[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"pack-detail-backend-engineer-ai-toolkit-zh":3,"seo:pack:backend-engineer-ai-toolkit:zh":92},{"code":4,"message":5,"data":6},200,"操作成功",{"pack":7},{"slug":8,"icon":9,"tone":10,"status":11,"status_label":12,"title":13,"description":14,"items":15,"install_cmd":91},"backend-engineer-ai-toolkit","⚙️","#1F2937","new","本周新建","后端工程师 AI 工具包","9 件套，给 Go\u002FRust\u002FPython\u002FNode 后端工程师：让 AI agent 真正帮你处理数据库 schema、API 契约、脚手架、生产环境排障 — 不是只补一个 for 循环。",[16,28,35,45,53,63,70,77,84],{"id":17,"uuid":18,"slug":19,"title":20,"description":21,"author_name":22,"view_count":23,"vote_count":24,"lang_type":25,"type":26,"type_label":27},660,"faa28c56-1377-4109-9e8f-a9b2be119d16","postgresql-mcp-sql-database-server-ai-agents-faa28c56","PostgreSQL MCP — SQL Database Server for AI Agents","MCP server that gives AI agents direct access to PostgreSQL databases. Run queries, explore schemas, manage tables, and analyze data through natural language. 3,000+ stars.","MCP Hub",211,0,"en","mcp","MCP",{"id":29,"uuid":30,"slug":31,"title":32,"description":33,"author_name":22,"view_count":34,"vote_count":24,"lang_type":25,"type":26,"type_label":27},3283,"216cb667-d5ae-5400-99d5-63dd528e1690","postgres-mcp-pro-index-tuning-safe-sql-tools","Postgres MCP Pro — Index Tuning + Safe SQL Tools","Postgres MCP Pro is an MCP server for PostgreSQL that runs safe SQL, explains plans, and recommends indexes so agents can tune databases faster.",69,{"id":36,"uuid":37,"slug":38,"title":39,"description":40,"author_name":41,"view_count":42,"vote_count":24,"lang_type":25,"type":43,"type_label":44},4367,"cb075415-dd85-4008-aec1-4b9781cc2bda","claude-code-agent-backend-architect-cb075415","Claude Code Agent: Backend Architect","Backend system architecture and API design specialist. Use PROACTIVELY for greenfield service design, monolith decomposition, API paradigm selection (REST\u002FgRPC\u002FGraphQL),...","TokRepo精选",28,"skill","Skill",{"id":46,"uuid":47,"slug":48,"title":49,"description":50,"author_name":51,"view_count":52,"vote_count":24,"lang_type":25,"type":43,"type_label":44},839,"00db0ed8-cdb7-4a83-bf67-a2fcae16f6bf","fastapi-build-ai-backend-apis-minutes-00db0ed8","FastAPI — Build AI Backend APIs in Minutes","Modern Python web framework for building AI backend APIs. FastAPI provides automatic OpenAPI docs, async support, Pydantic validation, and the fastest Python web performance.","Script Depot",188,{"id":54,"uuid":55,"slug":56,"title":57,"description":58,"author_name":59,"view_count":60,"vote_count":24,"lang_type":25,"type":61,"type_label":62},2668,"a951911e-4838-11f1-9bc6-00163e2b0d79","openapi-generator-generate-client-sdks-server-stubs-api-a951911e","OpenAPI Generator — Generate Client SDKs and Server Stubs from API Specs","A code generation tool that produces client libraries, server stubs, API documentation, and configuration from OpenAPI 2.0\u002F3.x specifications in over 50 languages.","AI Open Source",85,"config","Config",{"id":64,"uuid":65,"slug":66,"title":67,"description":68,"author_name":41,"view_count":69,"vote_count":24,"lang_type":25,"type":43,"type_label":44},4524,"f7ce3bb8-0b3e-44ff-b325-6e6362059a43","claude-code-agent-golang-pro-f7ce3bb8","Claude