evaluating-machine-learning-models

Evaluate trained machine learning models with the right metrics and comparison logic. Use for benchmark review, threshold selection, calibration, validation, and model comparison; not for feature engineering or leakage auditing.

Quality

68%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./bundled/skills/evaluating-machine-learning-models/SKILL.md

Quality

Content

37%

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

The body is a clean, concise overview but stops at description: it provides no executable guidance, no sequenced evaluation workflow, and does not reference the bundled scripts/assets that would make it actionable. It reads as a routing stub rather than an instructive skill.

Suggestions

Add a concrete "How to run" section that invokes the bundled scripts (e.g. `python scripts/evaluate_model.py --model <path> --dataset <path> --metrics precision,recall,f1`), turning description into executable guidance.

Sequence the evaluation as a numbered workflow with a validation/checkpoint step (run metrics -> compare candidates -> sanity-check against baselines -> report), since evaluation is a multi-step process.

Link to the actual bundle files in the body (e.g. "See scripts/metrics_calculator.py for metric computation" and "assets/visualization_script.py for plots") so the progressive-disclosure structure is discoverable.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence ("Use this skill when the model exists and the question is whether it is good enough"), with no padding or explanation of basic concepts; every section earns its place.	3 / 3
Actionability	The body only describes ("Comparing candidate models with consistent metrics", "Reviewing precision/recall/F1/AUC...") without any executable code, commands, or concrete steps, even though bundled scripts like scripts/evaluate_model.py exist and are never referenced.	1 / 3
Workflow Clarity	Model evaluation is a multi-step task, yet the body offers no sequenced steps or validation checkpoints; it lists "When to Use" and "Typical Outputs" rather than an ordered workflow with feedback loops.	1 / 3
Progressive Disclosure	Sections are well-organized and related skills are signaled, but the body never points to the actual bundle files (scripts/, assets/visualization_script.py), so the available progressive-disclosure structure is not surfaced.	2 / 3
	Total	7 / 12 Passed

Description

100%

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 is concise, concrete, and well-scoped: it states what the skill does, when to use it with natural trigger terms, and explicit exclusions that reduce conflict risk. It hits the top anchor on every dimension.

Dimension	Reasoning	Score
Specificity	Names the core action ("Evaluate trained machine learning models with the right metrics and comparison logic") and lists multiple concrete use cases ("benchmark review, threshold selection, calibration, validation, and model comparison"), matching the multi-action anchor.	3 / 3
Completeness	Explicitly states what it does and provides an explicit "Use for..." trigger clause, satisfying both the what and the when.	3 / 3
Trigger Term Quality	Natural, user-sayable terms like "benchmark review", "threshold selection", "calibration", "validation", and "model comparison" give good coverage; not merely technical jargon.	3 / 3
Distinctiveness Conflict Risk	The "not for feature engineering or leakage auditing" boundary and the focus on evaluating already-trained models carve a clear niche unlikely to trigger for the wrong skill.	3 / 3
	Total	12 / 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: foryourhealth111-pixel/Vibe-Skills
Path: bundled/skills/evaluating-machine-learning-models/SKILL.md
Commit: f627ab5

Reviewed: about 8 hours 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.