mcpField
M
mongodb-atlas

mongodb-atlas

github.com/thelobbi数据与数据库2151 次安装1 天前更新
代码仓库

关于 mongodb-atlas

mongodb-atlas 是面向 MongoDB 与 MongoDB Atlas 开发工作的技能，适用于处理数据模型、查询和数据库运维的开发者。内容涵盖 Prisma、索引、聚合、性能分析及 Atlas CLI 示例。

主要功能

  1. 01MongoDB 查询与模式示例
  2. 02聚合管道模式
  3. 03Prisma MongoDB 操作
  4. 04索引与查询优化
  5. 05Atlas CLI 集群命令

适用场景

  1. 01设计 MongoDB 模式与集合
  2. 02构建成员数据聚合报告
  3. 03分析索引使用情况和慢查询
  4. 04通过 Atlas CLI 管理集群

安装

复制下面的命令，在终端中运行。

$npx skills add https://github.com/thelobbi/claude --skill mongodb-atlas

安装后在 Claude Code 中输入/mongodb-atlas确认已加载。

详细介绍

MongoDB Atlas 技能

为 Alpha Members Platform 提供全面的 MongoDB 和 MongoDB Atlas 能力。

何时使用此技能

在处理以下任务时启用此技能：

  • MongoDB 查询开发
  • 架构设计与建模
  • 聚合管道
  • 索引优化
  • Prisma MongoDB 集成
  • Atlas 集群管理
  • 数据库性能调优

快速参考

Docker 命令
# Start local MongoDB
docker-compose up mongodb -d

# Connect to MongoDB shell
docker exec -it mongodb mongosh -u root -p rootpassword

# Check MongoDB logs
docker logs mongodb -f --tail=100
MongoDB Shell 命令
// Switch to database
use member_db

// List collections
show collections

// Find documents
db.members.find({ status: "ACTIVE" }).limit(10)

// Insert document
db.members.insertOne({ email: "[email protected]", status: "PENDING" })

// Update document
db.members.updateOne(
  { email: "[email protected]" },
  { $set: { status: "ACTIVE" } }
)

// Aggregation
db.members.aggregate([
  { $match: { status: "ACTIVE" } },
  { $group: { _id: "$profile.company", count: { $sum: 1 } } }
])
连接字符串
# Local Docker
mongodb://root:rootpassword@localhost:27017/member_db?authSource=admin

# Atlas (example)
mongodb+srv://user:[email protected]/member_db?retryWrites=true&w=majority

架构设计

Prisma MongoDB 架构
datasource db {
  provider = "mongodb"
  url      = env("DATABASE_URL")
}

model Member {
  id              String          @id @default(auto()) @map("_id") @db.ObjectId
  keycloakUserId  String          @unique
  email           String          @unique
  firstName       String
  lastName        String
  status          MemberStatus    @default(PENDING)
  profile         MemberProfile?  // Embedded
  addresses       MemberAddress[] // Embedded array
  memberships     Membership[]    // Relation
  createdAt       DateTime        @default(now())
  updatedAt       DateTime        @updatedAt

  @@index([status])
  @@index([email])
}

type MemberProfile {
  jobTitle    String?
  company     String?
  timezone    String  @default("UTC")
}

type MemberAddress {
  type        AddressType
  street1     String
  city        String
  state       String
  postalCode  String
}

集合

members
{
  _id: ObjectId("..."),
  keycloakUserId: "uuid",
  email: "[email protected]",
  firstName: "John",
  lastName: "Doe",
  status: "ACTIVE",
  profile: {
    jobTitle: "Engineer",
    company: "Acme Corp"
  },
  addresses: [
    { type: "HOME", street1: "123 Main St", ... }
  ],
  createdAt: ISODate("..."),
  updatedAt: ISODate("...")
}
memberships
{
  _id: ObjectId("..."),
  memberId: ObjectId("..."),
  type: "INDIVIDUAL",
  level: "PREMIUM",
  status: "ACTIVE",
  startDate: ISODate("..."),
  payment: {
    method: "card",
    amount: 99.99,
    currency: "USD"
  }
}
member_activities
{
  _id: ObjectId("..."),
  memberId: ObjectId("..."),
  action: "LOGIN",
  category: "AUTH",
  ipAddress: "192.168.1.1",
  createdAt: ISODate("...")
}

