playbooks

home / skills / duc01226 / easyplatform / skill-create

skill-create skill

/.claude/skills/skill-create

This skill scaffolds a new agent skill in .claude/skills, generating structure, docs, and starter tests for rapid deployment.

npx playbooks add skill duc01226/easyplatform --skill skill-create

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
2.0 KB
---
name: skill-create
description: "[Tooling & Meta] Create a new agent skill"
argument-hint: [prompt-or-llms-or-github-url]
infer: true
---

Use `skill-plan` and `claude-code` skills.
Use `docs-seeker` skills to search for documentation if needed.

## Your mission
Create a new skill in `.claude/skills/` directory.

## Requirements
<user-prompt>$ARGUMENTS</user-prompt>

## Rules of Skill Creation:
Base on the requirements:
- Always keep in mind that `SKILL.md` and reference files should be token consumption efficient, so that **progressive disclosure** can be leveraged at best.
- `SKILL.md` is always short and concise, straight to the point, treat it as a quick reference guide.
- If you're given nothing, use `AskUserQuestion` tool for clarifications and `researcher` subagent to research about the topic.
- If you're given an URL, it's documentation page, use `Explore` subagent to explore every internal link and report back to main agent, don't skip any link.
- If you receive a lot of URLs, use multiple `Explore` subagents to explore them in parallel, then report back to main agent.
- If you receive a lot of files, use multiple `Explore` subagents to explore them in parallel, then report back to main agent.
- If you're given a Github URL, use [`repomix`](https://repomix.com/guide/usage) command to summarize ([install it](https://repomix.com/guide/installation) if needed) and spawn multiple `Explore` subagents to explore it in parallel, then report back to main agent.

**IMPORTANT:**
- Skills are not documentation, they are practical instructions for Claude Code to use the tools, packages, plugins or APIs to achieve the tasks.
- Each skill teaches Claude how to perform a specific development task, not what a tool does.
- Claude Code can activate multiple skills automatically to achieve the user's request.

## IMPORTANT Task Planning Notes

- Always plan and break many small todo tasks
- Always add a final review todo task to review the works done at the end to find any fix or enhancement needed

Overview

This skill automates creating a new agent skill in the .claude/skills/ folder. It provides a repeatable, step-by-step workflow that plans the work, generates code artifacts, and performs a final review using the available subagents and tooling.

How this skill works

The skill uses planning and code-generation subagents to decompose the request into small tasks, generate the skill files, and run sanity checks. It leverages docs-seeker to fetch any required documentation, skill-plan to produce a prioritized task list, and claude-code to generate and refine code snippets and file contents. A final review task verifies structure, token-efficiency, and completeness.

When to use it

  • You need to add a new skill implementation to .claude/skills/ quickly and consistently.
  • The request includes requirements, URLs, or many files that need organized exploration.
  • You want automated planning, parallel exploration of multiple sources, and code generation.
  • You need to ensure the new skill follows token-efficiency and progressive disclosure patterns.
  • You need a final checklist and review to catch missing items before integration.

Best practices

  • Start by running the planner to break the job into small, testable tasks before generating code.
  • Use docs-seeker to fetch and traverse documentation links when an external URL is provided.
  • Generate minimal, concise files that favor progressive disclosure and low token usage.
  • Parallelize exploration tasks if there are many URLs or files to inspect.
  • Always include a final review task that validates structure, naming, and edge cases.

Example use cases

  • Create a new C# skill that implements a specific agent capability from a provided spec.
  • Ingest multiple documentation pages and synthesize a skill that wraps an external API.
  • Break a complex feature request into incremental tasks, generate files, and run a review.
  • Receive a GitHub repo URL and spawn parallel explorers to gather implementation details.

FAQ

What subagents are used?

This workflow uses skill-plan for task breakdown, docs-seeker for documentation discovery, and claude-code for code generation and refinement.

How are many URLs handled?

The skill spawns parallel exploration subagents to fetch and summarize each URL, then consolidates findings for code synthesis.

How does it ensure token-efficiency?

Files are generated concise by design, tasks emphasize progressive disclosure, and the final review enforces minimal, clear content.