pagespeed-insights-mcp
This MCP server integrates Google PageSpeed Insights to analyze website performance, accessibility, best practices, SEO, and PWA on mobile and desktop, returning detailed audits and optimization opportunities.
README
PageSpeed Insights MCP Server
FastMCP server for Google PageSpeed Insights analysis through the PageSpeed Insights REST API.
Features
- Run PageSpeed Insights analysis for desktop or mobile.
- Return compact summaries with Lighthouse category scores, core metrics, field data, opportunities, warnings, and failed audits.
- Compare mobile and desktop results.
- List performance opportunities sorted by estimated savings.
- Fetch a specific Lighthouse audit by id, including its full details payload.
- Expose local API metadata and supported categories/strategies.
Google API endpoint
This server calls:
GET https://pagespeedonline.googleapis.com/pagespeedonline/v5/runPagespeed
The required query parameter is url. Optional parameters used by this server include strategy, repeated category, locale, captchaToken, and key.
API docs:
- https://developers.google.com/speed/docs/insights/rest
- https://developers.google.com/speed/docs/insights/rest/v5/pagespeedapi/runpagespeed
Create .env file
An API key is optional for trial calls, but recommended for repeated automated usage.
# Optional
PAGESPEED_API_KEY=google_api_key_here
# Optional alternative key name
GOOGLE_API_KEY=google_api_key_here
# Optional
PAGESPEED_API_URL=https://pagespeedonline.googleapis.com
PAGESPEED_TIMEOUT_SECONDS=120
Tools
get_pagespeed_api_metadata
Returns endpoint details, whether an API key is configured, supported strategies, supported categories, aliases, and documentation links.
run_pagespeed
Runs PageSpeed analysis for a URL.
Parameters:
url: URL to analyze. Must start withhttp://,https://,url:http(s)://, ororigin:http(s)://.strategy:DESKTOPorMOBILE. Defaults toDESKTOP.categories: optional list. Supported values:PERFORMANCE,ACCESSIBILITY,BEST_PRACTICES,SEO,PWA, orall.locale: optional locale such asen-US.utm_campaign: optional campaign name for analytics.utm_source: optional campaign source for analytics.captcha_token: optional captcha token.full_response: return the full Google API JSON response when true.api_key: optional per-call API key override.
analyze_url
Runs analysis and returns a compact report. By default it runs performance, accessibility, best practices, and SEO.
compare_mobile_desktop
Runs mobile and desktop analysis for the same URL and returns score deltas.
list_opportunities
Runs performance analysis and returns opportunity audits sorted by estimated savings.
get_lighthouse_audit
Returns one Lighthouse audit by audit id. Use categories=["all"] if the audit may not be part of the Performance category.
Create a virtual environment
cd PROJECT_FOLDER
python3 -m venv .venv
Activate it
source .venv/bin/activate
Install dependencies
pip install fastmcp requests python-dotenv
Freeze the requirements
pip freeze > requirements.txt
Test it
python3 server.py
You should see:
Starting MCP server 'pagespeed-insights-mcp-server' with transport 'stdio'
Configure OpenCode
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"pagespeed-insights": {
"type": "local",
"command": [
"PATH_TO_YOUR_MCP_PROJECT/.venv/bin/python",
"PATH_TO_YOUR_MCP_PROJECT/server.py"
],
"enabled": true
}
}
}
Docker / HTTP mode
docker compose up --build
HTTP mode listens on port 8004.
Docker Compose reads values from .env automatically when the file exists, but the container can run without one for unauthenticated trial calls.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。