# Skill To Skill

> Generate a complete, standards-compliant SKILL.md for a new Anthropic Skill based on a given capability description.

- Skill: `rinkokawa/skill-to-skill` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add rinkokawa/skill-to-skill`
- Raw SKILL.md: https://api.skillmd.com/api/skills/rinkokawa/skill-to-skill/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: RinKokawa (https://skillmd.com/u/rinkokawa)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/rinkokawa/skill-to-skill

---


# Skill Generator

## Overview

This skill is used to generate other **Anthropic-compatible Skills**.  
It produces a complete `SKILL.md` file that follows the official Anthropic Skills conventions, including metadata, usage guidance, and examples.

The output of this skill is **not executable code**, but a **Skill definition document** intended to be reviewed, saved, and registered by a developer.

---

## When to Use

Use this skill when:

- A new capability needs to be exposed as an Anthropic Skill
- You want to standardize Skill documentation across a project
- You are designing a Skill registry or Skill-based agent system
- A developer asks to “create a new Skill” or “define a Skill”

---

## When NOT to Use

Do not use this skill when:

- The user is asking to execute an existing Skill
- The user only wants API code or function implementations
- The task does not involve defining a reusable, callable capability

---

## Instructions

When using this skill, follow these rules strictly:

1. Generate **one and only one** `SKILL.md`
2. The generated Skill **must include**:
   - YAML frontmatter with `name` and `description`
   - Clear sections for purpose, usage guidance, and examples
3. The Skill name must:
   - Be lowercase
   - Use kebab-case
   - Represent a single, atomic capability
4. The content must be written **for an AI model**, not for an end user
5. Do not invent implementation details that were not provided
6. Prefer clarity and constraints over verbosity

---

## Expected Input

The user will usually provide:

- A description of the capability
- Optional parameters or inputs
- Constraints or safety considerations
- The intended role of the Skill in a larger system

---

## Output Format

The output must be a **valid `SKILL.md` file**, written in Markdown, and ready to be saved to disk.

Do not include explanations outside the generated Skill document.

---

## Example

### Input (from user)

> “I want a Skill that validates JSON files and reports schema errors.”

### Output (generated by this skill)

A complete `SKILL.md` defining a `json-validator` Skill, including metadata, usage guidance, and examples.

---

## Model Guidance

- Treat Skill generation as a **specification task**, not creative writing
- Optimize for reusability, auditability, and correctness
- Assume the generated Skill will be used by other agents