Token导航 LogoToken导航TokenDH.com
前端设计敏感数据github未标认证来源可访问clear审计异常

moai-library-toon摩艾图书馆卡通

Agent Skill

moai-library-toon 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

188

周安装

8

GitHub Stars

公开资料未说明

下载量

66
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:moai-library-toon(摩艾图书馆卡通)
来源仓库:https://github.com/rdmptv/adbautoplayer
仓库路径:skills/moai-library-toon
安装命令:
npx skills add https://github.com/rdmptv/adbautoplayer --skill moai-library-toon
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/rdmptv/adbautoplayer --skill moai-library-toon

简介

moai-library-toon 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态、代码变更或协作事项进行整理。
  • 通过 npx skills add 命令从 adbautoplayer 仓库安装并使用。
  • 安装前建议确认权限范围和维护状态,注意是否会触发联网或文件读写操作。
  • 具体用法可结合来源仓库 README 进一步核验实际功能边界。

SKILL.md

Quick Reference (30 seconds)

TOON v4.0 = YAML-based hierarchical format inspired by BMAD Method

TOON (Token-Optimized Object Notation) adopts BMAD's clean YAML patterns for agent and workflow definitions. This provides 40-60% token reduction while maintaining full expressiveness.

Key Insight: BMAD's *.agent.yaml and workflow.yaml ARE the optimal token-efficient format.

Core Benefits:

  • 40-60% token reduction vs JSON
  • Clean hierarchical YAML structure
  • Runtime variable substitution ({project-root}, {config_source}:key)
  • Human-readable and LLM-parseable
  • Proven in BMAD Method v6 (650+ files, 25+ agents)

TOON Agent Definition Pattern (BMAD-Style)

Standard Agent YAML Structure

# {agent-name}.agent.yaml
agent:
  metadata:
    id: "{bmad_folder}/module/agents/{name}.md"
    name: "Display Name"
    title: "Agent Title"
    icon: "emoji"
    module: "module-code"  # Optional: for module-scoped agents

  persona:
    role: "Expert Title"
    identity: "Background, expertise, and approach description"
    communication_style: "How the agent communicates with users"
    principles: |
      - Core principle 1
      - Core principle 2
      - Core principle 3

  critical_actions:  # MANDATORY steps on activation
    - "FIRST: Read all context before any action"
    - "SECOND: Validate inputs against requirements"
    - "THIRD: Execute with full audit trail"

  menu:
    - trigger: command-name
      workflow: "{project-root}/{bmad_folder}/module/workflows/workflow-name/workflow.yaml"
      description: "What this command does"

    - trigger: another-command
      exec: "shell command to execute"
      description: "Direct shell execution"

    - trigger: load-template
      tmpl: "{project-root}/{bmad_folder}/module/templates/template.md"
      description: "Display template file"

Menu Command Types (6 Types)

TypePurposeExample
workflowExecute multi-step workflowworkflow: "{path}/workflow.yaml"
execRun shell commandexec: "npm test"
actionPerform named actionaction: "validate_schema"
tmplDisplay template filetmpl: "{path}/template.md"
dataLoad data filedata: "{path}/data.json"
validate-workflowValidate workflow structurevalidate-workflow: "{path}/"

Example: MoAI Developer Agent

# expert-backend.agent.yaml
agent:
  metadata:
    id: ".claude/agents/moai/expert-backend.md"
    name: expert-backend
    title: Backend Expert Agent
    icon: "🔧"

  persona:
    role: Senior Backend Engineer
    identity: Expert in API design, database integration, and server architecture with deep knowledge of Python, TypeScript, and cloud platforms.
    communication_style: "Technical and precise. References file paths and line numbers. Provides code examples."
    principles: |
      - Follow TRUST 5 quality standards
      - Red-green-refactor TDD cycle
      - API-first design approach
      - Comprehensive error handling
      - Performance-conscious implementation

  critical_actions:
    - "READ existing codebase structure before implementation"
    - "VERIFY API contracts match SPEC requirements"
    - "WRITE tests before implementation code"
    - "VALIDATE all endpoints with integration tests"

  menu:
    - trigger: implement-api
      workflow: "{project-root}/.claude/workflows/api-implementation/workflow.yaml"
      description: "Implement API endpoints from SPEC"

    - trigger: design-schema
      workflow: "{project-root}/.claude/workflows/schema-design/workflow.yaml"
      description: "Design database schema"

    - trigger: run-tests
      exec: "pytest tests/ -v --cov"
      description: "Run backend test suite"

TOON Workflow Definition Pattern (BMAD-Style)

Standard Workflow YAML Structure

# workflow.yaml
name: workflow-name
description: "What this workflow accomplishes"
author: "Author Name"

# Configuration references (runtime variable resolution)
config_source: "{project-root}/{bmad_folder}/module/config.yaml"
output_folder: "{config_source}:output_folder"
user_name: "{config_source}:user_name"
communication_language: "{config_source}:communication_language"

# Project context (glob pattern)
project_context: "**/project-context.md"

# Workflow components
installed_path: "{project-root}/{bmad_folder}/module/workflows/workflow-name"
instructions: "{installed_path}/instructions.md"
checklist: "{installed_path}/checklist.md"

