メニュー
Create, validate, and convert skills for the agent ecosystem. Enforces standardized structure for consistency. Enables self-evolution by creating new skills on demand, converting MCP servers and codebases to skills.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/oimiragieo/agent-studio --skill skill-creatorコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Agent frontmatter enhancements — disallowedTools, mcpServers scoping, fork_eligible field
Context management — pre-compact persistence, threshold alignment, microcompact/circuit breaker detection
Hook enhancements — updatedInput for bash safety prefixes, suppressOutput for verbose blocks, denial-based routing feedback
SKILL.md を表示中
SOC 職業分類に基づく
| name | skill-creator |
| description | Create, validate, and convert skills for the agent ecosystem. Enforces standardized structure for consistency. Enables self-evolution by creating new skills on demand, converting MCP servers and codebases to skills. |
| version | 2.3.0 |
| model | sonnet |
| invoked_by | both |
| user_invocable | true |
| tools | ["Read","Write","Edit","Bash","Glob","Grep","WebSearch","WebFetch"] |
| args | <action> [options] |
| best_practices | ["Always use standardized structure","Include Memory Protocol section","Create scripts/main.cjs for executable logic","Validate after creation"] |
| error_handling | graceful |
| streaming | supported |
| verified | true |
| lastVerifiedAt | 2026-04-20 |
| dependencies | ["research-synthesis"] |
| source | builtin |
| trust_score | 100 |
| provenance_sha | eb1c31c856e8d240 |
Mode: Script-First - Use scripts/create.cjs as the canonical path for new skill creation, then use this guide for research, review, and integration follow-through.
Create, validate, install, and convert skills for the multi-agent ecosystem without skipping routing, catalog, and registry integration.
Use this skill to:
| Action | Use when | Primary command |
|---|---|---|
create | Creating a brand-new skill | node .claude/skills/skill-creator/scripts/create.cjs --name <skill-name> --description "<summary>" |
convert | Converting an MCP server into a skill | node .claude/skills/skill-creator/scripts/convert.cjs --source <package-or-url> |
validate | Checking an existing skill definition | node .claude/tools/cli/validate-integration.cjs .claude/skills/<skill-name>/SKILL.md |
install | Importing a skill from GitHub | Follow the preserved install flow in enterprise bundle details |
convert-codebase | Turning an external tool or codebase into a skill | Follow the preserved conversion notes in |
consolidate | Folding many narrow skills into domain experts | Use the preserved consolidation reference in enterprise bundle details |
convert-rules | Migrating legacy rules into skills | Use the preserved rule-conversion reference in enterprise bundle details |
assign | Assigning a skill to one or more agents | Use the checklist below and the detailed matrix in integration reference |
register-hooks / register-schemas | Wiring existing assets into a skill | Use the preserved action notes in enterprise bundle details |
show-structure | Reviewing the expected folder layout | Use the template and directory guidance below |
Before creating any skill file, check whether the skill already exists.
test -f .claude/skills/<skill-name>/SKILL.md && echo "EXISTS" || echo "NEW"
artifact-updater.unified-creator-guard.cjs is expected to block unsafe creation paths.Run the duplicate detector before proceeding:
const { checkDuplicate } = require('.claude/lib/creation/duplicate-detector.cjs');
const result = checkDuplicate({
artifactType: 'skill',
name: proposedName,
description: proposedDescription,
keywords: proposedKeywords || [],
});
Handle results exactly as before:
EXACT_MATCH -> stop and route to skill-updaterREGISTRY_MATCH -> investigate registry/file drift before creatingSIMILAR_FOUND -> review candidates and decide whether to create or updateNO_MATCH -> continue to Step 0.5Run the companion check before creation:
.claude/lib/creators/companion-check.cjs.checkCompanions("skill", "{skill-name}").If the skill provides behavioral guidance that should persist outside explicit invocation, confirm whether it also needs .claude/rules/<skill-name>.md.
