Agenten stärken
Der eigene Befehl des Herausgebers. Kein PlugMCP-Endpunkt.
Host a copy with our keys and logs.
npx -y @danielsogl/[email protected]Claude Code hat keinen Installationslink für einen lokalen Befehl. Kopieren Sie den Befehl unten und führen Sie ihn in einem Terminal aus.
claude mcp add danielsogl-lighthouse-mcp-server -- npx -y @danielsogl/lighthouse-mcp@2.0.1In einem Terminal ausführen.
{ "mcpServers": { "danielsogl-lighthouse-mcp-server": { "command": "npx", "args": [ "-y", "@danielsogl/[email protected]" ] } }}claude_desktop_config.json
Remote-URL oder ein Befehl auf Ihrem Rechner? Worin sich die beiden Installationen unterscheiden
Auf PlugMCP ausführen
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
StartenA Model Context Protocol (MCP) server that provides comprehensive web performance auditing and analysis capabilities using Google Lighthouse. This server enables LLMs and AI agents to perform detailed website performance assessments, accessibility audits, SEO analysis, security checks, and Core Web Vitals monitoring.
outputSchema and returns validated structuredContent, so clients get typed data instead of a JSON string to parseInstall the Lighthouse MCP server with your preferred client using one of the configurations below:
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": ["@danielsogl/lighthouse-mcp@latest"]
}
}
}
If you need authenticated sessions, launch with a persistent Chrome profile and run headed:
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": [
"@danielsogl/lighthouse-mcp@latest",
"--profile-path",
"<profile-path>",
"--no-headless"
]
}
}
}
You can pass extra Chrome flags with --chrome-flag, for example --chrome-flag=--disable-gpu.
If the flag value starts with -- and matches a known option name, prefer --chrome-flag=... to avoid parsing it as a top-level option.
Profile mode disables Lighthouse's storage reset so cookies and local storage persist between runs.
If --user-data-dir points to a missing directory, it will be created and treated as a fresh profile.
Set --profile-path to the Profile Path shown in chrome://version (e.g. .../Default).
Note: Chrome's remote debugging requires a non-default user data directory, so reuse a dedicated profile directory instead of the system default.
You can also pass --user-data-dir + --profile-directory separately if you prefer.
Attaching with --chrome-port alone does not preserve storage; include a profile flag to keep sessions.
Supported runtime flags for the MCP server:
--profile-path <path>: use the Profile Path from chrome://version (auto-derives user data dir + profile name)--user-data-dir <path>: reuse a Chrome profile directory for persistent sessions--profile-directory <name>: select a profile within the user data dir--chrome-path <path>: explicit path to the Chrome/Chromium executable (overrides auto-detection; also respects the CHROME_PATH environment variable)--chrome-flag <flag> or --chrome-flag=<flag>: pass through extra Chrome flags (repeatable)--chrome-port <port> or --remote-debugging-port <port>: attach to an existing Chrome instance launched with remote debugging enabled--headless: force headless mode--no-headless: force headed modeLighthouse logs to stderr. The server keeps this at error so it does not flood your MCP
client's logs; set LIGHTHOUSE_LOG_LEVEL to silent, info or verbose when debugging
(for example when Chrome fails to launch).
LIGHTHOUSE_LOG_LEVEL=verbose npx @danielsogl/lighthouse-mcp@latest
If the wrong Chrome binary is picked up (e.g. Windows Chrome instead of the Linux binary on WSL2), set the path explicitly:
# Via CLI flag
npx @danielsogl/lighthouse-mcp@latest --chrome-path /usr/bin/google-chrome
# Via environment variable
CHROME_PATH=/usr/bin/google-chrome npx @danielsogl/lighthouse-mcp@latest
In your MCP config:
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": ["@danielsogl/lighthouse-mcp@latest", "--chrome-path", "/usr/bin/google-chrome"]
}
}
}
Run a real audit with a persistent profile (use an existing profile directory and log in once if needed):
npm run smoke:profile -- --url https://example.com \
--profile-path "<profile-path>" \
--no-headless
Start Chrome with remote debugging enabled:
/path/to/GoogleChromeExecutable \
--remote-debugging-port=9222 \
--user-data-dir /path/to/chrome-profile
Replace /path/to/GoogleChromeExecutable with your platform's Chrome/Chromium binary path.