Code Agent: Golang Pro","Use when building Go applications requiring concurrent programming, high-performance systems, microservices, or cloud-native architectures where idiomatic patterns, error handling excellence, and efficiency are critical. Specifically:\\\\n\\\\n\u003Cexample>\\\\nConte...",23,{"id":71,"uuid":72,"slug":73,"title":74,"description":75,"author_name":41,"view_count":76,"vote_count":24,"lang_type":25,"type":43,"type_label":44},4433,"853ee7a5-7149-49bb-9cf9-43692f89f748","claude-code-agent-microservices-architect-853ee7a5","Claude Code Agent: Microservices Architect","Use when designing distributed system architecture, decomposing monolithic applications into independent microservices, or establishing communication patterns between services...",27,{"id":78,"uuid":79,"slug":80,"title":81,"description":82,"author_name":41,"view_count":83,"vote_count":24,"lang_type":25,"type":43,"type_label":44},4393,"3259265d-42dc-4ab3-84c7-f910c20f934e","claude-code-agent-debugger-3259265d","Claude Code Agent: Debugger","Use this agent when you need to diagnose and fix bugs, identify root causes of failures, or analyze error logs and stack traces to resolve issues. Specifically: Context:...",21,{"id":85,"uuid":86,"slug":87,"title":88,"description":89,"author_name":51,"view_count":90,"vote_count":24,"lang_type":25,"type":43,"type_label":44},3235,"67236d34-9663-4f58-a975-0a218f099c6d","logfire-python-observability-on-opentelemetry","Logfire — Python Observability on OpenTelemetry","Logfire is Pydantic’s Python SDK for traces\u002Fmetrics\u002Flogs on OpenTelemetry, helping teams add observability with minimal code and query data with SQL.",44,"tokrepo install pack\u002Fbackend-engineer-ai-toolkit",{"pageType":93,"pageKey":8,"locale":94,"title":95,"metaDescription":96,"h1":97,"tldr":98,"bodyMarkdown":99,"faq":100,"schema":116,"internalLinks":122,"citations":135,"wordCount":148,"generatedAt":149},"pack","zh","后端工程师 AI 工具包 — Go\u002FRust\u002FPython\u002FNode 工程师的 9 件套","PostgreSQL MCP \u002F Postgres MCP Pro \u002F Backend Architect \u002F FastAPI \u002F OpenAPI Generator \u002F Golang Pro \u002F Microservices Architect \u002F Debugger \u002F Logfire。9 个工具让 AI agent 真正帮你处理 schema、API 契约、脚手架、生产排障 — 含推荐安装顺序和取舍说明。","后端工程师 AI 工具包 — 真能写进 CLAUDE.md 的 9 件套","一个在职后端工程师的 AI 战机配置：让 agent 安全地碰数据库、API 契约、服务骨架、线上事故 — 又不把事情搞砸。先数据层，再 API 层，然后脚手架 + 语言专家，最后调试 + 可观测。安装顺序经过精心安排，全是真东西、没有编造的数字。","## 这个 pack 包含什么\n\n受众：你以写 **Go \u002F Rust \u002F Python \u002F Node** 服务为生。编辑器里已经有 LLM。它现在还做不好的事情是：任何**碰到生产现实**的部分 — 80 列的真 schema、半年前下游团队批准的契约、不是你选的微服务拓扑、凌晨 3 点 trace 突然断掉的事故。\n\n这个 pack 9 个工具就是为这个缺口设计的。每一个都**可以被 agent 直接操作**（MCP server 或 skill），不是又一个你 `go get` 就能装的库。三层：\n\n- **数据层** — 给 agent 一个安全、有结构的 Postgres 通道，让它能读 schema、提索引建议、dry-run SQL，又不烧掉你的生产集群。\n- **API + 服务层** — 让它能生成框架正确的 handler、OpenAPI 客户端、符合团队风格的 Go 服务骨架。\n- **调试 + 可观测** — 出事故的时候，给它真的 debugger 和真的 telemetry，不是一个 `print`。\n\n## 推荐安装顺序\n\n顺序很重要。每一层解锁下一层。\n\n1. **PostgreSQL MCP**（id 660）— 从这里开始。一个只读 MCP server，连到一个非生产副本上。