Token导航 LogoToken导航

Agent Skill · 前端设计

master-architect(建筑师大师)

master-architect 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。

master-architect 是前端设计类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描已通过,按安装量在前端设计类的 10,216 个 Skill 里排第 9,098。

总安装 238周安装 10GitHub Stars 2下载 83核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

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:master-architect(建筑师大师)
来源仓库:https://github.com/cycleuser/skills
仓库路径:skills/master-architect
安装命令:
npx skills add https://github.com/cycleuser/skills --skill master-architect
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。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/cycleuser/skills --skill master-architect

这个 Skill 是做什么的

前端设计类 Skill 关注页面结构、视觉规范和组件写法,输出通常是可运行的界面代码。

master-architect 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时使用。

通过 npx skills add 命令从指定 GitHub 仓库安装,需确认权限和维护状态。使用前建议核实是否会触发联网、命令执行或文件读写操作。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/master-architect

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/cycleuser/skills --skill master-architect;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

master-architect 怎么安装?

在终端执行 npx skills add https://github.com/cycleuser/skills --skill master-architect。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

master-architect 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

master-architect 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

master-architect 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

master-architect 用的人多吗?

公开数据是累计安装 238 次,近一周安装 10 次,下载 83 次,GitHub 2 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

master-architect 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理前端设计类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

前端设计类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
frontend-design

前端设计 · Codex、Claude

8,823,360126,072只读MIT通过
vercel-composition-patterns

Vercel 组合模式 · Codex、Claude

3,858,81625,893需要联网MIT通过
vercel-react-native-skills

Vercel React native skills 前端 · Codex、Claude

2,468,66425,878只读MIT通过
adapt

adapt 搜索 · Codex、Claude

2,027,04023,125操作浏览器未明确通过
bolder

bolder 分析 · Codex、Claude

2,004,28823,130只读未明确通过
master-architect

建筑师大师 · 本页

2382需要联网未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Master Architect

The supreme architect agent for complex software development projects.

Philosophy

┌─────────────────────────────────────────────────────────────────────────┐
│                        MASTER ARCHITECT                                  │
│                    "Excellence Through Iteration"                        │
├─────────────────────────────────────────────────────────────────────────┤
│                                                                          │
│    ┌──────────┐    ┌──────────┐    ┌──────────┐    ┌──────────┐        │
│    │ ANALYZE  │───▶│  DESIGN  │───▶│ DECOMPOSE │───▶| ITERATE  │        │
│    │          │    │          │    │           │    │          │        │
│    └────┬─────┘    └────┬─────┘    └─────┬─────┘    └────┬─────┘        │
│         │               │                │               │              │
│         │               │                │               │              │
│    ┌────▼─────┐    ┌────▼─────┐    ┌─────▼─────┐    ┌────▼─────┐        │
│    │ QUALITY  │    │ QUALITY  │    │  QUALITY  │    │ QUALITY  │        │
│    │  GATE 1  │    │  GATE 2  │    │   GATE 3  │    │  GATE 4  │        │
│    └────┬─────┘    └────┬─────┘    └─────┬─────┘    └────┬─────┘        │
│         │               │                │               │              │
│         └───────────────┴────────────────┴───────────────┘              │
│                                   │                                      │
│                            ┌──────▼──────┐                               │
│                            │  INTEGRATE  │                               │
│                            │   & DELIVER │                               │
│                            └─────────────┘                               │
│                                                                          │
└─────────────────────────────────────────────────────────────────────────┘

Phase 1: Requirement Analysis

Objective: Achieve deep understanding of the task, its context, constraints, and success criteria.

Process: The requirement analysis phase proceeds through four key activities. First, stakeholder analysis identifies who the users are, what pain points they experience, and what outcomes they expect. Second, constraint mapping documents technical constraints around languages, frameworks, and platforms, along with resource constraints for time, budget, and personnel, plus quality constraints for performance, security, and usability. Third, success criteria definition establishes measurable outcomes, acceptance thresholds, and validation methods. Fourth, risk assessment identifies technical risks, integration risks, and timeline risks.

Quality Gate 1 checks that all user personas are documented, no constraints remain unresolved, all metrics are quantified, and all high risks have mitigation plans. Exit condition: All gate criteria passed. Architecture review approved.

Phase 2: Architecture Design

Objective: Create the system blueprint with clear module boundaries and interfaces.

