메뉴
Guide for creating effective skills. Use when creating a new skill, updating an existing skill, or verifying skills work before deployment. Covers skill structure, creation process, testing methodology, and packaging.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | skill-creator |
| description | Guide for creating effective skills. Use when creating a new skill, updating an existing skill, or verifying skills work before deployment. Covers skill structure, creation process, testing methodology, and packaging. |
Create effective, tested skills that extend Claude's capabilities with specialized knowledge, workflows, and tools.
Skills are modular packages that provide:
The context window is a public good. Only add context Claude doesn't already have. Challenge each piece: "Does Claude really need this?" Prefer concise examples over verbose explanations.
Three-level loading to manage context efficiently:
skill-name/
├── SKILL.md # Required: frontmatter + instructions
├── scripts/ # Optional: executable code (Python/Bash)
├── references/ # Optional: docs loaded into context as needed
└── assets/ # Optional: files used in output (templates, icons)
Only name and description are required. The description is the primary trigger — it determines when the skill loads.
---
name: my-skill
description: Use when [specific triggering conditions and symptoms]
---
Description rules:
Do NOT include: README.md, CHANGELOG.md, INSTALLATION_GUIDE.md, or auxiliary docs about the creation process.
Ask the user:
For each example, identify what scripts, references, and assets would help when executing these workflows repeatedly.
Create the skill directory with a SKILL.md containing frontmatter (name + description) and the appropriate resource subdirectories (scripts/, references/, assets/) as needed.
Consult these guides based on your needs:
Start with reusable resources, then write SKILL.md. Use imperative/infinitive form.
Verify the skill has valid frontmatter (name + description, kebab-case name, under 1024 chars), then package with zip -r skill-name.skill skill-name/ if distributing.
Skill creation IS Test-Driven Development applied to process documentation.
Follow RED-GREEN-REFACTOR:
RED — Baseline test (before writing/editing the skill):
GREEN — Write minimal skill:
REFACTOR — Close loopholes:
The Iron Law: No skill without a failing test first. This applies to new skills AND edits.
For the complete testing methodology (pressure scenarios, pressure types, plugging holes, meta-testing), see references/testing-skills-with-subagents.md.
For psychology of persuasion techniques in discipline-enforcing skills, see references/persuasion-principles.md.
Future Claude needs to FIND your skill:
condition-based-waiting not async-test-helpers)| Type | Examples | Test With |
|---|---|---|
| Discipline (rules) | TDD, verification | Pressure scenarios, rationalization resistance |
| Technique (how-to) | condition-based-waiting | Application + edge cases |
| Pattern (mental model) | flatten-with-flags | Recognition + counter-examples |
| Reference (docs/APIs) | API guides | Retrieval + application accuracy |
Pattern 1: High-level guide linking to references for advanced features Pattern 2: Domain-specific organization (only load relevant domain file) Pattern 3: Conditional details (basic inline, advanced linked)
Keep references one level deep from SKILL.md. For files >100 lines, include a table of contents.
Use ONLY for non-obvious decisions, process loops, or A-vs-B choices. Never for reference material, code examples, or linear instructions.
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes - four-phase framework with built-in backward tracing for deep-stack failures, ensuring root-cause understanding before implementation
Turn the current session into a coordination thread that routes per-branch implementation work to durable, reusable child agents. Use when the user says 'orchestrator on', wants this session to act as chief-of-staff across branches, or asks to route work without implementing locally.
Test-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions "red-green-refactor", wants integration tests, or asks for test-first development.
Adversarial questioning and collaborative shaping in one skill. Two modes: stress-test (challenge a plan/design until gaps surface) and shape (iterate on requirements and solution options until alignment). Use when user says 'grill me', 'stress-test this', 'challenge this plan', 'shape this', 'let's explore options', or wants to pressure-test any design before committing.
Compact the current session into a handoff document for another agent or future session to pick up. Captures mission, repo state, role, scope, locked decisions, dispatch sequence, and a pickup checklist. Studio-aware. Use when pausing work, ending a session, switching contexts, or preparing for a fresh agent to continue.
Initialize a new project with PROJECT.md + ROADMAP.md in .project/, set up jira and studio, and route to the first sprint. Use when starting a new project, onboarding a new codebase, or setting up workflow infrastructure for the first time.