Agent skill

skill-creator

Use PROACTIVELY when creating new Claude Code skills from scratch. Automated generation tool following Claudex marketplace standards with intelligent templates, pattern detection, and quality validation. Supports guided creation, quick start templates, clone-and-modify, and validation-only modes. Not for modifying existing skills or non-skill Claude Code configurations.

Stars 4
Forks 2

Install this agent skill to your Project

npx add-skill https://github.com/cskiro/claudex/tree/main/plugins/skill-creator/skills/skill-creator

SKILL.md

Skill Creator

Automates creation of Claude Code skills through interactive guidance, template generation, and quality validation.

When to Use

Trigger Phrases:

  • "create a new skill for [purpose]"
  • "generate a skill called [name]"
  • "scaffold a [type] skill"
  • "set up a new skill"

Use Cases:

  • Creating new skills from scratch
  • Following Claudex marketplace standards
  • Learning skill structure through examples

Quick Decision Matrix

User Request Mode Action
"create skill for [purpose]" Guided Interactive creation
"create [type] skill" Quick Start Template-based
"skill like [existing]" Clone Copy pattern
"validate skill" Validate Quality check

Mode 1: Guided Creation (Default)

Use when: User wants full guidance and customization

Process:

  1. Gather basic info (name, description, author)
  2. Define purpose, category, triggers
  3. Assess complexity → determine skill type
  4. Customize directory structure
  5. Select pattern (mode-based, phase-based, validation, data-processing)
  6. Generate files from templates
  7. Run quality validation
  8. Provide installation and next steps

Workflow: workflow/guided-creation.md

Mode 2: Quick Start

Use when: User specifies skill type directly (minimal, standard, complex)

Process:

  1. Confirm skill type
  2. Gather minimal required info
  3. Generate with standardized defaults
  4. Flag ALL customization points

Advantages: Fast, minimal questions Trade-off: More TODO sections to customize

Mode 3: Clone & Modify

Use when: User wants to base skill on existing one

Process:

  1. Read existing skill's structure
  2. Extract organizational pattern (not content)
  3. Generate new skill with same structure
  4. Clear example-specific content

Advantages: Proven structure, familiar patterns

Mode 4: Validation Only

Use when: User wants to check existing skill quality

Process:

  1. Read existing skill files
  2. Run quality checklist
  3. Generate validation report
  4. Offer to fix issues automatically

Use Case: Before submission, after modifications

Skill Types

Type Complexity Directories Pattern
Minimal Low SKILL.md, README.md only phase-based
Standard Medium + data/, examples/ phase-based or validation
Complex (mode) High + modes/, templates/ mode-based
Complex (data) High + scripts/, data/ data-processing

Generated Files

Required (all skills):

  • SKILL.md - Main skill manifest (with YAML frontmatter)
  • README.md - User documentation
  • CHANGELOG.md - Version history

Optional (based on type):

  • modes/ - Mode-specific workflows
  • data/ - Reference materials
  • examples/ - Example outputs
  • templates/ - Reusable templates
  • scripts/ - Automation scripts

Note: plugin.json is NOT required. The marketplace.json is the single source of truth for plugin metadata.

Quality Validation

Validates against data/quality-checklist.md:

  • File existence (all required files)
  • Syntax (YAML frontmatter, JSON)
  • Content completeness
  • Security (no secrets)
  • Naming conventions (kebab-case)
  • Quality grade (A-F)

Success Criteria

  • All required files generated (SKILL.md, README.md, CHANGELOG.md)
  • Valid YAML frontmatter with name and description
  • name matches directory name (Anthropic spec requirement)
  • No security issues (no secrets in files)
  • Kebab-case naming (lowercase + hyphens only)
  • Version 0.1.0 for new skills
  • Description includes capabilities AND trigger context
  • Quality grade C or better

Reference Materials

Templates

  • templates/SKILL.md.j2 - Main manifest with frontmatter
  • templates/README.md.j2 - User documentation
  • templates/CHANGELOG.md.j2 - Version history

Patterns

  • patterns/mode-based.md - Multi-mode skills
  • patterns/phase-based.md - Sequential workflows
  • patterns/validation.md - Audit skills
  • patterns/data-processing.md - Data analysis

Reference Data

  • data/categories.yaml - Valid categories
  • data/skill-types.yaml - Type definitions
  • data/quality-checklist.md - Validation criteria

Examples

  • examples/minimal-skill/
  • examples/standard-skill/
  • examples/complex-skill/

Quick Commands

bash
# Check existing skills
ls ~/.claude/skills/

# View skill structure
tree ~/.claude/skills/[skill-name]/

# Validate frontmatter syntax
head -20 ~/.claude/skills/[skill-name]/SKILL.md

# Run marketplace validation
python3 scripts/validate-skills.py

Error Handling

Error Solution
Name exists Suggest alternatives or confirm overwrite
Invalid name Explain kebab-case, provide corrected suggestion
Permission denied Check ~/.claude/skills/ write access
Template fails Fallback to manual creation with guidance

Version: 0.1.0 | Author: Connor

Expand your agent's capabilities with these related and highly-rated skills.

cskiro/claudex

e2e-testing

Use PROACTIVELY when setting up end-to-end testing, debugging UI issues, creating visual regression suites, or automating browser testing. Uses Playwright with LLM-powered visual analysis, screenshot capture, and fix recommendations. Zero-setup for React, Next.js, Vue, Node.js, and static sites. Not for unit testing, API-only testing, or mobile native apps.

4 2
Explore
cskiro/claudex

github-repo-setup

Use PROACTIVELY when user needs to create a new GitHub repository or set up a project with best practices. Automates repository creation with four modes - quick public repos (~30s), enterprise-grade with security and CI/CD (~120s), open-source community standards (~90s), and private team collaboration with governance (~90s). Not for existing repo configuration or GitHub Actions workflow debugging.

4 2
Explore
cskiro/claudex

sub-agent-creator

Use PROACTIVELY when creating specialized Claude Code sub-agents for task delegation. Automates agent creation following Anthropic's official patterns with proper frontmatter, tool configuration, and system prompts. Generates domain-specific agents, proactive auto-triggering agents, and security-sensitive agents with limited tools. Not for modifying existing agents or general prompt engineering.

4 2
Explore
cskiro/claudex

accessibility-audit

Use PROACTIVELY when user asks for accessibility review, a11y audit, WCAG compliance check, screen reader testing, keyboard navigation validation, or color contrast analysis. Audits React/TypeScript applications for WCAG 2.2 Level AA compliance with risk-based severity scoring. Includes MUI framework awareness to avoid false positives. Not for runtime accessibility testing in production, automated remediation, or non-React frameworks.

4 2
Explore
cskiro/claudex

otel-monitoring-setup

Use PROACTIVELY when setting up OpenTelemetry monitoring for Claude Code usage tracking, cost analysis, or productivity metrics. Provides local PoC mode (full Docker stack with Grafana) and enterprise mode (centralized infrastructure). Configures telemetry collection, imports dashboards, and verifies data flow. Not for non-Claude telemetry or custom metric definitions.

4 2
Explore
cskiro/claudex

json-outputs-implementer

Use PROACTIVELY when extracting structured data from text/images, classifying content, or formatting API responses with guaranteed schema compliance. Implements Anthropic's JSON outputs mode with Pydantic/Zod SDK integration. Covers schema design, validation, testing, and production optimization. Not for tool parameter validation or agentic workflows (use strict-tool-implementer instead).

4 2
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results