这一步之后，agent 就能 `describe table users`、列出索引、采样 100 行。就这一个动作，把「AI 写看起来合理的 SQL」变成了「AI 写在你真实 schema 上能编译的 SQL」。\n2. **Postgres MCP Pro**（id 3283）— 只读用顺了，再加索引调优 + 安全 SQL 这一层。同样的思路，更大的面：`EXPLAIN` 集成、索引推荐、加了安全模式的写工具。**别**在第 1 步之前装这个 — 你需要先吃过 agent 误读 schema 的亏。\n3. **Backend Architect agent**（id 4367）— 一个 Claude Code skill：吃需求描述，吐出服务形状：表、端点、队列边界、失败模式。只动嘴不动代码 — 这是「想清楚」的环节。\n4. **FastAPI**（id 839）— 强意见、类型驱动的 Python 框架。哪怕你生产环境跑 Go，FastAPI 也是给 agent 做原型 \u002F 内部 AI 端点的最快目标栈。Pydantic 模型同时充当 LLM 的 schema 真相源。\n5. **OpenAPI Generator**（id 2668）— 契约层。把你的 OpenAPI spec 给 agent，30 种语言的客户端 + 服务端 stub 一把出。别再让 agent 写手撸 HTTP 客户端、然后跟 spec 慢慢漂移了。\n6. **Golang Pro agent**（id 4524）— 一个 Go 专家 Claude Code agent。懂幂等 error 包装、`context.Context` 传播、table test、标准项目布局。**在 Backend Architect 定下形状之后**用。\n7. **Microservices Architect agent**（id 4433）— 需求跨服务时上。讨论服务边界、transactional outbox、idempotency key、saga vs 2PC。可选 — 单体团队请绕开。\n8. **Debugger agent**（id 4393）— 一个跑四阶段 root cause 协议的 Claude Code skill：复现、隔离、假设、验证。配合真 debugger（Delve \u002F pdb \u002F node --inspect）用。专治「LLM 用 try\u002Fexcept 把症状盖过去」这个失败模式。\n9. **Logfire**（id 3235）— 基于 OpenTelemetry 的 Python 可观测。哪怕不是 Python 服务，OTel 管道 + Logfire 的结构化视图也是一个合理的默认选择，让 Debugger agent 在凌晨 3 点有东西可读。\n\n## 它们怎么协同\n\n```\n         ┌─────────────────────┐\n         │  Backend Architect  │  ← 「设计这个功能」\n         │     (4367)          │     （只说话，还不写代码）\n         └──────────┬──────────┘\n                    │ schema + 端点方案\n                    ▼\n   ┌────────────────┴────────────────┐\n   │      数据层（只读）              │\n   │  PostgreSQL MCP (660)           │  ← agent 读真 schema\n   │  Postgres MCP Pro (3283)        │  ← 再推索引 \u002F dry-run SQL\n   └────────────────┬────────────────┘\n                    │\n                    ▼\n   ┌─────────────────────────────────┐\n   │      API + 服务层                │\n   │  OpenAPI Generator (2668) ──┐   │  ← 契约即真相源\n   │                             │   │\n   │  FastAPI (839)   Golang Pro │   │  ← agent 写出栈正确的代码\n   │                  (4524)     │   │\n   │  Microservices Architect (4433) │  ← 跨边界调用而非进程内\n   └────────────────┬────────────────┘\n                    │\n                    ▼\n   ┌─────────────────────────────────┐\n   │   调试 + 可观测                  │\n   │  Logfire (3235) ─── spans ──┐   │\n   │                             ▼   │\n   │  Debugger agent (4393) ◀────────┤  ← 读 span + 重现 bug\n   └─────────────────────────────────┘\n```\n\n关键的闭环：**Architect 起草 → MCP server 把草稿落地到真 schema → 代码生成 agent 出框架正确的代码 → 可观测 + debugger 等生产说真话再回填反馈。** 任何一层缺位，下一层就开始幻觉。\n\n## 你会遇到的取舍\n\n- **只读 DB MCP vs 可写 DB MCP** — PostgreSQL MCP（660）默认只读，Postgres MCP Pro（3283）在安全模式后面开了写工具。先在非生产副本上跑只读至少两周；信任 agent 的 SQL 之后再放开。\n- **懂 ORM 的 agent vs 写裸 SQL 的 agent** — Backend Architect 和 Golang Pro 都不预设 ORM。如果团队规范是 GORM \u002F sqlx \u002F SQLAlchemy，在 CLAUDE.md 写一段说明 — 否则一个 PR 里会出现一半裸 SQL 一半 ORM。\n- **单体 vs 微服务工具** — Microservices Architect（4433）对边界类工作真有用，但它倾向于「能拆服务就拆」。服务数 \u003C 5 的团队可以装，但少用。\n- **手撸客户端 vs OpenAPI Generator** — `openapi-generator` 出的代码量大，部分丑。赢的不是代码好看 — 是 spec 和客户端漂移会变成编译错误，而不是凌晨 4 点的 bug。\n- **Logfire vs 既有 OTel 栈** — Logfire 偏 Python 风但说 OTel，Go \u002F Rust 服务也能把 span 导过去。