CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

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

91%

Does it follow best practices?

Impact

98%

1.24x

2 of 3 eval scenarios. Add 1 more for a full score.

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

81%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.

A strong, highly actionable walkthrough with an exemplary testing/iteration loop and well-organized one-level-deep references. The main weakness is redundancy: the description-anatomy deep dive repeats Phase 3's formula, adding tokens without new guidance.

Suggestions

Fold "Deep dive: description anatomy" into Phase 3 or move it into references/description-cookbook.md — it restates the 4-part formula and per-sentence good/bad examples already given in Phase 3.

Move the worked example (post-mortem walkthrough) to a references/example-skill.md file and link it, keeping the body well under the 250-line limit the skill itself prescribes.

Trim the "The framework: how skills work" section to the three practical implications, cutting the general explanation of progressive disclosure that Phase 5 and the length-test guidance already imply.

DimensionReasoningScore

Conciseness

The body is mostly efficient, but "Deep dive: description anatomy" (~30 lines) restates the 4-part description formula that Phase 3 already presents in full with good/bad examples, and the second deep dive re-walks the same six phases. This is more than minor trimmable over-explanation (score 4) but well short of heavily padded (score 2).

3 / 5

Actionability

Fully actionable for an instruction-only skill: a copy-paste SKILL.md structure template, concrete reference-naming patterns ("references/[noun]-template.md"), a testing checklist with explicit pass/fail criteria, a symptom-to-fix mapping, and a concrete deliverables list in the Output format section. No code is required, and the guidance is directly executable.

5 / 5

Workflow Clarity

Phases 1-6 are clearly sequenced, and Phase 6 provides explicit validation checkpoints (trigger, edge-case, false-positive, output, and length tests) plus a feedback loop mapping observed failure to a specific fix. No destructive or batch operations, so no workflow cap applies.

5 / 5

Progressive Disclosure

Good structure: all three cited reference files exist, are one level deep, and are well-signaled both inline and in a dedicated Reference files section with one-line descriptions. It falls short of a 5 only because the deep-dive sections duplicate Phase 3 content inline that the skill's own rules suggest belongs in references/description-cookbook.md.

4 / 5

Total

17

/

20

Passed

Description

100%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.

The description exemplifies its own four-part formula: concrete actions, explicit trigger phrases with direct user-quote examples, edge-case coverage via "Also triggers when...", and an audience clause. It is dense with signal rather than padded, and conflict risk is minimal.

DimensionReasoningScore

Specificity

"From deciding whether a skill is the right tool to writing the SKILL.md file, structuring reference material, and making it trigger reliably" names multiple specific concrete actions that comprehensively cover the skill-creation process. It clearly matches the anchor for comprehensive coverage rather than the score-4 anchor with minor gaps.

5 / 5

Completeness

Explicitly answers both what ("Step-by-step guide for creating your own Claude Skills...") and when ("Use when you want to package a workflow...", "Also triggers when someone asks...") with concrete trigger phrases. The 'Use when...' clause is present and specific, so the completeness cap does not apply.

5 / 5

Trigger Term Quality

Includes natural phrases users would actually say — "package a workflow, framework, or repeated task into a reusable Skill", "an existing skill is not triggering or not loading the right context", "auditing a skill that is underperforming", and the direct quotes "how do I make a skill" and "what makes a good skill". Coverage includes synonyms (workflow/framework/repeated task) and is comprehensive.

5 / 5

Distinctiveness Conflict Risk

Occupies a clear meta-niche (authoring skills themselves) with quoted, distinctive trigger phrases that are unlikely to fire for unrelated skills. Voice is third person ("Step-by-step guide for creating...") so no specificity penalty applies.

5 / 5

Total

20

/

20

Passed

Validation

87%

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

CriteriaDescriptionResult

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

Path

skills/skill-creation-walkthrough/SKILL.md

Reviewed

Table of Contents

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.