concierge MCP — Anwendungsfälle, Installieren & Live-Demo

Warum nutzen

Hauptfunktionen

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

Live-Demo

In der Praxis

concierge.replay ▶ bereit

0/0

Installieren

Wählen Sie Ihren Client

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

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

Öffne Claude Desktop → Settings → Developer → Edit Config. Nach dem Speichern neu starten.

~/.cursor/mcp.json · .cursor/mcp.json

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

Cursor nutzt das gleiche mcpServers-Schema wie Claude Desktop. Projektkonfiguration schlägt die globale.

VS Code → Cline → MCP Servers → Edit

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

Klicken Sie auf das MCP-Servers-Symbol in der Cline-Seitenleiste, dann "Edit Configuration".

~/.codeium/windsurf/mcp_config.json

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

Gleiche Struktur wie Claude Desktop. Windsurf neu starten zum Übernehmen.

~/.continue/config.json

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

Continue nutzt ein Array von Serverobjekten statt einer Map.

~/.config/zed/settings.json

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

In context_servers hinzufügen. Zed lädt beim Speichern neu.

claude mcp add concierge -- uvx concierge

Einzeiler. Prüfen mit claude mcp list. Entfernen mit claude mcp remove.

Anwendungsfälle

Praxisnahe Nutzung: 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

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

Voraussetzungen

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

Ablauf

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.✓ Kopiert

→ 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.✓ Kopiert

→ 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.✓ Kopiert

→ Claude sees ~10 tools, not 300

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

Fallstricke

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

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

Ablauf

Declare the workflow

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

→ Workflow config
Share state

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

→ State-object code
Handle failure

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

→ Retry/backtrack logic

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

Fallstricke

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

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

Ablauf

Install and scaffold

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

→ Working Python file
Run and connect

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

→ Tool calls succeed in Claude

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

Fallstricke

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

Kombinationen

Mit anderen MCPs für 10-fache Wirkung

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.✓ Kopiert

Werkzeuge

Was dieses MCP bereitstellt

Werkzeug	Eingaben	Wann aufrufen	Kosten
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

Kosten & Limits

Was der Betrieb kostet

API-Kontingent: None — you're the author
Tokens pro Aufruf: Depends on your tool design; progressive disclosure shrinks system prompts 5-10x
Kosten in €: Free and open source
Tipp: Don't eagerly expose every tool. Start with 'search' + current-step tools. Add eager tools only if profiling shows the LLM always needs them.

Sicherheit

Rechte, Secrets, Reichweite

Credential-Speicherung: Your MCP server handles its own credentials — concierge doesn't impose a model

Datenabfluss: 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.

Fehlerbehebung

Häufige Fehler und Lösungen

Tools not appearing in the expected step

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

Prüfen: 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).

Alternativen

concierge vs. andere

Alternative	Wann stattdessen	Kompromiss
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

Mehr

Ressourcen

📖 Offizielle README auf GitHub lesen

🐙 Offene Issues ansehen

🔍 Alle 400+ MCP-Server und Skills durchsuchen