concierge MCP — Сценарии использования, Установка & Живое демо

Зачем использовать

Ключевые функции

Progressive disclosure — only expose tools relevant to the current workflow step
Workflow state shared across steps without manual plumbing
Semantic tool search — worth it once you cross ~100 tools
Multiple transports: stdio, HTTP, SSE
Session-isolated — concurrent workflows don't clobber each other

Живое демо

Как выглядит на практике

concierge.replay ▶ готово

0/0

Установка

Выберите клиент

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "concierge": {
      "command": "uvx",
      "args": [
        "concierge"
      ],
      "_inferred": true
    }
  }
}

Откройте Claude Desktop → Settings → Developer → Edit Config. Перезапустите после сохранения.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "concierge": {
      "command": "uvx",
      "args": [
        "concierge"
      ],
      "_inferred": true
    }
  }
}

Cursor использует ту же схему mcpServers, что и Claude Desktop. Конфиг проекта приоритетнее глобального.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "concierge": {
      "command": "uvx",
      "args": [
        "concierge"
      ],
      "_inferred": true
    }
  }
}

Щёлкните значок MCP Servers на боковой панели Cline, затем "Edit Configuration".

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "concierge": {
      "command": "uvx",
      "args": [
        "concierge"
      ],
      "_inferred": true
    }
  }
}

Тот же формат, что и Claude Desktop. Перезапустите Windsurf для применения.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "concierge",
      "command": "uvx",
      "args": [
        "concierge"
      ]
    }
  ]
}

Continue использует массив объектов серверов, а не map.

~/.config/zed/settings.json

{
  "context_servers": {
    "concierge": {
      "command": {
        "path": "uvx",
        "args": [
          "concierge"
        ]
      }
    }
  }
}

Добавьте в context_servers. Zed перезагружается автоматически.

claude mcp add concierge -- uvx concierge

Однострочная команда. Проверить: claude mcp list. Удалить: claude mcp remove.

Сценарии использования

Реальные сценарии: concierge

Wrap a giant REST API (200+ endpoints) as an MCP without blowing the context

👤 Platform engineers exposing internal APIs to LLMs ⏱ ~90 min advanced

Когда использовать: Your company's API has 300 endpoints and a naive MCP dumps all of them into the system prompt, wrecking latency and hit rate.

Предварительные требования

Python 3.9+ — pyenv / uv
OpenAPI spec or endpoint catalog — Your API gateway / Swagger

Поток

Scaffold with concierge-sdk

Using concierge-sdk, scaffold an MCP server that wraps my OpenAPI spec at ./openapi.yaml. Make it use semantic search rather than listing all tools up front.✓ Скопировано

→ Boilerplate code + search handler
Define workflow stages

Group the endpoints into 3 workflows: 'read operations', 'create operations', 'admin'. Each workflow exposes only its own tools.✓ Скопировано

→ Workflow definitions with tool allowlists
Test with Claude

Connect Claude Desktop to this server and verify that listing tools only shows the search tool + current workflow tools — not all 300.✓ Скопировано

→ Claude sees ~10 tools, not 300

Итог: A large API surface usable by an LLM without collapsing on system prompt length.

Подводные камни

Semantic search returns the wrong tool when descriptions are too similar — Write distinctive one-line tool descriptions; test search with held-out queries

Build a guided multi-step workflow (e.g. customer onboarding)

👤 Devs building structured agent flows ⏱ ~60 min advanced

Когда использовать: You want the LLM to follow a defined sequence: collect info → validate → create record → notify. Each step has its own tools.

Поток

Declare the workflow

Define a concierge workflow 'customer_onboarding' with steps [collect, validate, create, notify], each with its own tool set.✓ Скопировано

→ Workflow config
Share state

Pass the customer_data dict from step 1 into steps 2, 3, 4 via shared state. Show me how.✓ Скопировано

→ State-object code
Handle failure

If validate fails, return to step 1 with the specific fix needed.✓ Скопировано

→ Retry/backtrack logic

Итог: A robust guided flow that keeps the LLM on-rails.

Подводные камни

LLM tries to skip steps — Concierge enforces order at the tool-visibility level — skipping is physically impossible if you wired it right

Ship your first MCP server in 15 minutes

👤 Developers new to MCP ⏱ ~15 min beginner

Когда использовать: You want to expose 3-5 functions to Claude and don't want to learn the raw MCP spec.

Поток

Install and scaffold

pip install concierge-sdk. Generate a minimal server exposing two tools: add(a, b) and greet(name). Stdio transport.✓ Скопировано

→ Working Python file
Run and connect

Add to Claude Desktop config and test that both tools are callable.✓ Скопировано

→ Tool calls succeed in Claude

Итог: A working MCP server you wrote end-to-end in an afternoon.

Подводные камни

Missing type hints — SDK relies on them for tool schema — Always type your args and return — concierge generates the MCP schema from them

Комбинации

Сочетайте с другими MCP — эффект x10

concierge + gateway

Serve concierge-built MCP behind a security gateway for PII redaction

Start concierge server on :8001, then add it as an upstream to mcp-gateway with Presidio plugin.✓ Скопировано

Инструменты

Что предоставляет этот MCP

Инструмент	Входные данные	Когда вызывать	Стоимость
concierge.workflow(name)	name, steps, initial_state	Define a named multi-step flow	framework only
concierge.tool(workflow, step)	step allowlist	Attach a function as a tool scoped to one or more steps	framework only
concierge.search_tools	query: str	Auto-exposed when you have too many tools to list eagerly	free
concierge.serve()	transport: 'stdio'\|'http'\|'sse'	Entry point of your script	free

Стоимость и лимиты

Во что обходится

Квота API: None — you're the author
Токенов на вызов: Depends on your tool design; progressive disclosure shrinks system prompts 5-10x
Деньги: Free and open source
Совет: Don't eagerly expose every tool. Start with 'search' + current-step tools. Add eager tools only if profiling shows the LLM always needs them.

Безопасность

Права, секреты, радиус поражения

Хранение учётных данных: Your MCP server handles its own credentials — concierge doesn't impose a model

Исходящий трафик: Whatever your tools call

As an SDK, concierge's threat model is inherited from how you use it. Validate inputs in each tool.
Session isolation protects state per-user but shared external resources still need their own locks.

Устранение неполадок

Частые ошибки и исправления

Tools not appearing in the expected step

Check the step arg on @concierge.tool. A tool with step=['create'] is invisible during collect.

Проверить: Call the search_tools tool; if nothing returns, your allowlist is wrong

Semantic search returns no results

Tool descriptions may be empty. Concierge builds embeddings from docstrings — fill them in.

Workflow state lost between calls

Concierge state is session-scoped. Ensure your client is sticky (same MCP session across calls).

Альтернативы

concierge в сравнении

Альтернатива	Когда использовать	Компромисс
fastmcp (Python)	You want the most popular Python MCP SDK without progressive disclosure	Lighter, but no built-in workflow/tool-gating features
Official MCP Python SDK	You want zero framework, closest to the spec	More boilerplate; you build gating yourself
TypeScript MCP SDK	Your stack is TS/Node	Different language; no direct concierge equivalent

Ещё

Ресурсы

📖 Читать официальный README на GitHub

🐙 Открытые задачи

🔍 Все 400+ MCP-серверов и Skills