# Workflow settings
standalone: true
web_bundle: false

Runtime Variable Substitution

VariableSourceDescription
{project-root}Auto-detectProject root directory
{bmad_folder}ConfigUsually .claude or .bmad
{config_source}:keyYAML lookupRead value from config file
{installed_path}ComputedCurrent workflow directory
{date}SystemCurrent date (system-generated)

Example: MoAI TDD Workflow

# .claude/workflows/tdd-implementation/workflow.yaml
name: tdd-implementation
description: "Execute RED-GREEN-REFACTOR cycle for SPEC implementation"
author: "MoAI-ADK"

# Configuration
config_source: "{project-root}/.moai/config/config.json"
test_coverage_target: "{config_source}:constitution.test_coverage_target"
enforce_tdd: "{config_source}:constitution.enforce_tdd"

# Project context
project_context: "**/CLAUDE.md"
spec_file: "{project-root}/.moai/specs/SPEC-{spec_id}/SPEC.md"

# Workflow components
installed_path: "{project-root}/.claude/workflows/tdd-implementation"
instructions: "{installed_path}/instructions.md"
checklist: "{installed_path}/checklist.md"
validation: "{installed_path}/validation-rules.md"

# Micro-file steps
steps:
  - name: "Phase 1 - RED"
    file: "{installed_path}/step-01-red.md"
    checkpoint: true
  - name: "Phase 2 - GREEN"
    file: "{installed_path}/step-02-green.md"
    checkpoint: true
  - name: "Phase 3 - REFACTOR"
    file: "{installed_path}/step-03-refactor.md"
    checkpoint: false

standalone: true
web_bundle: false

TOON Configuration Pattern

Config YAML Structure

# config.yaml
project:
  name: "Project Name"
  version: "1.0.0"
  language: "python"

settings:
  output_folder: ".moai/output"
  user_name: "Developer"
  communication_language: "ko"
  document_output_language: "en"

constitution:
  enforce_tdd: true
  test_coverage_target: 90

agents:
  enabled:
    - expert-backend
    - expert-frontend
    - manager-tdd
  default_model: "sonnet"

Variable Reference Syntax

# In workflow.yaml, reference config values:
output_folder: "{config_source}:settings.output_folder"
test_target: "{config_source}:constitution.test_coverage_target"

# Nested key access with dot notation
user_lang: "{config_source}:settings.communication_language"

Token Efficiency Analysis

TOON YAML vs JSON Comparison

# TOON YAML (45 tokens)
agent:
  metadata:
    name: expert-backend
    title: Backend Expert
  persona:
    role: Senior Engineer
    principles: |
      - TDD first
      - API-first design
// JSON equivalent (78 tokens)
{
  "agent": {
    "metadata": {
      "name": "expert-backend",
      "title": "Backend Expert"
    },
    "persona": {
      "role": "Senior Engineer",
      "principles": ["TDD first", "API-first design"]
    }
  }
}

Token Savings: 42% reduction

Cumulative Savings per Agent

ComponentJSON TokensTOON YAMLReduction
Metadata803556%
Persona1507053%
Critical Actions1005050%
Menu (5 items)2009055%
Total53024554%

Migration from Legacy TOON

Legacy TOON v2 (CSV-like)

# OLD FORMAT - DEPRECATED
scripts[N]{name,skill,lines,deps}:
script1.py,skill-a,200,2
script2.py,skill-b,300,3

New TOON v4 (YAML-based)

# NEW FORMAT - RECOMMENDED
scripts:
  - name: script1.py
    skill: skill-a
    lines: 200
    deps: 2
  - name: script2.py
    skill: skill-b
    lines: 300
    deps: 3

Migration Note: CSV-like format is still valid for tabular data, but YAML is preferred for agent/workflow definitions.


Implementation Modules

For detailed patterns:

  • Core Implementation: modules/core.md
  • Advanced Patterns: modules/advanced-patterns.md
  • BMAD Integration: modules/bmad-integration.md (NEW)

Works Well With

Agents:

  • builder-workflow - UV script generation with TOON patterns
  • builder-agent - Agent YAML creation
  • builder-skill - Skill definition with TOON metadata

Skills:

  • moai-foundation-core - Execution rules and delegation
  • moai-foundation-claude - Claude Code configuration
  • moai-lang-unified - Multi-language patterns

Commands:

  • /moai:1-plan - SPEC generation with TOON metadata
  • /moai:2-run - TDD execution with TOON workflows
  • /moai:3-sync - Documentation with TOON configs

BMAD Method Attribution

TOON v4.0 patterns are inspired by BMAD Method v6.0.0-alpha.12:

  • Agent YAML structure (*.agent.yaml)
  • Workflow YAML structure (workflow.yaml)
  • Runtime variable substitution patterns
  • Menu command system (6 types)
  • Configuration cascade pattern

Key Difference: MoAI uses Markdown directly (no XML compilation) while maintaining BMAD's clean YAML patterns.


Version: 4.0.0 Status: Active (BMAD YAML Patterns) Last Updated: 2025-12-01

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

28.59%
按下载量换算19

windsurf

21.47%
按下载量换算14

OpenCode

19.36%
按下载量换算13

Codex

13.51%
按下载量换算9

Antigravity

8.17%
按下载量换算5

Gemini CLI

3.3%
按下载量换算2

安全审计

Gen Agent Trust Hub

未通过

Socket

通过

Snyk

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。

来源信息

继续浏览同类 Skills