[{"TODO":"lowercase-hyphen-case-name"}]
About
This skill provides a template for creating new Claude Skills, with placeholders for descriptions, use cases, and technical specifications. It should be used when developers need to structure and document custom skills for Claude Code with clear implementation guidelines. The template includes sections for setup instructions, dependencies, and configuration examples.
Quick Install
Claude Code
Recommended/plugin add https://github.com/jezweb/claude-skillsgit clone https://github.com/jezweb/claude-skills.git ~/.claude/skills/[{"TODO":"lowercase-hyphen-case-name"}]Copy and paste this command in Claude Code to install this skill
Documentation
[TODO: Skill Display Name]
Status: [TODO: Production Ready / Beta / Experimental] Last Updated: [TODO: YYYY-MM-DD] Dependencies: [TODO: List prerequisite skills or tools, or write "None"] Latest Versions: [TODO: package@version, package2@version]
Quick Start ([TODO: X] Minutes)
1. [TODO: First Major Step]
[TODO: Provide clear, actionable instructions]
# [TODO: Example command]
Why this matters:
- [TODO: Explain the reasoning]
- [TODO: Key benefits]
2. [TODO: Second Major Step]
[TODO: Continue with step-by-step instructions]
// [TODO: Code example with comments]
CRITICAL:
- [TODO: Important notes that prevent common mistakes]
3. [TODO: Third Major Step]
[TODO: Keep going until setup is complete]
The [TODO: Number]-Step Setup Process
Step 1: [TODO: First Step Name]
[TODO: Detailed instructions]
[TODO: Code examples, configurations, etc.]
Key Points:
- [TODO: Important details]
- [TODO: What to avoid]
Step 2: [TODO: Second Step Name]
[TODO: Continue with detailed steps]
Critical Rules
Always Do
✅ [TODO: Best practice #1] ✅ [TODO: Best practice #2] ✅ [TODO: Best practice #3]
Never Do
❌ [TODO: Anti-pattern #1] ❌ [TODO: Anti-pattern #2] ❌ [TODO: Anti-pattern #3]
Known Issues Prevention
This skill prevents [TODO: number] documented issues:
Issue #1: [TODO: Issue Name]
Error: [TODO: Exact error message or description] Source: [TODO: GitHub issue link, Stack Overflow link, or official docs] Why It Happens: [TODO: Root cause] Prevention: [TODO: How this skill fixes it]
Issue #2: [TODO: Issue Name]
[TODO: Repeat pattern for each issue]
Configuration Files Reference
[TODO: config-file.ext] (Full Example)
[TODO: Complete, working configuration file]
Why these settings:
- [TODO: Explain key configuration decisions]
Common Patterns
Pattern 1: [TODO: Pattern Name]
// [TODO: Code example]
When to use: [TODO: Specific scenarios]
Pattern 2: [TODO: Pattern Name]
[TODO: More examples]
Using Bundled Resources
[TODO: Only include this section if you have scripts/, references/, or assets/]
Scripts (scripts/)
[TODO: List and explain any executable scripts]
Example Usage:
./scripts/[TODO: script-name].sh
References (references/)
[TODO: List documentation files that Claude can load when needed]
references/[TODO: filename].md- [TODO: What it contains]
When Claude should load these: [TODO: Specific scenarios]
Assets (assets/)
[TODO: List template files, images, etc. that get used in output]
assets/[TODO: filename]- [TODO: What it's for]
Advanced Topics
[TODO: Advanced Topic 1]
[TODO: Deep dive into complex scenarios]
[TODO: Advanced Topic 2]
[TODO: Additional advanced content]
Dependencies
Required:
- [TODO: package@version] - [TODO: What it's used for]
- [TODO: package@version] - [TODO: What it's used for]
Optional:
- [TODO: package@version] - [TODO: What it's used for]
Official Documentation
- [TODO: Technology Name]: [TODO: Official docs URL]
- [TODO: Related Tool]: [TODO: Official docs URL]
- Context7 Library ID: [TODO: /org/project if applicable]
Package Versions (Verified [TODO: YYYY-MM-DD])
{
"dependencies": {
"[TODO: package]": "^[TODO: version]"
},
"devDependencies": {
"[TODO: package]": "^[TODO: version]"
}
}
Production Example
[TODO: Optional - provide evidence of production usage]
This skill is based on [TODO: project name]:
- Live: [TODO: URL if applicable]
- Build Time: [TODO: time]
- Errors: 0 (all [TODO: number] known issues prevented)
- Validation: ✅ [TODO: What was tested]
Troubleshooting
Problem: [TODO: Common issue]
Solution: [TODO: How to fix]
Problem: [TODO: Another issue]
Solution: [TODO: How to fix]
Complete Setup Checklist
Use this checklist to verify your setup:
- [TODO: First verification step]
- [TODO: Second verification step]
- [TODO: Third verification step]
- [TODO: All package versions correct]
- [TODO: Dev server runs without errors]
- [TODO: Production build succeeds]
- [TODO: Deployed successfully (if applicable)]
Questions? Issues?
- Check [TODO: reference/common-issues.md] if you created one
- Verify all steps in the setup process
- Check official docs: [TODO: URL]
- Ensure [TODO: critical requirement] is configured
INSTRUCTIONS FOR USING THIS TEMPLATE
- Search for [TODO:** - Replace every [TODO: ...] with actual content
- Delete this section when done
- Remove sections that don't apply to your skill
- Add sections as needed for your specific use case
- Test everything before committing
- Verify against ONE_PAGE_CHECKLIST.md
Key Reminders:
- Use third-person for description: "This skill provides..." not "This skill helps you..."
- Write in imperative form: "To do X, run Y" not "You should run Y to do X"
- Include actual error messages and GitHub issue links
- Document all package versions with verification date
- Test all code examples
- Remove all [TODO:] markers before committing
Example Good Description:
description: |
This skill provides comprehensive knowledge for building React applications with
Vite and TypeScript. It should be used when scaffolding new Vite projects,
configuring TypeScript settings, or encountering HMR issues.
Keywords: vite, react, typescript, hmr, vite config, react setup
Example Bad Description:
description: "A skill for Vite and React" # ❌ Too vague, no keywords
Ready to build? Fill in all [TODO:] markers and create an amazing skill!
GitHub Repository
Related Skills
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
evaluating-llms-harness
TestingThis Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.
llamaguard
OtherLlamaGuard is Meta's 7-8B parameter model for moderating LLM inputs and outputs across six safety categories like violence and hate speech. It offers 94-95% accuracy and can be deployed using vLLM, Hugging Face, or Amazon SageMaker. Use this skill to easily integrate content filtering and safety guardrails into your AI applications.
langchain
MetaLangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.
