create-m4-skill

Document the process of creating new M4 skills with structured frontmatter and provenance.

40|17|Updated Oct 27, 2025
One-click install
npx skills add https://github.com/hannesill/m4 --skill create-m4-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-m4-skill
Source: https://github.com/hannesill/m4/tree/main/src/m4/skills/system/create-m4-skill
Command: npx skills add https://github.com/hannesill/m4 --skill create-m4-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a guided blueprint to create new M4 skills with standardized structure, provenance tracking, and tier categorization.

Core Features & Use Cases

  • Establishes a repeatable workflow to define a skill's frontmatter (name, description, tier, category), body, and provenance.
  • Enforces validation checks and provenance documentation to ensure future maintainability and auditability.
  • Useful for contributors who want to document a clinical concept or framework feature as a new M4 skill.

Quick Start

Outline the concept and start generating the required frontmatter and provenance for the new skill.

Frequently Asked Questions about create-m4-skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
What is an M4 skill and when do I need to create one?

To create an M4 skill, you generate the required frontmatter (name, description, tier, category), write a structured SKILL.md body, and add a PROVENANCE.yaml file to track origins and ensure auditability.

How do I structure frontmatter and provenance for a new skill?

Skill creation frontmatter requires name, description, tier, and category fields. Provenance tracking requires a PROVENANCE.yaml file to document the skill's origins, ensuring future maintainability and auditability.

Can I include optional resource directories when building an M4 skill?

Yes, M4 skill creation supports optional resources directories for scripts, references, and assets. These directories are integrated into the skill structure when present to provide additional supporting documentation.

What's the best way to validate a system skill before finalizing it?

The best way to validate a system skill is to apply the enforced validation checks during the creation workflow. This ensures the frontmatter, body, and provenance meet the standardized structure required for maintainability.

Does M4 skill creation work for dataset-agnostic system skills?

Yes, M4 skill creation applies to system-skill creation using a dataset-agnostic frontmatter structure. This allows contributors to document concepts independently of specific datasets while maintaining standardized provenance.