CoreDash MCP Server

CoreDash MCP Server

Real User Monitoring for Core Web Vitals. Query LCP, INP, and CLS field data from real visitors.

Category
访问服务器

README

CoreDash MCP Server

Real User Monitoring for Core Web Vitals. Query LCP, INP, and CLS field data from real visitors through the Model Context Protocol.

The MCP server is hosted at https://app.coredash.app/api/mcp. This repo holds the public connector docs and registry manifest. You do not need to install or run a server. You connect your MCP client directly to the hosted endpoint.

What CoreDash is

CoreDash is a Real User Monitoring platform built around Core Web Vitals. It collects LCP, INP, and CLS from real visitors and lets you segment by page template, country, device, browser, OS, and 20 plus other dimensions. Data is stored in the EU. The product runs without consent banners. Paid plans start at $19 per month. More at coredash.app.

Tools exposed via MCP

Tool Purpose
get_metrics Current performance scores at p75. Filter and group by any dimension. Use for "what is" questions.
get_timeseries Metric over time with an improving, stable, or regressing summary. Use for regressions and trends.
get_histogram Distribution shape of a single metric across roughly 40 buckets. Use to see whether traffic is bimodal, long tailed, or evenly slow.

All three tools return rated buckets (good, improve, poor) matching the Core Web Vitals thresholds. Filter dimensions include device (d), country (cc), page path (ff), URL (u), browser, OS, and the LCP / INP / CLS attribution element selectors.

Authentication

The server supports two methods:

  1. OAuth (recommended). Point your MCP client at the server URL and complete the browser sign-in when prompted — no API key required. The server follows the WWW-Authenticate discovery flow and publishes /.well-known/oauth-authorization-server and /.well-known/oauth-protected-resource.
  2. API key. Best for CI or headless use. Generate one at coredash.app/settings/api and pass it as Authorization: Bearer cdk_YOUR_API_KEY.

Anonymous calls succeed for initialize (so clients can introspect server capabilities) and fail for any tool call.

Connecting

Claude Code

Add the server, then authenticate in the browser:

claude mcp add --transport http coredash https://app.coredash.app/api/mcp

Run /mcp inside Claude Code, select coredash, and complete the OAuth sign-in.

To use an API key instead (e.g. in CI), pass it as a header:

claude mcp add --transport http coredash https://app.coredash.app/api/mcp \
  --header "Authorization: Bearer cdk_YOUR_API_KEY"

Cursor

Add to ~/.cursor/mcp.json, then click Login on the server in Cursor's MCP settings to complete OAuth:

{
  "mcpServers": {
    "coredash": {
      "url": "https://app.coredash.app/api/mcp"
    }
  }
}

To use an API key instead, add a "headers" block with "Authorization": "Bearer cdk_YOUR_API_KEY".

VS Code

Add to .vscode/mcp.json in your workspace (or the user config via the MCP: Open User Configuration command). VS Code prompts you to sign in on first use:

{
  "servers": {
    "coredash": {
      "type": "http",
      "url": "https://app.coredash.app/api/mcp"
    }
  }
}

To use an API key instead, add a "headers" block with "Authorization": "Bearer cdk_YOUR_API_KEY".

Windsurf

Same JSON shape as Cursor, in the Windsurf MCP settings.

Any other MCP client

Point it at https://app.coredash.app/api/mcp with the streamable-http transport. Clients that support OAuth discover the flow via the WWW-Authenticate header; otherwise pass an Authorization: Bearer cdk_YOUR_API_KEY header.

Example queries

Why is LCP slow on mobile in Germany this week?
Has INP regressed on the product detail template in the last 24 hours?
Show me the LCP distribution on /checkout/* in the US.

Registry listing

This server is published on the Official MCP Registry as io.github.corewebvitals/coredash. From there it propagates to PulseMCP, Smithery, Glama, and Windsurf.

The registry manifest is in server.json. To publish a new version, edit the file and run mcp-publisher publish.

Related projects

  • cwv-superpowers. Skills package for Claude Code, Cursor, and Gemini that drives this MCP server to diagnose and fix Core Web Vitals issues.
  • Core Web Vitals Visualizer. Chrome extension. Lets you see CWV metrics overlayed on any page.

License

MIT. See LICENSE.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选