skill-create
Skill2 starsChanged 9 months agoMITView on GitHub
Create and validate Claude Code skills with proper structure, checklists, and validation scripts. Use when creating a new skill, auditing existing skills, or ensuring skills meet quality standards. Generates complete skill scaffolds with SKILL.md, check scripts, guides, and templates.
We split this skill into the rules it gives an agent
15 rules. 0 are common, 12 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 12 edited versions we read keep these.
12
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Create complete, well-structured Claude Code skills.
- Very specific: line 82
Use template from templates/SKILL.md.template:
- Distinctive: line 84
Write frontmatter (name, description, allowed-tools)
- Distinctive: line 85
Add "When to Use" section with clear scenarios
- Distinctive: line 87
Add guides table linking to detailed content
- Distinctive: line 89
Add "Related Skills" section
- Distinctive: line 124
Add to "Complete Skill Reference" table
- Distinctive: line 125
Add to skill flow diagrams
- Distinctive: line 126
Add to delegation rules if needed
- Distinctive: line 228
All skills - This skill helps create and validate them
- Distinctive: line 233
After creating a skill:
- Very specific: line 234
Run validate.dart --skill {name} to check quality gates
- Very specific: line 235
Update SKILL_STRUCTURE.md to register the skill
- Distinctive: line 236
Test the skill with representative tasks
- Distinctive: line 237
Verify Claude finds reference files correctly
“Common” is measured: kept by at least 6 in 10 of the 12 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Skill Creation & Validation
Create complete, well-structured Claude Code skills. Generates scaffolds, validates skill completeness, and ensures quality standards.
When to Use This Skill
- Creating a new skill from scratch
- Auditing existing skills for completeness
- Validating a skill before deployment
- Understanding skill best practices
- User asks "create skill", "new skill", or "validate skill"
Questions to Ask
- Skill category: Code-heavy, Auth, Config, Knowledge, or Orchestration?
- Core purpose: What problem does this skill solve? (1-2 sentences)
- Trigger words: What keywords should invoke this skill?
- Reference code: Will it include Dart templates, or just guides?
- Validation needs: What should the check script verify?
Quick Reference
Skill Categories
| Category | Examples | Focus |
| Code-Heavy | domain, data, presentation | Generate detailed code from specs |
| Auth | auth, phone-auth, social-login | Authentication with platform setup |
| Config | release, ci-cd, analytics | Platform configs and workflows |
| Knowledge | design, i18n, testing, a11y | Guidelines and patterns |
| Orchestration | plan, init, feature-init | Coordinate other skills |
Directory Structure
.claude/skills/{skill-name}/
├── SKILL.md # Main entry (<300 lines)
├── {topic}-guide.md # Detailed guides (reference/ for code)
├── templates/ # Platform configs (XML, YAML)
└── scripts/check.dart # Validation scriptSKILL.md Requirements
| Section | Required | Purpose |
| YAML frontmatter | YES | name, description, allowed-tools |
| When to Use | YES | Clear trigger scenarios |
| Workflow | YES | Numbered implementation steps |
| Guides table | YES | Links to detailed content |
| Checklist | YES | Verification items |
| Related Skills | YES | Cross-references |
| Quick Reference | Recommended | Decision tables, key patterns |
Workflow
Phase 1: Define Skill
- Determine skill category and core purpose
- Identify trigger words for description
- List what the skill should NOT handle (delegate to other skills)
- Define expected outputs (files, code, configs)
Phase 2: Create Structure
# Create skill directory
mkdir -p .claude/skills/{skill-name}/{scripts,templates,reference}
# Validate structure
dart run .claude/skills/skill-create/scripts/validate.dart --skill {skill-name}Phase 3: Write SKILL.md
Use template from templates/SKILL.md.template:
- Write frontmatter (name, description, allowed-tools)
- Add "When to Use" section with clear scenarios
- Define workflow with numbered steps
- Add guides table linking to detailed content
- Include checklist for verification
- Add "Related Skills" section
Phase 4: Create Supporting Files
Based on category:
Code-Heavy Skills:
reference/- Full Dart implementationsscripts/check.dart- Validates generated code
Config Skills:
templates/- Platform config files (XML, YAML, properties){topic}-guide.md- Setup instructionsscripts/check.dart- Validates config presence
Knowledge Skills:
{topic}-guide.md- Best practices and patternsexamples.md- Code examplesscripts/check.dart- Audits code for issues
Phase 5: Validate
# Run full validation
dart run .claude/skills/skill-create/scripts/validate.dart --skill {skill-name}
# Check specific aspects
dart run .claude/skills/skill-create/scripts/validate.dart --skill {skill-name} --check frontmatter
dart run .claude/skills/skill-create/scripts/validate.dart --skill {skill-name} --check structure
dart run .claude/skills/skill-create/scripts/validate.dart --skill {skill-name} --check referencesPhase 6: Register
Update SKILL_STRUCTURE.md:
- Add to "Complete Skill Reference" table
- Add to skill flow diagrams
- Add to delegation rules if needed
Commands
dart run .claude/skills/skill-create/scripts/validate.dart --skill {name} # Validate
dart run .claude/skills/skill-create/scripts/validate.dart --all # Validate all
dart run .claude/skills/skill-create/scripts/validate.dart --list-checks # List checks
dart run .claude/skills/skill-create/scripts/validate.dart --skill {name} --report # DetailedQuality Gates
A skill must pass these gates before deployment:
Gate 1: Structure (Required)
- SKILL.md exists
- SKILL.md < 300 lines
- Valid YAML frontmatter
- Required sections present
Gate 2: Content (Required)
- Description < 1024 chars
- Description in third person
- Workflow has numbered steps
- Checklist has > 3 items
- All internal links resolve
Gate 3: Quality (Recommended)
- No code blocks > 10 lines in SKILL.md
- Reference files have header comments
- Check script exists and runs
- Related Skills section accurate
Gate 4: Integration (Recommended)
- Registered in SKILL_STRUCTURE.md
- Cross-references to/from other skills accurate
- No duplicate functionality with existing skills
See: quality-gates.md for detailed criteria.
