agentleFS
I-OnlabsI-Onlabs/claude-code-skills/skill-creation/SKILL.md

skill-creation

Skill4 starsChanged 26 days agoMITView on GitHub

Use when creating new skills, agents, or improving existing ones. Provides the 10-step systematic process for high-quality skill development.

14 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Skill Creation Methodology

When This Skill Activates

  • Creating a new Claude Code skill or agent
  • "How do I make a skill for X?"
  • "Create a skill that does Y"
  • Systematizing a repetitive workflow
  • Improving existing skills

The 10-Step Process

StepNameOutput
1UNDERSTANDProblem statement (1-2 paragraphs)
2EXPLORE3-5 failure modes without guidance
3RESEARCHNotes with sources
4SYNTHESIZE5-10 core principles
5DRAFTFirst SKILL.md
6SELF-CRITIQUEGap list
7ITERATEImproved draft
8TESTBefore/after comparison
9FINALIZEProduction-ready file
10DOCUMENTEcosystem integration

Quick Start Template

---
name: <skill-name>
description: <When to activate - be precise!>
---

# <Skill Name>

## When This Skill Activates
- <Trigger condition 1>
- <Trigger condition 2>

## Methodology

### Step 1: <First Step>
<What to do>

### Step 2: <Second Step>
<What to do>

## Examples

<Concrete examples>

## Quality Checklist
- [ ] Criteria 1
- [ ] Criteria 2

Quality Checklist

Every skill MUST:

  • Address documented failure modes
  • Be actionable, not theoretical
  • Have concrete examples
  • Have precise trigger conditions
  • Integrate with relevant agents

Skill vs Agent Decision

NeedSolution
Methodology guidanceSkill
Autonomous executionAgent
BothSkill + Agent pair

File Locations

~/.claude/skills/<name>/SKILL.md    # Skill file
~/.claude/agents/<name>.md          # Agent file

Integration

Agent: skill-architect (for autonomous execution of this methodology)

Delegate to skill-architect when:

  • Complex skill requiring deep research
  • Full 10-step process needed
  • User wants hands-off skill creation

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.