Agent skill
skill-creator
Create new Claude Code skills with proper structure, documentation, and best practices. Use when building custom skills for specific domains, workflows, or organizational needs.
Install this agent skill to your Project
npx add-skill https://github.com/vamseeachanta/workspace-hub/tree/main/.claude/skills/_internal/builders/skill-creator
SKILL.md
Skill Creator
Overview
This skill guides the creation of new Claude Code skills. Skills are specialized instruction sets that enhance Claude's capabilities for specific domains, tasks, or workflows.
When to Use
- Building custom skills for specific domains
- Creating reusable workflow templates
- Standardizing organizational processes
- Extending Claude Code capabilities
- Documenting specialized knowledge
Quick Start
- Define scope - Answer: What problem? Who uses it? What outputs?
- Create structure -
.claude/skills/skill-name/SKILL.md - Write frontmatter - Name, description, version, category
- Add content - Overview, Instructions, Examples, Best Practices
- Test - Verify skill triggers correctly
# Create skill directory
mkdir -p .claude/skills/my-new-skill
# Create SKILL.md with template
cat > .claude/skills/my-new-skill/SKILL.md << 'EOF'
---
name: my-new-skill
description: Action-oriented description. Use for X, Y, and Z.
version: 1.0.0
category: [builders|tools|content-design|communication|meta]
tags: [topic-keyword, tool-name, use-case] # discovery tags for search/filtering
related_skills: [companion-skill-name] # cross-references to sibling skills
freedom: medium # high = open-ended (brainstorming, research); medium = structured output (reports, templates); low = strict procedure (deploy, migrate, verify)
---
# My New Skill
## Overview
[1-2 sentences explaining purpose]
## When to Use
- Scenario 1
- Scenario 2
## Prerequisites
- Dependency 1
- Dependency 2
## Version History
- **1.0.0** (YYYY-MM-DD): Initial release
EOF
Prerequisites
- Familiarity with YAML frontmatter
- Understanding of markdown structure
- Knowledge of the skill category taxonomy
Deprecation
When retiring a skill:
---
name: old-skill-name
description: DEPRECATED - Use new-skill-name instead. [Original description]
deprecated: true
deprecated_date: 2026-01-02
replacement: new-skill-name
---
Quality Tiers
Skills are classified into quality tiers (A/B/C/D) based on enforcement level:
- Tier A — Script-wired: frontmatter
scripts:list with >=1 entry - Tier B — Exec-pattern: body contains
bash scripts/,uv run, orbash .claude/skills/ - Tier D — Decomposition candidate: >500 words with no script refs (split recommended)
- Tier C — Focused prose: everything else
Target Tier A or B for new skills. See config/skills/quality-tiers.yaml.
Related Skills
- session-start-routine - Skill library maintenance
- sparc-workflow - Development methodology
- mcp-builder - MCP server creation
Version History
- 2.2.0 (2026-03-05): Deduplicated hub SKILL.md; trimmed to <200 lines
- 2.1.0 (2026-03-04): Synced critical rules with Anthropic official guide
- 2.0.0 (2026-01-02): Upgraded to v2 template
- 1.0.0 (2024-10-15): Initial release
Sub-Skills
- Official Plugin Reference
- Canonical Reference
- Instructions
- File Structure (+1)
- Error Handling
- Metrics
- 1. Clear Triggering (+1)
- Detailed Instructions
- 3. Multi-Stage Workflow Architecture (+1)
- Bad: Conceptual only
- Good: Actionable
- Step 1: Define the Skill Scope (+2)
- [Task Category 1] (+1)
- Technical Skill Template
- Installation
- Operation 1 (+1)
- Configuration Options
- Common Errors
- Step 1: [Name] (+2)
- [Template Name]
- Creative Skill Template
- Principle 1: [Name] (+1)
- Style 1: [Name] (+1)
- Phase 1: [Name] (+1)
- Technical Implementation
- Quality Checklist
- Inspiration & Examples
- Examples
- Best Practices
- Advanced Usage
- Execution Checklist
- Content Quality (+3)
- Versioning
- Skill Discovery (+2)
- Process Skill Template
- [Scenario]
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
gsd-complete-milestone
Archive completed milestone and prepare for next version
gsd-reapply-patches
Reapply local modifications after a GSD update
gsd-verify-work
Validate built features through conversational UAT
gsd-thread
Manage persistent context threads for cross-session work
clinical-trial-protocol
Generate clinical trial protocols for medical devices or drugs through a modular, waypoint-based architecture with research-only and full protocol modes.
single-cell-rna-qc
Performs quality control on single-cell RNA-seq data (.h5ad or .h5 files) using scverse best practices with MAD-based filtering and comprehensive visualizations.
Didn't find tool you were looking for?