MCP Queen

com.mcpqueen由 mcpqueen.com 發布官方發布者v0.3.0更新於 Nov 2, 2026

The evidence layer for MCP: live operational grades plus Trust Receipts for every registry server.

已驗證Streamable HTTP可網頁執行網頁搜尋與抓取開發者工具安全與監控

概覽

AI 產生的概覽

MCP Queen 是公開的 MCP 伺服器探索與證據服務，讓助理在連線前搜尋伺服器及其工具，並查看運作評級與 Trust Receipt。

可以做什麼
  • 以自然語言任務搜尋 MCP 伺服器，可依驗證、延遲、類別與運作評級篩選
  • 搜尋已評級伺服器實際提供的工具，依能力或資料類型查找
  • 列出評級最高的伺服器，並取得單一伺服器的評級與逐字探測證據
  • 讀取某個伺服器的 Trust Receipt，並檢索已發布的信任證據與經審閱的實測回報
  • 提交你實際使用過的伺服器實測回饋，進入隔離的人工審閱佇列
適用情境
適合在挑選要連線的 MCP 伺服器時使用，希望取得有日期、有證據支撐的工具、可靠性、驗證與存取限制資訊，而非宣傳說法。也適合在安裝前確認某個伺服器是否提供特定能力。
執行需求
遠端 Streamable HTTP 端點 mcpqueen.com/mcp，限速評估不需帳號。對於無法直連遠端 HTTP 端點的用戶端，提供可選的零依賴本機 stdio 橋接（Node.js 或 Docker），並可用 MCPQUEEN_MCP_URL 指向其他端點。
安裝前請注意
運作評級並非安全、隱私、資料品質或合規認證，缺少證據代表未經稽核而非安全。submit_feedback 會寫入審閱佇列，請勿提交敏感細節；範例代理刻意不開放此工具給模型。查詢與回饋會傳送給第三方託管服務。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 MCP Queen，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/com.mcpqueen%2Fregistry/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "queen": {
      "type": "http",
      "url": "https://mcpqueen.com/mcp"
    }
  }
}

README

MCP Queen

MCP Queen is a public discovery and evidence service for Model Context Protocol servers. It publishes dated operational observations, tool metadata, and separate Trust Receipts so developers and agents can inspect evidence before deciding what to connect.

An MCP Queen operational grade is not a security, privacy, data-quality, or compliance certification. Missing evidence is unaudited, not safe.

Use the hosted endpoint

The public Streamable HTTP endpoint requires no account for rate-limited evaluation:

text
https://mcpqueen.com/mcp

It exposes six read-only discovery/evidence tools and one additive feedback tool:

  • search_servers
  • search_tools
  • list_grades
  • get_server_grade
  • get_trust_receipt
  • search_trust_evidence
  • submit_feedback — writes only to a quarantined review queue

For human-readable setup and current service limits, use mcpqueen.com/connect.

Claude Code

bash
claude mcp add --transport http mcpqueen https://mcpqueen.com/mcp

Direct protocol examples

The model-free examples under examples/clients call the hosted endpoint directly:

bash
node examples/clients/node-http.mjs "GitHub issue triage"

The Python version uses the official MCP SDK. Additional examples show bounded integration with LangChain, LlamaIndex, Cloudflare Agents, Hugging Face, and the OpenAI Responses API. Agent examples allowlist only the six read-only tools; submit_feedback is excluded from automatic model access.

Optional local stdio bridge

This repository contains a small zero-dependency stdio bridge for clients that cannot connect to a remote Streamable HTTP endpoint directly. The bridge does not contain MCP Queen's hosted implementation or evidence corpus; it forwards JSON-RPC messages to https://mcpqueen.com/mcp.

bash
git clone https://github.com/mcpqueen/mcpqueen.gitcd mcpqueennpm cinpm start

To point the bridge at another compatible endpoint during local testing, set MCPQUEEN_MCP_URL in the process environment.

Docker is also supported:

bash
docker build -t mcpqueen-bridge .docker run --rm -i mcpqueen-bridge

What this public repository contains

The production Worker, deployment configuration, evidence database, submission packages, operational records, and private source history are not part of this public developer surface. This repository is not a production deployment source.

Verify locally

bash
npm cinpm test

The tests exercise the bridge against a local mock endpoint and verify that public documentation and package scripts reference files that are actually in this repository.

License and support

Code in this repository is available under the MIT License. Product documentation, privacy, terms, and support are available from mcpqueen.com.

來源：README.md，提交 51168b7

工具

7

search_servers

Search MCP Servers

Search the MCP evidence registry using a natural multi-word task (e.g. 'reliable no-auth drug interaction server with citations'). Tokenizes and expands common synonyms, ranks metadata plus observed tool descriptions, and supports operational-grade/auth/latency/category filters. Use get_trust_receipt or search_trust_evidence for security, data, citation, claim, and response evidence.

輸入參數authlimitquery必填categoryminimum_grademax_latency_ms

search_tools

Search MCP Tools

Search across the actual tools that graded MCP servers expose (their tool names and descriptions, captured live from tools/list) — not just server metadata. Use this when you need a specific capability or data type, e.g. 'get weather', 'query postgres', 'device recall', 'FDA 510k'. Returns the matching tools with the server that offers each, its grade, and the remote endpoint so you can connect directly.

輸入參數authlimitquery必填categoryminimum_grademax_latency_ms

list_grades

List Operational Grades

List the top graded MCP servers from the mcpqueen registry (deterministic probe grades with evidence). Returns grade, score 0-100, latency, tool count and auth state per server.

輸入參數limit

get_server_grade

Get Server Grade

Get the full grade and verbatim probe evidence for one MCP server, by its official registry name (e.g. 'com.healthai/radar').

輸入參數name必填

get_trust_receipt

Get Trust Receipt

Get one MCP server's complete evidence receipt: operational grade, deterministic security/data-integrity/citation/claim observations, and reviewed real-usage field reports. Missing observations are explicitly unaudited, never treated as a pass.

輸入參數name必填

search_trust_evidence

Search Trust Evidence

Search published MCP trust evidence and reviewed real-usage field reports. Use for questions such as which servers expose citations, have access caveats, make unverifiable corpus claims, or show security concerns. Returns verbatim observations with dates and source type, not a synthetic trust score.

輸入參數limitquery必填statusdimension

submit_feedback

Submit Field Report

Submit a field report about an MCP server you have actually used (what worked, what failed, surprising behavior). Reports are quarantined for human review and never auto-published.

輸入參數report必填agent_nameserver_name必填

版本歷史

1
  1. v0.3.0最新Sep 16, 2026

評分

已滿足適用檢查項的 100%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分