create-skill
Skill24 starsChanged 9 months agoMITView on GitHub
Create or update Claude Code skills following Anthropic's official skill standard. Use when user says "create a skill", "new skill", "write a skill", "update a skill", "skill standard", or any time a new SKILL.md file needs to be created. Also use when reviewing existing skills for standards compliance.
144 other projects use a version of this.
Adopted by
Most recently taken up by
- neomjs/neo-agent-skillstheir own version, first seen Nov 15, 2026
- shishtpal/SpiAgentHarnesstheir own version, first seen Nov 15, 2026
- shoodoow/kurutheir own version, first seen Nov 15, 2026
- CherryHQ/cherry-studiotheir own version, first seen Nov 15, 2026
- xSolutions365/skillstheir own version, first seen Nov 15, 2026
- lancy69/skillstheir own version, first seen Nov 8, 2026
Used with
Projects with it also use review-bugbot, migrate-to-skills 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.
0mlkrizzz655335vSkillreview-bugbot
缺陷导向代码审查。用 Bugbot 子代理找 bug。
Also in 5 projects that use this
32103210Skillmigrate-to-skills
Convert 'Applied intelligently' Cursor rules (.cursor/rules/*.mdc) and slash commands (.cursor/commands/*.md) to Agent Skills format (.cursor/skills/). Use when you want to migrate rules or commands to skills, convert .mdc rules to SKILL.md format, or consolidate commands into the skills directory.
Also in 8 projects that use this
32103210Skillupdate-cursor-settings
Modify Cursor/VSCode user settings in settings.json. Use when you want to change editor settings, preferences, configuration, themes, font size, tab size, format on save, auto save, keybindings, or any settings.json values.
Also in 8 projects that use this
PandelisZSkillupdate-cli-config
View and modify Cursor CLI configuration settings in ~/.cursor/cli-config.json. Use when the user wants to change CLI settings, configure permissions, switch approval mode, enable vim mode, toggle display options, configure sandbox, or manage any CLI preferences.
Also in 7 projects that use this
yawningwhaleAISkillbuild-premium-website
Build a premium, animated marketing website (React + Vite + Tailwind + GSAP) for any industry. Use when the user asks to "build a website", "make a landing page for [business]", "build a marketing site", or wants a high-end animated company website. Adapts color, copy, services, and signature animation to the industry.
Also in 6 projects that use this
yawningwhaleAISkillinstantly-campaign
Create a new Instantly cold-email campaign in the proven 5-email structure of the user's top-performing campaign (~16% reply rate). Use when the user wants to launch a new outbound campaign, write a cold email sequence, create an Instantly campaign, draft outreach for a new service or niche, or set up a sales sequence.
Also in 5 projects that use this
- +1 more from 32103210
Other versions144 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 144
- microsoft/vscode★ 195ksame name
- CherryHQ/cherry-studio★ 54ksame name
- videojs/video.js★ 40ksame name
- BuilderIO/agent-native★ 7.7ksame name
- stagewise-io/stagewise★ 6.8ksame name
- dotnet/skills★ 6.2ksame name
- xstongxue/best-skills★ 2.9ksame name
- Lilac-Labs/gini-agent★ 1.9ksame name
- Open-Curiosity/gini-agent★ 1.9ksame name
- mattmireles/gemma-tuner-multimodal★ 1.5ksame name
134 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.
In this repository11 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- sync-skillsSkill
- better-waySkill
- confluence-to-mdSkill
- diagram-generatorSkill
- draft-emailSkill
- jira-ticket-creatorSkill
- md-to-confluenceSkill
- presentation-builderSkill
- teach-meSkill
- youtube-transcriptSkill
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 4 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 134 edited versions we read keep these.
4
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
- Very specific: line 30
Keep SKILL.md under 500 lines.
- Distinctive: line 60
Every skill body should include these components:
- Distinctive: line 102
Write instructions as commands: "Open with a personal admission" not "You should open with a personal admission."
- Distinctive: line 105
Write instructions that work across varied inputs.
- Very specific: line 117
Use init_skill.py to create the folder and register in skills.json in one step:
- Very specific: line 134
When creating skills for the claude-skills public repo, replace all personal/project references:
- Distinctive: line 152
Before finalizing any skill:
“Common” is measured: kept by at least 6 in 10 of the 134 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 Guide
Reference for creating skills that follow the Anthropic skill standard (github.com/anthropics/skills).
