create-skill

Guides users through creating effective Agent Skills for Cursor. Use when the user wants to create, write, or author a new skill, or asks about skill structure, best practices, or SKILL.md format.

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

77%

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

A thorough, actionable meta-skill with clear workflows and concrete templates, weakened mainly by length and self-redundancy plus a single-file structure that doesn't practice the progressive disclosure it preaches.

Suggestions

Collapse the duplicate requirement-gathering lists in "Before You Begin" and "Phase 1: Discovery" into a single section to remove redundancy and trim length.

Move the extensive pattern examples, anti-patterns, and the complete worked example into a reference file (e.g. examples.md) referenced one level deep from SKILL.md to honor the progressive-disclosure principle the skill teaches.

Cut explanations of concepts Claude already knows (e.g., that the context window is shared and tokens compete) to tighten token efficiency.

Dimension	Reasoning	Score
Conciseness	At ~495 lines it approaches its own 500-line cap and contains redundancy — "Before You Begin: Gather Requirements" and "Phase 1: Discovery" repeat near-identical lists, and some explanations (context window as a shared good, token competition) restate what Claude already knows.	2 / 3
Actionability	Provides concrete, copy-paste-ready guidance: metadata field tables with exact constraints, directory layouts, code/markdown templates, anti-patterns with explicit do/don't pairs, and a full worked example — highly actionable for an instruction-only skill.	3 / 3
Workflow Clarity	The four-phase Skill Creation Workflow (Discovery → Design → Implementation → Verification) is clearly sequenced, with a Verification phase as a validation checkpoint and multiple checklists plus a Feedback Loop Pattern section.	3 / 3
Progressive Disclosure	No bundle files exist and the skill is a single ~495-line monolith; well-organized into sections, but extensive pattern examples, anti-patterns, and the complete worked example are inline rather than split into one-level-deep reference files as the skill itself prescribes.	2 / 3
	Total	10 / 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 well-crafted description that follows its own advice: specific capabilities, natural trigger terms, and explicit what/when guidance in third person. No vagueness or over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "creating effective Agent Skills for Cursor", "create, write, or author a new skill", and answering questions about "skill structure, best practices, or SKILL.md format" — matching the multi-action anchor.	3 / 3
Completeness	Explicitly answers both what ("Guides users through creating effective Agent Skills for Cursor") and when ("Use when the user wants to create, write, or author a new skill, or asks about skill structure...") with an explicit "Use when" trigger clause.	3 / 3
Trigger Term Quality	Uses natural terms a user would actually say ("create, write, or author a new skill", "skill structure", "best practices", "SKILL.md format") with good coverage of phrasing variations.	3 / 3
Distinctiveness Conflict Risk	Targets a clear niche (authoring Agent Skills for Cursor) with distinct triggers unlikely to fire for unrelated 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
relative_links	Relative link issues: 4 missing	Warning
referenced_paths_exist	Referenced path issues: 5 missing	Warning

	Total	14 / 16 Passed

Repository: xstongxue/best-skills
Path: skills/skill-create/SKILL.md
Commit: fd28854

Reviewed: 5 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.