
Specificity	It names the domain and a couple of concrete actions ('Build and evaluate classification models'), but does not enumerate multiple specific actions like preprocessing, feature selection, or metrics reporting, only the broad build/evaluate pair.	2 / 3
Completeness	It states what the skill does and includes an explicit 'Use when requesting...' clause, but the trigger guidance is partly boilerplate ('Trigger with relevant phrases based on skill purpose') rather than fully concrete, so it does not cleanly reach the clearly-explicit level 3.	2 / 3
Trigger Term Quality	It lists several natural trigger phrases ('build a classifier', 'create classification model', 'train classifier'), but the trailing 'Trigger with relevant phrases based on skill purpose' is generic boilerplate and common variations like 'detect spam' or 'predict churn' are missing.	2 / 3
Distinctiveness Conflict Risk	'Supervised learning tasks with labeled data' gives it a niche, but classification is a broad area that overlaps with general ML/data-analysis skills, and the generic trailing trigger guidance weakens distinctiveness.	2 / 3
	Total	8 / 12 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: jeremylongshore/claude-code-plugins-plus-skills
Commit: 2cafb23

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

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.