Process: The architecture design phase creates a comprehensive system blueprint. The system blueprint includes a high-level architecture diagram, component responsibilities, and data flow patterns. Module boundaries are defined with single responsibility per module, clear interface contracts, and a dependency graph. Technology decisions cover language and framework selection with rationale, library choices with alternatives considered, and infrastructure requirements. Non-functional design addresses performance targets, scalability approach, and security model.

Architecture Documentation Template

The architecture document should follow a standard structure. Section 1 provides an overview with one paragraph describing the system's purpose and approach. Section 2 presents the high-level design with an architecture diagram. Section 3 details module decomposition, defining for each module its responsibility and dependencies. For example: Core handles business logic with no dependencies, API provides external interface depending on Core, CLI offers command interface depending on API, and GUI provides graphical interface depending on API. Section 4 documents interface contracts including API interface specifications and data model definitions. Section 5 describes the technology stack with rationale for each layer: Python 3.10+ for backend due to rich ecosystem and rapid development, Flask for web due to being lightweight and flexible, PySide6 for GUI due to cross-platform native feel. Section 6 specifies quality attributes including performance targets (response time under 100ms), scalability targets (handle 10,000 concurrent users), and security requirements (input validation, output sanitization).

Quality Gate 2 verifies that each module has single responsibility (module cohesion), no circular dependencies exist (coupling analysis), all interfaces are documented (interface clarity), and all technology choices are justified (technology rationale). Exit condition: Architecture review approved by stakeholder simulation.

Phase 3: Task Decomposition

Objective: Break architecture into granular, actionable sub-tasks with clear dependencies.

Process: The task decomposition phase transforms the architecture into actionable work items. Feature extraction lists all features from requirements, prioritizes them by value and risk, and maps features to modules. Task breakdown converts each feature into multiple tasks, with each task targeting under 4 hours of work and clear completion criteria. Dependency graph builds a task dependency tree, identifies the critical path, and plans parallel execution opportunities. Iteration planning assigns iteration targets, defines iteration scope, and sets quality targets per iteration.

Task Decomposition Template: For each module and feature, define the priority (High/Medium/Low), dependencies (list of prerequisite features), and estimated complexity (Simple/Medium/Complex). Then break into sub-tasks with ID, description, time estimate, and completion criteria. Finally, create an iteration plan grouping tasks into logical iterations.

Quality Gate 3 verifies all tasks are under 4 hours (task granularity), no ambiguous dependencies exist (dependency clarity), all features are mapped to tasks (coverage), and the critical path is identified and documented. Exit condition: Task breakdown reviewed and approved.

Phase 4: Iterative Development

Objective: Develop each module through rigorous iteration until excellence.

Iteration Cycle:

┌─────────────────────────────────────────────────────────────────────┐
│                    SINGLE MODULE ITERATION                           │
├─────────────────────────────────────────────────────────────────────┤
│                                                                      │
│    ┌──────────┐    ┌──────────┐    ┌──────────┐    ┌──────────┐   │
│    │  BUILD   │───▶│  TEST    │───▶│ EVALUATE │───▶│ REFINE   │   │
│    │          │    │          │    │          │    │          │   │
│    └──────────┘    └──────────┘    └──────────┘    └──────────┘   │
│         │               │               │               │          │
│         │               │               │               │          │
│         │               │               ┌───────┐       │          │
│         │               │               │SCORE  │       │          │
│         │               │               │≥ 90?  │       │          │
│         │               │               └───┬───┘       │          │
│         │               │              YES  │  NO        │          │
│         │               │               ┌───┴───┐       │          │
│         │               │               ▼       ▼       │          │
│         │               │          [DONE]  [NEXT       │          │
│         │               │                   ITERATION] │          │
│                                                                      │
└─────────────────────────────────────────────────────────────────────┘

Iteration Scoring: The iteration uses a weighted scoring system across four metrics. Test coverage carries 30% weight with a target of 80% or higher. Code quality carries 25% weight targeting a lint score of zero. Functionality carries 25% weight requiring all tests to pass. Documentation carries 20% weight requiring completeness. Minimum score to proceed: 90/100.

Iteration Report Template: Each iteration report documents build summary (files created, modified, lines added), test results (tests run, passed, coverage percentage), quality metrics (lint errors, type errors, complexity score), the overall score, improvements applied, and next iteration focus areas.

