Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计通过

architecture-review架构评论

Agent Skill

architecture-review 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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

269

周安装

11

GitHub Stars

58

下载量

87
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

2

许可证

unknown

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:architecture-review(架构评论)
来源仓库:https://github.com/codenamev/ai-software-architect
仓库路径:skills/architecture-review
安装命令:
npx skills add https://github.com/codenamev/ai-software-architect --skill architecture-review
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/codenamev/ai-software-architect --skill architecture-review

简介

architecture-review 执行多视角架构评审,协调团队成员从各自专业角度分析系统结构。

  • 适用于版本发布、功能开发或组件重构前的系统性架构评估与协作讨论。
  • 自动加载团队配置,生成结构化评审报告,支持优先级排序与问题追踪。
  • 安装前应确认权限范围、仓库访问能力及是否允许执行 git 或文件操作。
  • 建议结合 .architecture/members.yml 配置与模板文档理解具体流程要求。

SKILL.md

Architecture Review

Conducts comprehensive multi-perspective architecture reviews with all team members.

Process Overview

  1. Determine Scope - Identify what to review (version, feature, or component)
  2. Load Team - Read members from .architecture/members.yml and check pragmatic mode
  3. Analyze System - Examine architecture using Read, Glob, Grep, and git tools
  4. Individual Reviews - Each member reviews from their specialized perspective
  5. Collaborative Discussion - Synthesize findings and establish priorities
  6. Create Document - Generate comprehensive review using template
  7. Report Results - Summarize findings and next steps for user

Detailed guidance: references/review-process.md

Workflow Steps

1. Determine Scope

Identify review target and create filename:

  • Version: "version X.Y.Z" → X-Y-Z.md
  • Feature: "feature name" → feature-kebab-case.md
  • Component: "component name" → component-kebab-case.md

Apply input validation (see _patterns.md § Filename Sanitization).

2. Load Configuration and Team

cat .architecture/config.yml     # Check pragmatic_mode.enabled
cat .architecture/members.yml    # Load all members

Include Pragmatic Enforcer if pragmatic mode enabled for reviews.

3. Analyze the Target

Use available tools to examine the system:

  • Read - Code, configs, documentation
  • Glob - Find files by pattern
  • Grep - Search for specific patterns
  • Bash(git:*) - Git history and status

Focus based on review type:

  • Version: Overall architecture, components, patterns, technical debt
  • Feature: Implementation, integration, security, performance
  • Component: Structure, dependencies, boundaries, interfaces

4. Conduct Individual Member Reviews

For each member in members.yml, write a review including:

  • Perspective statement
  • Key observations (3-5)
  • Strengths (3-5)
  • Concerns with impact and recommendations (3-7)
  • Prioritized recommendations with effort estimates (3-7)

Format details: references/review-process.md § Individual Member Review Format

Pragmatic integration: If enabled, add pragmatic analysis after each member. See references/pragmatic-integration.md

5. Facilitate Collaborative Discussion

Synthesize findings:

  • Identify common concerns
  • Discuss disagreements
  • Establish consensus
  • Prioritize: Critical (0-2 weeks) | Important (2-8 weeks) | Nice-to-Have (2-6 months)

Discussion format: references/review-process.md § Collaborative Discussion

6. Create Review Document

Load template and fill in all sections:

cat .claude/skills/architecture-review/assets/review-template.md

Include:

  • Executive summary and overall assessment
  • Individual member reviews
  • Collaborative discussion
  • Consolidated findings (strengths, improvements, debt, risks)
  • Recommendations (immediate, short-term, long-term)
  • Success metrics and follow-up plan

Save to .architecture/reviews/[filename].md

Template: assets/review-template.md

7. Report to User

Architecture Review Complete: [Target]

Location: .architecture/reviews/[filename].md
Overall Assessment: [Strong | Adequate | Needs Improvement]

Top 3 Priorities:
1. [Priority 1]
2. [Priority 2]
3. [Priority 3]

Immediate Actions:
- [Action 1]
- [Action 2]

Next Steps:
- Review with team
- "Start architecture recalibration for [target]"
- Create ADRs for key decisions

Related Skills

Before: architecture-status, list-members During: specialist-review, create-adr After: architecture-recalibration, create-adr

Documentation

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

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

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

Codex

35.58%
按下载量换算31

Claude

32.26%
按下载量换算28

Cursor

19.55%
按下载量换算17

Gemini CLI

9.4%
按下载量换算8

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills