[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"pack-detail-backend-engineer-ai-toolkit-fr":3,"seo:pack:backend-engineer-ai-toolkit:fr":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","Nouveau · cette semaine","Boîte à Outils IA du Développeur Backend","Neuf outils pour ingénieurs Go\u002FRust\u002FPython\u002FNode qui veulent que leurs agents IA aident vraiment sur les schémas, contrats, scaffolding et debug en prod — pas juste compléter une boucle for.",[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.",78,{"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",30,"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",89,"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...",26,{"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...",28,{"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:...",23,{"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":25,"title":94,"metaDescription":95,"h1":96,"tldr":97,"bodyMarkdown":98,"faq":99,"schema":115,"internalLinks":121,"citations":134,"wordCount":147,"generatedAt":148},"pack","Backend Engineer's AI Toolkit — 9 Picks for Go\u002FRust\u002FPython\u002FNode Builders","PostgreSQL MCP, Postgres MCP Pro, Backend Architect, FastAPI, OpenAPI Generator, Golang Pro, Microservices Architect, Debugger, Logfire. Nine tools so AI agents help with schemas, contracts, scaffolding and prod debugging — installed in deliberate order, with the tradeoffs spelled out.","Backend Engineer's AI Toolkit — Nine Picks That Earn Their Place in CLAUDE.md","A working backend engineer's rig for letting AI agents touch the database, the API contract, the service skeleton, and the prod incident — without making things worse. Data layer first, API layer next, scaffolding and language pros third, debugging and observability last. Opinionated install order, no fan-fiction stats.","## What's in this pack\n\nAudience: you write services for a living in **Go, Rust, Python, or Node**. You already have an LLM in your editor. The thing it's still bad at is anything that touches **production reality** — a real schema with 80 columns, a contract that downstream teams ratified six months ago, a microservice topology you didn't choose, a 3 a.m. incident where the trace just stops.\n\nThis pack is nine tools that close that gap. Every pick is **operable by an agent (MCP server or skill)**, not just a library you could already `go get`. Three buckets:\n\n- **Data layer** — give the agent safe, structured access to Postgres so it can read schemas, propose indexes, and dry-run SQL without burning your production cluster.\n- **API + service layer** — let it generate framework-correct handlers, OpenAPI clients, and Go service skeletons that match your house style.\n- **Debugging + observability** — when something is on fire, give it a real debugger and real telemetry, not a `print` statement.\n\n## Install in this order\n\nOrder matters. Each tier unlocks the next.\n\n1. **PostgreSQL MCP** (id 660) — start here. A read-only MCP server pointed at a non-prod replica. Now the agent can `describe table users`, list indexes, sample 100 rows. This single step changes \"AI writes plausible SQL\" into \"AI writes SQL that compiles against your actual schema.\"\n2. **Postgres MCP Pro** (id 3283) — once read-only is comfortable, add the index-tuning + safe-SQL layer. Same idea, broader surface: `EXPLAIN` integration, index recommendations, write tools behind a safety mode. Don't install this before #1; you want to have already learned where the agent gets schema things wrong.\n3. **Backend Architect agent** (id 4367) — a Claude Code skill that takes a feature description and outputs a service shape: tables, endpoints, queue boundaries, failure modes. Read-only, no code yet — this is the thinking step.\n4. **FastAPI** (id 839) — opinionated, type-driven Python framework. Even if you ship Go in production, FastAPI is the fastest way to give an agent a target stack for prototyping or building internal AI endpoints. Pydantic models double as the LLM's schema source of truth.\n5. **OpenAPI Generator** (id 2668) — the contract layer. Hand the agent your OpenAPI spec, generate clients + server stubs in 30 languages. Stop letting agents write hand-rolled HTTP clients that drift from the spec.