engineering-standards

engineering-standards

Enforces team engineering standards across Git, code review, Rails, frontend, deployment, incidents, observability, API design, database, ADRs, and technical debt, with tools for branch name and commit message validation.

Category
访问服务器

README

                ███████╗███╗   ██╗ ██████╗    ███████╗████████╗██████╗ ███████╗
                ██╔════╝████╗  ██║██╔════╝    ██╔════╝╚══██╔══╝██╔══██╗██╔════╝
                █████╗  ██╔██╗ ██║██║  ███╗   ███████╗   ██║   ██║  ██║███████╗
                ██╔══╝  ██║╚██╗██║██║   ██║   ╚════██║   ██║   ██║  ██║╚════██║
                ███████╗██║ ╚████║╚██████╔╝██╗███████║   ██║   ██████╔╝███████║
                ╚══════╝╚═╝  ╚═══╝ ╚═════╝ ╚═╝╚══════╝   ╚═╝   ╚═════╝ ╚══════╝

                          ⚡ Engineering Standards MCP Server ⚡

                          Your engineering brain, exported as a protocol.

Good engineering habits shouldn't live in your head — they should be encoded, shareable, and enforceable.


     ┌──────────────┐          ┌──────────────────┐          ┌──────────────┐
     │              │  query   │                  │  read    │  ◆ git       │
     │  Claude Code ├─────────►│  MCP Server      ├─────────►│  ◆ reviews   │
     │  / Desktop   │          │  (this thing)    │          │  ◆ PRs       │
     │              │◄─────────┤                  │◄─────────┤  ◆ testing   │
     │              │ response │                  │ standard │  ◆ rails     │
     │              │          │                  │          │  ◆ frontend  │
     │              │          │                  │          │  ◆ deploy    │
     │              │          │                  │          │  ◆ incidents │
     │              │          │                  │          │  ◆ observe   │
     │              │          │                  │          │  ◆ api       │
     │              │          │                  │          │  ◆ database  │
     │              │          │                  │          │  ◆ ADRs      │
     │              │          │                  │          │  ◆ debt      │
     │              │          │                  │          │  ◆ staff     │
     └──────────────┘          └──────────────────┘          └──────────────┘
      YOU + AI                  Protocol Layer                 Standards DB

🛠️ Tools

Tool Description
get_standard Retrieve a specific engineering standard (14 available — see Standards Included below)
review_branch_name Check if a branch name follows conventions
review_commit_message Check if a commit message follows conventions
staff_engineer_review Get the staff engineer thinking checklist for a given phase of work

📚 Standards Included

General

Standard Key Covers
Git Conventions git Branch naming, commits, rebase workflow, commit hygiene
Code Review code_review Tone, structure, what to look for, giving and receiving feedback
PR Standards pr Size, description, pre-merge checklist, hotfix protocol
Staff Engineer Checklist staff_engineer Thinking prompts for each phase of work

Rails

Standard Key Covers
Rails Standards rails Architecture, security, performance, migrations, Hotwire, deployment
Testing Philosophy testing Rails/RSpec strategy, FactoryBot, system specs, N+1 detection, CI integration

Frontend

Standard Key Covers
Frontend Standards frontend TypeScript, components, accessibility, state management, performance, testing, security

Operations

Standard Key Covers
Deployment Standards deployment Three-phase migration strategy, strong_migrations, zero-downtime deploys, rollback strategy
Incident Response incident_response Severity levels, incident commander, postmortems, on-call, game days
Observability observability Structured logging, metrics (RED/USE), alerting, SLIs/SLOs, dashboards, tracing

Architecture

Standard Key Covers
API Design api_design REST conventions, versioning, error format, pagination, idempotency
Database Design database_design PostgreSQL/Rails, indexing, constraints, partitioning, connection management
Architecture Decision Records architecture_decisions ADR template, when to write them, lifecycle, ADR vs RFC
Technical Debt technical_debt Classification, severity, when to pay it down, prevention, metrics

🤖 Companion Agents

The MCP provides the what (your team's standards). These agents provide the how (code generation and implementation).

From awesome-claude-code-subagents:

Agent Role
rails-expert Rails 7.x/8.x implementation, version-aware patterns, Hotwire, deployment
frontend-developer React, Vue, Angular — TypeScript, accessibility, component architecture

When the MCP returns a standard, it includes a reference to the relevant agent. Claude Code can then delegate implementation to that agent automatically.


⚙️ Setup

1. Install

git clone https://github.com/rajgurung/claude-engineering-standards.git
cd claude-engineering-standards
npm install
npm run build

2. Register

Add to your MCP settings file:

Client Config file
Claude Code ~/.claude/.mcp.json (or project-level)
Claude Desktop ~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "engineering-standards": {
      "command": "node",
      "args": ["/absolute/path/to/engineering-standards-mcp/build/index.js"]
    }
  }
}

📋 Wiring It Into Your Projects

Once the MCP is registered, add a CLAUDE.md to your project root so Claude Code knows to consult it:

## Engineering Standards

This project follows team engineering standards via the engineering-standards MCP server.

### Before writing code
- Call `get_standard("rails")` for architecture and conventions
- Call `get_standard("testing")` for test strategy
- Call `get_standard("api_design")` before building or modifying API endpoints
- Call `get_standard("database_design")` before schema changes
- Call `get_standard("deployment")` before any migration or deploy work

### Before committing
- Call `review_branch_name` to validate your branch name
- Call `review_commit_message` to validate your commit message

### Before opening a PR
- Call `staff_engineer_review` with phase `before_pr`
- Call `get_standard("code_review")` for review expectations

### During incidents
- Call `get_standard("incident_response")` for severity levels and response protocol
- Call `get_standard("observability")` for debugging and monitoring guidance

### Implementation
- Use the rails-expert agent for Rails code generation
- Use the frontend-developer agent for frontend work

Adapt this to your stack — remove what doesn't apply, add project-specific rules. The CLAUDE.md stays small because the standards live in the MCP, not in every repo.


🎨 Customising

The standards are plain markdown files in src/standards/. Edit them to match your team's conventions, then rebuild:

src/standards/
├── git-conventions.md           ← branch naming, commits
├── code-review.md               ← review tone & structure
├── pr-standards.md              ← PR size, checklists
├── staff-engineer-checklist.md  ← thinking prompts
├── testing-philosophy.md        ← test strategy (Rails/RSpec)
├── rails-standards.md           ← Rails conventions + rails-expert agent
├── frontend-standards.md        ← Frontend conventions + frontend-developer agent
├── deployment-standards.md      ← Safe deploys, migrations, rollback strategy
├── incident-response.md         ← Severity levels, postmortems, on-call
├── observability.md             ← Logging, metrics, alerting, SLIs/SLOs
├── api-design.md                ← REST conventions, versioning, error format
├── database-design.md           ← PostgreSQL/Rails, indexing, partitioning
├── architecture-decisions.md    ← ADR template and lifecycle
└── technical-debt.md            ← Classification, prioritisation, prevention
npm run build

🧑‍💻 Development

npm run dev    # Watch mode — recompiles on change
npm run build  # One-off build
npm start      # Run the server

📄 License

MIT

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选