ml-training-recipe

Find implementable ML training recipes from papers, datasets, docs, and code. Use when the user wants to fine-tune, train, reproduce, or choose a practical ML method, dataset, hyperparameter setup, or benchmark recipe.

Quality

88%

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

Quality

Content

92%Weight 40%Scale 1-5

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

An efficient, well-structured delegating skill that points to a single slash-command workflow with clear agents and output. Its only weakness is that actionability rests entirely on the delegated workflow, with no inline examples of common recipe-finding cases.

Suggestions

Add one or two concrete examples of the output (e.g., a sample ranked recipe brief) so the expected result is unambiguous without inspecting the /recipe workflow.

Briefly note the criteria the /recipe workflow uses to rank recipes, so the skill communicates value even before the slash command expands.

Clarify when the optional `verifier` agent should be invoked, turning the implicit validation step into an explicit checkpoint.

Dimension	Reasoning	Score
Conciseness	Six lean lines that assume Claude's competence; every line earns its place, including the useful guard against reading a relative prompt-template path.	5 / 5
Actionability	Gives a concrete executable command ('Run the `/recipe` workflow') and names the output artifact and location, but does not illustrate common cases or the workflow's internal steps.	4 / 5
Workflow Clarity	A simple single-purpose skill whose single action is unambiguous, with agents and output destination stated; the simple-skill exception applies since no destructive/batch operation is exposed here.	5 / 5
Progressive Disclosure	Under 50 lines with no bundle files and no need for external references; the body is cleanly organized into action, agents, and output sections.	5 / 5
	Total	19 / 20 Passed

Description

83%Weight 40%Scale 1-5

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

A strong description that clearly states both capability and trigger conditions with concrete, natural-language terms. It is specific and well-scoped, with only minor room to broaden trigger synonyms and reduce overlap ambiguity.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions/sources ('Find implementable ML training recipes from papers, datasets, docs, and code') plus 'fine-tune, train, reproduce, or choose', giving good concrete coverage with minor gaps.	4 / 5
Completeness	Explicitly answers both what ('Find implementable ML training recipes from papers, datasets, docs, and code') and when ('Use when the user wants to fine-tune, train, reproduce, or choose...') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural user terms ('fine-tune, train, reproduce, choose ... method, dataset, hyperparameter setup, benchmark recipe') with synonyms, though a few common phrasings like 'model' or 'training run' are missing.	4 / 5
Distinctiveness Conflict Risk	The 'implementable ML training recipes' niche is mostly distinct with clear triggers, but carries minor overlap risk with general ML research or coding skills.	4 / 5
	Total	17 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: companion-inc/feynman
Path: skills/ml-training-recipe/SKILL.md
Commit: 95c51d9

Reviewed: about 9 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.