\n6. **Golang Pro agent** (id 4524) — a Go-specialist Claude Code agent. Knows idiomatic error wrapping, `context.Context` propagation, table tests, and the standard project layout discussions. Use after Backend Architect has set the shape.\n7. **Microservices Architect agent** (id 4433) — for when the feature crosses service boundaries. Talks about service boundaries, transactional outbox, idempotency keys, saga vs 2PC. Optional — skip if you're a monolith team and proud of it.\n8. **Debugger agent** (id 4393) — a Claude Code skill that runs a four-phase root-cause protocol: reproduce, isolate, hypothesize, verify. Pair with a real debugger (Delve \u002F pdb \u002F node --inspect). Replaces the failure mode where the LLM \"fixes\" the symptom by adding a try\u002Fexcept.\n9. **Logfire** (id 3235) — Python observability on OpenTelemetry. Even for non-Python services, the OTel pipeline + Logfire's structured views are a sane default to wire spans into so the debugger agent has something to read at 3 a.m.\n\n## How they fit together\n\n```\n         ┌─────────────────────┐\n         │  Backend Architect  │  ← \"design this feature\"\n         │     (4367)          │     (talks, doesn't code yet)\n         └──────────┬──────────┘\n                    │ schema + endpoint plan\n                    ▼\n   ┌────────────────┴────────────────┐\n   │       Data layer (read-only)    │\n   │  PostgreSQL MCP (660)           │  ← agent reads real schema\n   │  Postgres MCP Pro (3283)        │  ← then proposes indexes \u002F dry-run SQL\n   └────────────────┬────────────────┘\n                    │\n                    ▼\n   ┌─────────────────────────────────┐\n   │       API + service layer       │\n   │  OpenAPI Generator (2668) ──┐   │  ← contract is source of truth\n   │                             │   │\n   │  FastAPI (839)   Golang Pro │   │  ← agents write stack-correct code\n   │                  (4524)     │   │\n   │  Microservices Architect (4433) │  ← boundary calls, not in-process\n   └────────────────┬────────────────┘\n                    │\n                    ▼\n   ┌─────────────────────────────────┐\n   │   Debug + observability         │\n   │  Logfire (3235) ─── spans ──┐   │\n   │                             ▼   │\n   │  Debugger agent (4393) ◀────────┤  ← reads spans + repros bug\n   └─────────────────────────────────┘\n```\n\nThe loop that matters: **Architect drafts → MCP servers ground the draft in real schema → code-gen agents emit framework-correct code → observability + debugger close the feedback when production tells the truth.** Skip any one tier and the next tier starts hallucinating.\n\n## Tradeoffs you'll hit\n\n- **Read-only DB MCP vs write-enabled** — PostgreSQL MCP (660) is read-only by default, Postgres MCP Pro (3283) opens up write tools behind a safety mode. Start read-only on a non-prod replica for at least two weeks; only widen once you trust the agent's SQL.\n- **ORM-aware agent vs raw-SQL agent** — Backend Architect and Golang Pro don't assume an ORM. If your house style is GORM\u002Fsqlx\u002FSQLAlchemy, paste a CLAUDE.md snippet stating that — otherwise you'll get half raw-SQL, half ORM proposals on the same PR.\n- **Monolith vs microservice tooling** — Microservices Architect (4433) is genuinely useful for boundary work, but it tends to recommend a service split where a package would do. If you're under ~5 services, install but use sparingly.\n- **Hand-rolled clients vs OpenAPI Generator** — `openapi-generator` produces a lot of code, some of it ugly. The win isn't pretty code; it's that drift between spec and client becomes a compile error instead of a 4 a.m. bug.\n- **Logfire vs your existing OTel stack** — Logfire is Python-flavored but speaks OTel, so a Go or Rust service can export spans there too. If you already run Tempo or Honeycomb, treat Logfire as a side-channel for AI-readable views rather than ripping out what works.\n\n## Common pitfalls\n\n- **Pointing DB MCP at production** — don't. Use a logical replica, a read-only role, or a sanitized snapshot. The agent will eventually try a 12-way join on `events` and pin a CPU.\n- **Letting the agent invent endpoints not in the OpenAPI spec** — once #5 is installed, gate code review on \"does the spec say this endpoint exists.\" Otherwise the contract becomes fiction.\n- **Stacking two architecture agents on one task** — running Backend Architect + Microservices Architect on the same prompt produces 4× the text and zero extra signal. Pick one per planning round.