索引

推荐索引
// Members collection
db.members.createIndex({ keycloakUserId: 1 }, { unique: true })
db.members.createIndex({ email: 1 }, { unique: true })
db.members.createIndex({ status: 1, createdAt: -1 })
db.members.createIndex({ "profile.company": 1 })

// Text search
db.members.createIndex({
  firstName: "text",
  lastName: "text",
  email: "text",
  "profile.company": "text"
})

// Memberships
db.memberships.createIndex({ memberId: 1 })
db.memberships.createIndex({ status: 1, endDate: 1 })

// Activities (with TTL)
db.member_activities.createIndex({ memberId: 1, createdAt: -1 })
db.member_activities.createIndex(
  { createdAt: 1 },
  { expireAfterSeconds: 31536000 } // 1 year TTL
)

聚合管道

成员统计
db.members.aggregate([
  { $facet: {
      byStatus: [
        { $group: { _id: "$status", count: { $sum: 1 } } }
      ],
      byMonth: [
        { $group: {
            _id: { $dateToString: { format: "%Y-%m", date: "$createdAt" } },
            count: { $sum: 1 }
          }
        },
        { $sort: { _id: 1 } }
      ],
      total: [{ $count: "count" }]
    }
  }
])
成员及其会员关系
db.members.aggregate([
  { $match: { _id: ObjectId("...") } },
  { $lookup: {
      from: "memberships",
      localField: "_id",
      foreignField: "memberId",
      as: "memberships"
    }
  }
])

Prisma 操作

// Find member
const member = await prisma.member.findUnique({
  where: { email: "[email protected]" },
  include: { memberships: true }
});

// Create member with embedded profile
const newMember = await prisma.member.create({
  data: {
    email: "[email protected]",
    firstName: "Jane",
    lastName: "Doe",
    keycloakUserId: "uuid",
    profile: {
      jobTitle: "Manager",
      company: "Acme"
    }
  }
});

// Update with push to array
const updated = await prisma.member.update({
  where: { id: "..." },
  data: {
    addresses: {
      push: {
        type: "WORK",
        street1: "456 Office Blvd",
        city: "Boston",
        state: "MA",
        postalCode: "02102"
      }
    }
  }
});

// Aggregation with Prisma
const stats = await prisma.member.groupBy({
  by: ['status'],
  _count: true
});

性能优化

查询优化
// Use projection to limit returned fields
db.members.find(
  { status: "ACTIVE" },
  { firstName: 1, lastName: 1, email: 1 }
)

// Use covered queries
db.members.find(
  { email: "[email protected]" },
  { _id: 0, email: 1, status: 1 }
).hint({ email: 1, status: 1 })

// Explain query plan
db.members.find({ status: "ACTIVE" }).explain("executionStats")
索引分析
// Get index usage stats
db.members.aggregate([{ $indexStats: {} }])

// Find unused indexes
db.members.aggregate([
  { $indexStats: {} },
  { $match: { "accesses.ops": 0 } }
])

Atlas CLI 命令

# Login to Atlas
atlas auth login

# List clusters
atlas clusters list --projectId <projectId>

# Create cluster
atlas clusters create alpha-dev \
  --projectId <projectId> \
  --provider AWS \
  --region US_EAST_1 \
  --tier M10

# Get connection string
atlas clusters connectionStrings describe alpha-dev

项目文件

  • Prisma 架构：prisma/schema.prisma
  • MongoDB 初始化：infrastructure/mongo-init/01-init-db.js
  • Docker：docker/docker-compose.yml（mongodb 服务）

相关 Agent

  • mongodb-atlas-admin - 集群管理
  • mongodb-schema-designer - 架构设计
  • mongodb-query-optimizer - 性能调优
  • mongodb-aggregation-specialist - 复杂查询

故障排除

# Check MongoDB connection
docker exec mongodb mongosh --eval "db.adminCommand('ping')"

# Check replication status
docker exec mongodb mongosh --eval "rs.status()"

# Analyze slow queries
db.setProfilingLevel(1, { slowms: 100 })
db.system.profile.find().sort({ millis: -1 }).limit(10)

评论

支持 Markdown