Quality Gate 4 requires module score of 90 or higher, test coverage of 80% or higher, 100% tests passing, and complete documentation. Exit condition: Module achieves excellence score. Proceed to next module.

Phase 5: Integration & Validation

Objective: Ensure all modules work together coherently.

Process: The integration and validation phase verifies cross-module coherence. Interface validation verifies all contracts are honored, tests inter-module communication, and validates data flow. Integration testing covers end-to-end test scenarios, edge case coverage, and error handling verification. Performance validation includes load testing, stress testing, and resource usage profiling. User acceptance involves scenario walkthroughs, usability assessment, and documentation review.

Quality Gate 5 requires all integration tests to pass, performance targets to be met, user scenarios to be validated, and error handling to be complete.

Phase 6: Documentation Generation

Objective: Produce documentation compliant with all skill standards.

Documentation Checklist: Code-level documentation includes README.md in English, README_CN.md in Chinese, API documentation, code comments, and type hints. Architecture-level documentation includes system architecture document, module dependency diagram, interface specifications, and data model definitions. User-level documentation includes installation guide, user manual, troubleshooting guide, and FAQ. Development-level documentation includes CONTRIBUTING.md, development setup guide, testing guide, and release notes template.

Skill Compliance Matrix: Different skills have different documentation requirements. The python-project-developer skill requires ToolResult pattern documentation, CLI flags, and tests. The software-planner skill requires README sections and sample data. The academic-writer skill is needed only if paper documentation is required. |

Commands Reference

/architect design

Execute complete architecture design workflow.

/architect design "Build a text classification system with CLI and GUI"

→ Phase 1: Requirement Analysis
  - Analyzing stakeholders...
  - Mapping constraints...
  - Defining success criteria...
  ✓ Gate 1 passed

→ Phase 2: Architecture Design
  - Creating system blueprint...
  - Defining module boundaries...
  ✓ Gate 2 passed

→ Phase 3: Task Decomposition
  - Breaking into 15 tasks...
  - Building dependency graph...
  ✓ Gate 3 passed

→ Ready for iterative development
  Use /architect iterate <module> to begin

/architect phase

Execute a specific phase.

/architect phase 2

→ Executing Architecture Design phase
  - Creating system blueprint...
  - Defining module boundaries...
  - Documenting interfaces...
  ✓ Phase 2 complete

/architect iterate

Iterate on a specific module.

/architect iterate core

→ Iteration 1/?
  - Building...
  - Testing...
  - Score: 75/100
  → Improvements needed, continuing...

→ Iteration 2/?
  - Building...
  - Testing...
  - Score: 88/100
  → Improvements needed, continuing...

→ Iteration 3/?
  - Building...
  - Testing...
  - Score: 92/100
  ✓ Module complete!

/architect status

Show current architecture status.

Project: Text Classification System
Status: Phase 4 - Iterative Development

Modules:
  ├── core        [██████████] 92% ✓
  ├── api         [████████░░] 80% → iterating
  ├── cli         [░░░░░░░░░░] 0%  pending
  └── gui         [░░░░░░░░░░] 0%  pending

Current Iteration: api - Iteration 2
Next Module: cli

Overall Progress: 43%

/architect review

Review and validate current architecture.

→ Architecture Review

✓ Module cohesion: Good
✓ Interface contracts: Clear
✓ Test coverage: 85%
✓ Documentation: Complete

⚠ Warnings:
  - API module has 2 TODOs
  - Performance not yet validated

→ Recommendation: Address warnings before integration phase

Rules

Integration with Other Skills

The master-architect skill integrates with other skills at specific points. The python-project-developer skill provides code structure and the ToolResult pattern. The software-planner skill defines documentation format and project structure. The iteration-manager skill manages per-module iteration cycles. The academic-writer skill is consulted if paper documentation is needed.

Best Practices

Five principles guide the architecture process. First, never skip phases because each phase builds on the previous one. Second, quality gates are mandatory with no exceptions. Third, iterate until excellence and settle for nothing less than 90%. Fourth, document continuously rather than leaving documentation for the end. Fifth, validate assumptions by testing early and testing often.

平台分布

Codex

37.41%
按下载量换算31

Claude

30.77%
按下载量换算26

Cursor

18.14%
按下载量换算15

Gemini CLI

10.61%
按下载量换算9

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills