skill-creation-walkthrough

Step-by-step guide for creating your own Claude Skills, from deciding whether a skill is the right tool to writing the SKILL.md file, structuring reference material, and making it trigger reliably. Use when you want to package a workflow, framework, or repeated task into a reusable Skill, when an existing skill is not triggering or not loading the right context, when you are auditing a skill that is underperforming, or when you want to publish a skill for others. Also triggers when someone asks "how do I make a skill" or "what makes a good skill". Useful for individuals, teams, and anyone publishing skills publicly.

1.24x

Quality

—

Does it follow best practices?

Impact

98%

1.24x

Average score across 2 eval scenarios

Securityby

Passed

No known issues

Quality

Content

85%

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

A high-quality meta-skill body: actionable structure templates, a phased workflow with testing/feedback loops, and clean one-level progressive disclosure to three real reference files. The main weakness is redundancy between Phase 3 and the 'description anatomy' deep dive, which pushes conciseness down.

Suggestions

Collapse the 'Deep dive: description anatomy' into Phase 3 or move it to references/description-cookbook.md, since it restates the same four-part formula verbatim.

Trim the conceptual explanation of how progressive disclosure works (e.g., 'Claude does not read every skill on every turn...') — Claude already knows this; keep only the authoring implications.

Consider relocating the worked post-mortem example to a reference file to keep the SKILL.md body comfortably under its own 250-line guidance.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and well-structured, but the 'Deep dive: description anatomy' substantially re-covers the four-part description formula already given in Phase 3, and the body explains progressive-disclosure mechanics Claude already knows ('Claude does not read every skill on every turn...'). It could be tightened without losing guidance.	2 / 3
Actionability	As an instruction-only process skill it gives concrete, copy-ready guidance: a full SKILL.md structure template, reference naming patterns, a worked end-to-end example, and good/bad description contrasts — actionable rather than abstract.	3 / 3
Workflow Clarity	A clear Phase 1–6 sequence culminating in an explicit testing checklist (trigger, edge-case, false-positive, output, length) with a diagnose-then-fix feedback loop mapping symptoms to fixes, providing the validation checkpoints the rubric rewards.	3 / 3
Progressive Disclosure	The body is a lean overview pointing to three well-signaled, one-level-deep references (skill-template.md, description-cookbook.md, methodology-vs-implementation.md), all verified present in references/ and listed with one-line descriptions in the 'Reference files' section.	3 / 3
	Total	11 / 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.

A strong, complete description: it states concrete capabilities, gives multiple natural trigger phrases, includes explicit 'Use when' and 'Also triggers when' guidance, and scopes an audience. Minor note that 'Use when you want to...' uses second person, but it follows the accepted trigger-clause pattern and does not warrant a specificity reduction.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'deciding whether a skill is the right tool to writing the SKILL.md file, structuring reference material, and making it trigger reliably' — matching the 'multiple specific concrete actions' anchor.	3 / 3
Completeness	Explicitly answers both what (a step-by-step skill-creation guide) and when, with a 'Use when...' clause and an 'Also triggers when...' clause for edge cases, plus a 'Useful for...' audience line.	3 / 3
Trigger Term Quality	Covers natural trigger phrases a user would say: 'when you want to package a workflow', 'when an existing skill is not triggering or not loading the right context', 'when you are auditing a skill that is underperforming', and the quoted 'how do I make a skill' / 'what makes a good skill'.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (authoring/maintaining Claude Skills) with distinct meta-skill triggers unlikely to overlap with domain skills.	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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	14 / 16 Passed

Repository: rampstackco/claude-skills
Commit: f1bc195

Reviewed: 3 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.