Guides
| File | Content |
| checklist.md | Comprehensive skill creation checklist |
| quality-gates.md | Detailed quality criteria |
| examples.md | Well-structured skill examples |
Templates
| Template | Purpose |
templates/SKILL.md.template | Starting point for SKILL.md |
templates/check.dart.template | Validation script scaffold |
templates/guide.md.template | Guide file structure |
Anti-Patterns
| Pattern | Problem | Solution |
| Long SKILL.md | 500+ lines with code | < 200 lines, move code to reference/ |
| Vague description | "Helps with data stuff" | Specific: "Generate DTOs, repositories..." |
| Missing delegation | Inline i18n, testing, a11y | Use "Run /i18n" in Next Steps |
| No validation | No way to verify | Add scripts/check.dart |
See: examples.md for well-structured skill examples.
Checklist
Structure:
- Directory follows standard structure
- SKILL.md exists with valid frontmatter
- SKILL.md < 300 lines (ideally < 200)
- All internal links resolve
Content:
- "When to Use" section with clear scenarios
- Workflow with numbered steps
- Guides table with linked files
- Checklist with verification items
- Related Skills section
Quality:
- Description in third person
- Description includes trigger words
- No code blocks > 10 lines in SKILL.md
- Reference files have header comments
- Check script validates key requirements
Integration:
- Added to SKILL_STRUCTURE.md skill table
- Added to skill flow diagrams
- Cross-references accurate
- Delegation rules documented
Related Skills
- All skills - This skill helps create and validate them
- SKILL_STRUCTURE.md - Central documentation this skill supports
Next Steps
After creating a skill:
- Run
validate.dart --skill {name}to check quality gates - Update SKILL_STRUCTURE.md to register the skill
- Test the skill with representative tasks
- Verify Claude finds reference files correctly
Other versions13 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 13
- popup-studio-ai/bkit-claude-code★ 601same name
- ww-w-ai/bkit-claude-code★ 599same name
- ww-w-ai/bkit-gemini★ 66same name
- joris887/exosuit★ 7same name
- PatriotAgenticLLC/better_claude_skillz★ 3same name
- dfrysinger/skills★ 2same name
- risadams/ink-and-agency★ 2same name
- Michellog/skills★ 1same name
- TheKREATORS/my-claude-skills★ 1same name
- Marco-Satis/claude-setup★ 0same name
3 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
Used alongsideProjects with it also use pm-discovery, qa-phase and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
ww-w-aiSkillpm-discovery
PM Agent Team - Automated product discovery, strategy, and PRD generation. Runs PM analysis workflow to produce comprehensive PRD before PDCA Plan phase.
Also in 3 projects that use this
ww-w-aiSkillqa-phase
Run QA tests at various levels (L1-L5) for features under PDCA management. Integrates with MCP tools to execute structured test plans.
Also in 3 projects that use this
- +10 more from ww-w-ai
In this repository27 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- a11ySkill
- account-deletionSkill
- analyticsSkill
- authSkill
- ci-cdSkill
- coreSkill
- dataSkill
- deep-linkingSkill
- designSkill
- domainSkill
- feature-initSkill
- flavorsSkill
- force-updateSkill
- i18nSkill
- in-app-purchasesSkill
- initSkill
- local-authSkill
- network-connectivitySkill
- offlineSkill
- phone-authSkill
- planSkill
- presentationSkill
- push-notificationsSkill
- releaseSkill
- social-loginSkill
- testingSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