如果已经在跑 Tempo 或 Honeycomb，就把 Logfire 当作给 AI 看的副通道，别替换已经能用的东西。\n\n## 常见踩坑\n\n- **DB MCP 指向生产** — 别。用逻辑副本、只读角色、或脱敏快照。agent 早晚会在 `events` 表上跑一个 12 路 join 把一个 CPU 卡死。\n- **让 agent 发明 OpenAPI spec 里没有的端点** — 第 5 步装完之后，code review 必须卡「spec 是否声明了这个端点」。否则契约就变成小说。\n- **同一个任务上叠两个架构 agent** — Backend Architect + Microservices Architect 一起跑同一个 prompt，文本量乘 4 信号量乘 0。每轮规划用一个。\n- **「我们有 Datadog 了所以不要 Logfire」** — 行，可以跳。但 Debugger agent 就没东西可读了；要么把它指向你既有的 OTel exporter，要么接受第 8 步效果减半。\n- **把 Debugger agent 当一键「fix this」按钮** — 它是个四阶段协议；如果第一阶段（复现）失败，停下，别让它跳到「猜一个 fix」。",[101,104,107,110,113],{"q":102,"a":103},"为啥这么多 MCP \u002F agent，不直接选库？","因为后端工程师在 2026 年真正感觉到的缺口不是「我缺一个库」 — 是「我的 agent 不理解我的生产现实」。这里每个选择，要么给 agent 落地的上下文（Postgres MCP \u002F Logfire），要么是一个角色化的 skill（Architect \u002F Golang Pro \u002F Debugger）。一个普通的库帮你；一个 MCP server 或 skill 帮你的 agent 帮你。",{"q":105,"a":106},"我写 Go，为啥还塞 FastAPI 进来？","两个原因。一，FastAPI 的 Pydantic 类型系统是 agent 做原型最干净的目标栈 — 哪怕一个 Go 团队，也能用它一个周六下午起一个内部 AI 端点。二，请求 \u002F 响应 model schema 同时充当 agent 对你 API 形状的真相源。如果你 100% Go、完全不想碰 Python，跳过第 4 步保留另外 8 个，整套依然成立。",{"q":108,"a":109},"Debugger agent 真能找出 root cause 吗？","前提是你喂它真信号。光丢个 stack trace — 它只会猜。stack trace + Logfire 的 OTel span 树 + 一个能重跑的失败 test — 它会走完四阶段（复现 \u002F 隔离 \u002F 假设 \u002F 验证），效果接近一个资深工程师。agent 是一个流程，不是魔法；这个 pack 其它工具就是给这个流程提供输入用的。",{"q":111,"a":112},"Postgres MCP 和 Postgres MCP Pro 二选一？","两个都装，按顺序。PostgreSQL MCP（660）是轻量只读入口 — 上手快、爆炸半径小。Postgres MCP Pro（3283）加索引调优 + 写模式安全闸门，在 agent 度过「不懂你 schema」阶段之后很有价值。先装 Pro 会跳过「agent 经常误读 join」这堂便宜的课。",{"q":114,"a":115},"整套能用一个 agent（Claude Code）跑完吗？还是要多 agent 系统？","一个就够。这个 pack 里的「agent」（Backend Architect \u002F Golang Pro \u002F Microservices Architect \u002F Debugger）都是 Claude Code 的 skill \u002F subagent — 它们装到同一个 Claude Code 会话里，按任务自动切换。MCP server（Postgres \u002F Logfire）通过 MCP 协议接到同一个会话上。不需要多 agent 框架。",{"@context":117,"@type":118,"name":13,"description":119,"numberOfItems":120,"inLanguage":121},"https:\u002F\u002Fschema.org","ItemList","为 Go \u002F Rust \u002F Python \u002F Node 后端工程师精选的 9 个 MCP server 和 Claude Code skill — 覆盖数据层、API 契约、脚手架、调试和可观测，含推荐安装顺序。",9,"zh-CN",[123,127,131],{"url":124,"anchor":125,"reason":126},"\u002Fzh\u002Fai-tools-for\u002Fmcp","浏览全部 MCP server","9 件套里有 5 个是 MCP server — 用熟了可以来目录里挑更多",{"url":128,"anchor":129,"reason":130},"\u002Fzh\u002Fai-tools-for\u002Fagents","Claude Code subagent 目录","Backend Architect \u002F Golang Pro \u002F Microservices Architect \u002F Debugger 都属于更大的 agent 目录",{"url":132,"anchor":133,"reason":134},"\u002Fzh\u002Ftopics","其它主题 pack","相邻 pack 涵盖前端、AI 副业、Mac 工程师装机等主题",[136,140,144],{"claim":137,"source_name":138,"source_url":139},"FastAPI 是一个基于 Pydantic 和类型注解的现代 Python web 框架","FastAPI 官方文档","https:\u002F\u002Ffastapi.tiangolo.com\u002F",{"claim":141,"source_name":142,"source_url":143},"OpenAPI Generator 从 OpenAPI spec 生成 30+ 种语言的客户端 SDK 和服务端 stub","OpenAPI Generator 项目主页","https:\u002F\u002Fopenapi-generator.tech\u002F",{"claim":145,"source_name":146,"source_url":147},"Logfire 是 Pydantic 团队基于 OpenTelemetry 构建的 Python 可观测平台","Logfire 文档","https:\u002F\u002Flogfire.pydantic.dev\u002Fdocs\u002F",900,"2026-05-22T00:00:00Z"]