Skill Anatomy
skill-name/
├── SKILL.md (required)
│ ├── YAML frontmatter (name, description required)
│ └── Markdown instructions (<500 lines)
└── Bundled Resources (optional)
├── scripts/ - Executable code for deterministic tasks
├── references/ - Docs loaded into context as needed
└── assets/ - Templates, icons, fonts for outputsProgressive Disclosure
Skills load in three layers:
- Metadata (name + description) - Always in context (~100 words)
- SKILL.md body - Loaded when skill triggers (<500 lines)
- Bundled resources - Loaded as needed (unlimited size)
Keep SKILL.md under 500 lines. If approaching the limit, move content to references/ with clear pointers about when to read each file.
YAML Frontmatter
Two required fields:
---
name: skill-name
description: What the skill does AND when to trigger it. All "when to use" context goes here, not in the body.
---Description is the Trigger
The description field is the primary mechanism for skill invocation. Claude tends to undertrigger skills, so descriptions should be slightly pushy.
Weak: "Generate LinkedIn posts for professional audiences."
Strong: "Write high-performing LinkedIn posts. Use when user says 'write a LinkedIn post', 'LinkedIn content', 'post for LinkedIn', 'professional audience post', or wants to share insights, lessons, or updates with a professional network - even if they don't explicitly say LinkedIn."
Include:
- What the skill does (1 sentence)
- Explicit trigger phrases users might say
- Adjacent contexts where the skill applies ("even if they don't explicitly say X")
Body Structure
Every skill body should include these components:
1. Overview (optional, 1-2 paragraphs)
What this skill does and the critical distinction that shapes how it works. Name the book or methodology the skill is grounded in.
2. Workflow (required)
Step-by-step process. Use ### Step N: headings. Each step should be actionable and specific.
3. Anti-Patterns (required)
What NOT to do. Format as Bad/Good pairs so the model can calibrate:
## Anti-Patterns
**1. Descriptive name**
Bad: "example of wrong approach"
Good: "example of correct approach"4. Quality Checklist (required)
Checkbox list the model uses to verify output before delivering:
## Quality Checklist
- [ ] Check 1
- [ ] Check 2Writing Style
Explain the "Why" Instead of Bare Directives
The model follows instructions better when it understands the reasoning. Heavy-handed MUSTs without context are less effective than explaining why something matters.
Weak: "Never exceed 5 bullets per slide."
Strong: "Keep slides to 3-5 bullets. Beyond 5, the audience reads instead of listening - the slide competes with the speaker instead of supporting them."
Use Imperative Form
Write instructions as commands: "Open with a personal admission" not "You should open with a personal admission."
Generalize, Don't Overfit
Write instructions that work across varied inputs. Avoid narrow tweaks targeting only specific examples - look for underlying patterns.
Examples Are Valuable
Include concrete examples. Format with Input/Output or Bad/Good pairs. Examples calibrate the model's output better than abstract rules.
Ground Every Skill in a Book or Methodology
Skills grounded in a proven framework (book, methodology, model) are more credible, consistent, and unique. Name the source in the Overview section.
Good examples: Radical Candor (feedback), The Pyramid Principle (exec summaries), SPIN Selling (discovery calls), How Google Tests Software (risk-based testing).
Scaffolding a New Skill
Use init_skill.py to create the folder and register in skills.json in one step:
python3 scripts/init_skill.py <skill-name> --role <role> --description "<one-liner>" --tags "<tag1,tag2>"Valid roles: pm, engineer, qa, designer, content-creator, devops, shared, leadership, program-delivery-manager, customer-success, recruiter, consultant, sales, marketing, founder, data-scientist, data-engineer, security
This creates skills/by-role/<role>/<skill-name>/SKILL.md and adds an entry to skills.json. Always scaffold first, then write the SKILL.md content.
After writing, validate:
python3 scripts/validate_skill.pyPlaceholders (for this repo)
When creating skills for the claude-skills public repo, replace all personal/project references:
| Context | Placeholder |
| Personal name | [your name] |
| Newsletter or blog | [your newsletter] |
| Website domain | [your-domain] |
| Professional niche | [your niche] |
| Subject area | [your topic] |
| Target audience | [your audience] |
| Company name | [your company] |
| Team name | [your team] |
| Client name | [your client] |
| Customer name | [customer name] |
| Program or project | [your program] |
Creation Checklist
Before finalizing any skill:
- Scaffolded with
init_skill.py(creates folder + registers in skills.json) - YAML frontmatter has
nameanddescription - Description is pushy with multiple trigger phrases
- Body is under 500 lines
- Overview names the book or methodology the skill is grounded in
- Workflow has clear numbered steps with
### Step N:headings - Anti-patterns section exists with Bad/Good pairs
- Quality checklist exists
- Directives explain "why", not just "what"
- Examples are concrete and specific
- No personal/project-specific content (if for public repo)
- Placeholder brackets used for any context-specific references
- Overflow content moved to
references/with clear pointers python3 scripts/validate_skill.pypasses
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
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.

