skill-creator

This skill should be used when the user asks to create a new skill, build a skill, make a custom skill, develop a CLI skill, or wants to extend the CLI with new capabilities. Automates the entire skill creation workflow from brainstorming to installation.

Author

Eric Andrade

Install

Hot:12

Download and extract to your skills directory

Copy command and send to OpenClaw for auto-install:

Download and install this skill https://openskills.cc/api/download?slug=sickn33-skills-skill-creator&locale=en&source=copy

Skill Creator - CLI Skill Automation Creation Tool

Skill Overview


Skill Creator is a zero-configuration CLI skill scaffolding tool that automates the entire skill creation workflow—from brainstorming to installation—strictly following Anthropic’s official best practices.

Use Cases

1. Extending CLI Functionality


When you need to add custom capabilities to platforms such as GitHub Copilot CLI, Claude Code, or Codex, skill-creator can quickly generate a skill file structure that complies with the platform’s requirements, saving you the tedious steps of manually creating directories, writing metadata, and configuring templates.

2. Wrapping Automated Tasks


Encapsulate repetitive development workflows, debugging steps, or code review processes into reusable skills. The tool supports interactive brainstorming to help you clarify the skill’s purpose, trigger phrase type, and applicable scope.

3. Packaging Domain Knowledge


Package specialized knowledge for a particular domain (e.g., debugging techniques, framework conventions, deployment processes) into a team-shared skill library. Supports both local repository installation and global symbolic linking, making it easy for teams and individuals to collaborate and use.

Core Features

1. Interactive Skill Generation


Collect skill requirements through guided Q&A, including function description, trigger phrases, skill type (general/code/documentation/analysis), and target platform. Optionally integrate prompt-engineer skills to optimize prompts, ensuring skill descriptions are clear and precise.

2. Template-Based File Creation


Automatically generate standardized directory structures, including a SKILL.md main file, a README.md user document, a references extended documentation directory, an examples directory, and a scripts tool scripts directory. Supports automatic template variable substitution (skill name, author, date, version, etc.).

3. Standards Validation and Installation


Includes built-in YAML front-matter data validation, content quality checks (word count, writing style), and skill specification checks. After validation passes, it supports local repository installation, global symbolic linking installation, or both, and provides a detailed completion report with follow-up recommendations.

FAQs

Which platforms does Skill Creator support?


It supports three platforms: GitHub Copilot CLI, Claude Code, and Codex. The tool automatically detects which platforms are already installed on the system, and can generate the corresponding skill files for one or multiple selected platforms. Each platform’s skill directory location differs (.github/skills/, .claude/skills/, .codex/skills/), and the tool automatically adapts.

Is there a word limit for the generated skills?


Yes. The SKILL.md main file is recommended to stay within 1,500–2,000 words, with a maximum not exceeding 5,000 words. The README.md user documentation is recommended to be 300–500 words. When content is too long, detailed documentation should be moved to the references/ directory. The tool will validate the word count and provide prompts to help you follow the progressive disclosure principle and keep the skill’s core information concise.

How can I verify whether a skill meets the specifications?


The tool includes a built-in validation script that checks YAML front-matter completeness, description formatting (must use third-person), word count limits, writing style (disables second-person), and required fields. Installation is allowed only after validation passes. Common issues such as description format errors or word limit violations can be automatically fixed, while more complex problems will include detailed guidance.