Pick the narrowest action that matches the request:
create for a brand-new skillconvert for MCP server conversioninstall for importing a GitHub skillconvert-codebase for lifting an external tool or codebase into a skillvalidate when the artifact already exists and only needs verificationComplete the preserved research workflow in research gate details before you finalize the skill body. That reference keeps the original VoltAgent, Exa, arXiv, typed-artifact-search, evidence-quality, and external-content safety material.
For a brand-new skill, start with the managed scaffold:
node .claude/skills/skill-creator/scripts/create.cjs --name <skill-name> --description "<summary>"
Use the template below as the contract that the generated SKILL.md must satisfy. The detailed format notes, action examples, and layout guidance remain preserved in enterprise bundle details.
The enterprise bundle is now opt-in. Start from the minimal scaffold by default, and only add enterprise files when the request explicitly asks for --enterprise or the capability truly needs them. At minimum, decide whether the skill needs:
scripts/ for executable helpershooks/ for pre/post execution enforcementschemas/ for typed interfacestemplates/ or references/ for reusable authoring material.claude/tools/<skill-name>/.claude/workflows/The original bundle breakdown and acceptance checklist are preserved in enterprise bundle details.
Before completion:
node .claude/tools/cli/validate-integration.cjs .claude/skills/<skill-name>/SKILL.md.After scaffolding or updating a skill, verify the routing layer can still discover it. Treat any no matching agent/skill result as a routing gap that must be resolved before handoff.
Use this baseline structure in SKILL.md:
---
name: skill-name
description: What the skill does
version: 1.0.0
model: sonnet
invoked_by: user | agent | both
user_invocable: true | false
tools: [Read, Write, Bash]
args: "<required> [optional]"
agents: [developer, qa]
category: "Validation & Quality"
tags: [testing, validation]
frontmatter:
triggers:
- <pattern or trigger phrase that should invoke this skill>
token_budget: 10000 # estimated token cost for invoking this skill
# output_schema_ref: (set if a skill-*-output.schema.json exists for this skill)
# requires_skills: [] # list dependent skill names here
---
# Skill Name
## Purpose
What this skill accomplishes.
## Usage
How the skill is invoked and applied.
## Examples
Concrete invocation or workflow examples.
Required frontmatter fields that must stay explicit: name, description, version, agents, category, tags, tools, invoked_by, and user_invocable.
frontmatter blockThe frontmatter object is an optional, machine-parseable metadata block introduced in v3.1.0. Agents can inspect it without parsing full prose. The create action emits it by default in all new SKILL.md files.
| Field | Type | Required | Purpose |
|---|---|---|---|
triggers | string[] | No | Patterns that should cause the skill to be invoked |
token_budget | integer | No | Estimated token cost (min 1000); used by planner for budgeting |
output_schema_ref | string | No | Path to the skill's output schema (e.g. skill-foo-output.schema.json) |
requires_skills | string[] | No | Names of skills this skill depends on (resolved by skill index) |
All frontmatter fields are optional. Existing SKILL.md files without the frontmatter block remain valid — the schema uses additionalProperties: true at the root level, and the frontmatter property is not in required. Do NOT add frontmatter to existing skills during unrelated updates; add it only when the skill is being intentionally refreshed or verified.
create, convert, install, validate, or another supported action)..claude/skills/<skill-name>/SKILL.md.agents, category, and tags.CLAUDE.md, routing notes, and the skill catalog were updated if required.node .claude/tools/cli/validate-integration.cjs .claude/skills/<skill-name>/SKILL.md passes.node .claude/tools/cli/generate-skill-index.cjs was run when discoverability metadata changed.npm run gen:all-registries was run for new or re-registered skills.This creator skill must keep every new or updated skill aligned with the broader creator ecosystem:
agent-creator for ownership and execution pathstool-creator for executable helpershook-creator for enforcement and guardrailsrule-creator and semgrep-rule-creator for policy coveragetemplate-creator for reusable scaffoldsworkflow-creator for orchestrationcommand-creator for user-facing shortcutsBefore handoff, verify the related ecosystem updates:
validate-integration.cjs passes for the skill artifact.For new skill patterns, packaging approaches, or AI-adjacent methodologies: