Create New Skills

Standardize Skill creation with YAML frontmatter and progressive disclosure.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/declanshanaghy/fenrir-ledger --skill create-new-skills-declanshanaghy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Create New Skills
Source: https://github.com/declanshanaghy/fenrir-ledger/tree/main/.claude/skills/meta-skill
Command: npx skills add https://github.com/declanshanaghy/fenrir-ledger --skill create-new-skills-declanshanaghy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ambiguity in Skill creation leads to inconsistent results and longer onboarding. This guide provides a repeatable process to design, document, and package Claude Skills.

Core Features & Use Cases

  • Standardized workflow for defining Skill scope, directory structure, and YAML frontmatter requirements.
  • Progressive disclosure patterns that load metadata, instructions, and resources only as needed.
  • Templates and examples to accelerate authoring, testing, and sharing of Skills.

Quick Start

Create a new Skill directory under .claude/skills/, draft SKILL.md with YAML frontmatter, and begin the Instructions section.

Frequently Asked Questions about Create New Skills

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

FAQPage Schema
How do I create new agent skills with a standardized workflow?

Eliminate ambiguity in skill creation by following a step-by-step workflow from conception to validation, specifying required frontmatter and progressive-disclosure rules for safe packaging.

What is progressive disclosure in claude-code agent skills?

Progressive disclosure in agent skills loads metadata, instructions, and resources only as needed, optimizing performance and ensuring predictable behavior during workflow execution.

How do I structure a SKILL.md file with YAML frontmatter?

Structure SKILL.md by creating a directory under .claude/skills/, drafting required YAML frontmatter, and beginning the Instructions section to define the skill's processing logic.

Does this skill creation process work for documentation-heavy workflows?

Yes, this skill creation process applies across projects and documentation-heavy workflows, supporting step-by-step workflows from conception to validation for safe, predictable packaging.

Why does my claude-code skill produce inconsistent results?

Inconsistent results often stem from ambiguity in skill creation; applying a standardized workflow with required frontmatter and progressive-disclosure rules ensures safe, predictable packaging.