\n- **Skipping Logfire because \"we have Datadog\"** — fine, skip it. But your debugger agent then has nothing to read; either point it at your existing OTel exporter or accept that #8's effectiveness drops in half.\n- **Treating the Debugger agent as a 1-shot \"fix this\" button** — it's a four-phase protocol; if the first phase (reproduce) fails, stop, don't let it skip to a guess-fix.",[100,103,106,109,112],{"q":101,"a":102},"Why so many MCP \u002F agent picks instead of just libraries?","Because the gap a backend engineer actually feels in 2026 isn't \"I don't have a library\" — it's \"my agent doesn't understand my production reality.\" Every pick here either gives the agent grounded context (Postgres MCP, Logfire) or a role-shaped skill (Architect, Golang Pro, Debugger). A plain library would help you; an MCP server or skill helps your agent help you.",{"q":104,"a":105},"I write Go. Why is FastAPI in here?","Two reasons. One, FastAPI's Pydantic-driven type system is the cleanest target stack for an agent to prototype against — even a Go team uses it to spin up internal AI endpoints in a Saturday afternoon. Two, the request\u002Fresponse model schemas double as the agent's source of truth for what your API actually shapes look like. If you're 100% Go and don't want any Python at all, drop pick #4 and keep the other eight; the rig still works.",{"q":107,"a":108},"Will the Debugger agent actually find root causes?","Only if you feed it real signals. Stack trace alone — it'll guess. Stack trace + OTel span tree from Logfire + a failing test it can rerun — it'll work through the four-phase protocol (reproduce \u002F isolate \u002F hypothesize \u002F verify) much closer to how a senior engineer would. The agent is a process, not magic; the rest of the pack exists to give that process inputs.",{"q":110,"a":111},"Postgres MCP vs Postgres MCP Pro — pick one?","Install both, in that order. PostgreSQL MCP (660) is the lightweight read-only entry point — fast to set up, low blast radius. Postgres MCP Pro (3283) adds index tuning and write-mode safety gates that are valuable once the agent is past the 'doesn't know your schema' phase. Doing Pro first skips the cheap lesson about how often the agent misreads a join.",{"q":113,"a":114},"Can I run this whole stack with one agent (Claude Code), or do I need multiple?","One agent is fine. The 'agents' in this pack (Backend Architect, Golang Pro, Microservices Architect, Debugger) are Claude Code skills\u002Fsubagents — they install into the same Claude Code session and switch in based on the task. The MCP servers (Postgres, Logfire) plug into the same session over the MCP protocol. You don't need a multi-agent framework to run this pack.",{"@context":116,"@type":117,"name":118,"description":119,"numberOfItems":120,"inLanguage":25},"https:\u002F\u002Fschema.org","ItemList","Backend Engineer's AI Toolkit","Nine MCP servers and Claude Code skills curated for backend engineers in Go, Rust, Python and Node — covering data layer, API contracts, scaffolding, debugging and observability, in deliberate install order.",9,[122,126,130],{"url":123,"anchor":124,"reason":125},"\u002Fen\u002Fai-tools-for\u002Fmcp","Browse all MCP servers","Five of the nine picks are MCP servers — the catalog has more if you outgrow this pack",{"url":127,"anchor":128,"reason":129},"\u002Fen\u002Fai-tools-for\u002Fagents","Claude Code subagents directory","Backend Architect, Golang Pro, Microservices Architect and Debugger live in the broader agents catalog",{"url":131,"anchor":132,"reason":133},"\u002Fen\u002Ftopics","Other topic packs","Adjacent packs cover frontend, AI side hustles, and Mac engineering setup",[135,139,143],{"claim":136,"source_name":137,"source_url":138},"FastAPI is a modern, fast Python web framework based on Pydantic and type hints","FastAPI official docs","https:\u002F\u002Ffastapi.tiangolo.com\u002F",{"claim":140,"source_name":141,"source_url":142},"OpenAPI Generator produces client SDKs and server stubs from OpenAPI specs across 30+ languages","OpenAPI Generator project","https:\u002F\u002Fopenapi-generator.tech\u002F",{"claim":144,"source_name":145,"source_url":146},"Logfire is a Python observability platform built on OpenTelemetry by Pydantic","Logfire docs","https:\u002F\u002Flogfire.pydantic.dev\u002Fdocs\u002F",920,"2026-05-22T00:00:00Z"]