Then attach Lighthouse to that instance:
npm run smoke:profile -- --url https://example.com --chrome-port 9222
To preserve storage when attaching, pass the profile path so Lighthouse keeps cookies/local storage:
npm run smoke:profile -- --url https://example.com \
--chrome-port 9222 \
--profile-path "<profile-path>"
You can also install the Lighthouse MCP server using the VS Code CLI:
# For VS Code
code --add-mcp '{"name":"lighthouse","command":"npx","args":["-y","@danielsogl/lighthouse-mcp@latest"]}'
# For VS Code Insiders
code-insiders --add-mcp '{"name":"lighthouse","command":"npx","args":["-y","@danielsogl/lighthouse-mcp@latest"]}'
After installation, the Lighthouse MCP server will be available for use with your GitHub Copilot agent in VS Code.
Go to Cursor Settings → MCP → Add new MCP Server. Name it "lighthouse", use command type with the command npx @danielsogl/lighthouse-mcp@latest:
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": ["@danielsogl/lighthouse-mcp@latest"]
}
}
}
Follow the Windsurf MCP documentation. Use the following configuration:
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": ["@danielsogl/lighthouse-mcp@latest"]
}
}
}
Follow the MCP install guide, use the following configuration:
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": ["@danielsogl/lighthouse-mcp@latest"]
}
}
}
The Lighthouse MCP server provides the following tools for comprehensive web analysis:
The Lighthouse MCP server includes reusable prompts that help LLMs provide structured analysis and recommendations:
The Lighthouse MCP server provides built-in reference resources with essential guidelines and best practices:
auditResults: JSON audit results from Lighthouse toolsfocusArea: Specific category to focus on ("performance", "accessibility", "seo", "best-practices", "agentic-browsing")beforeAudit / afterAudit: Lighthouse audit results before and after changeschangesImplemented: Description of changes made between auditscurrentMetrics: Current performance metrics from auditstargetGoals: Specific performance targets or business goalstimeframe: Timeline for implementing improvementsframework: Frontend framework or technology stackconstraints: Technical or business constraintswebsiteType: Type of website (e.g., e-commerce, blog, corporate)targetAudience: Target audience or market informationcomplianceLevel: WCAG compliance level ("AA" or "AAA")userGroups: Specific user groups to consider for accessibilityurl (required): The website URL to analyzedevice: Target device ("desktop" or "mobile", default: "desktop")includeDetails: Include detailed audit information (default: false)throttling: Enable network/CPU throttling (default: false)categories: Lighthouse categories to audit (["performance", "accessibility", "best-practices", "seo", "agentic-browsing"])threshold: Custom thresholds for metrics (e.g., {"lcp": 2.5, "inp": 200, "cls": 0.1})budget: Performance budget limits (e.g., {"performanceScore": 90, "largestContentfulPaint": 2500})resourceTypes: Resource types to analyze (["images", "javascript", "css", "fonts", "other"])minBytes: Minimum file size threshold for analysis (default: 2048)checks: Security checks to perform (["https", "csp", "hsts", "origin-isolation", "clickjacking", "trusted-types", "third-party-cookies", "deprecations"])// Get overall performance score
{
"tool": "get_performance_score",
"arguments": {
"url": "https://example.com",
"device": "mobile"
}
}
// Check Core Web Vitals with custom thresholds
{
"tool": "get_core_web_vitals",
"arguments": {
"url": "https://example.com",
"device": "mobile",
"includeDetails": true,
"threshold": {
"lcp": 2.5,
"inp": 200,
"cls": 0.1
}
}
}
// Comprehensive security audit
{
"tool": "get_security_audit",
"arguments": {
"url": "https://example.com",
"checks": ["https", "csp", "hsts"]
}
}
// Find optimization opportunities
{
"tool": "analyze_resources",
"arguments": {
"url": "https://example.com",
"resourceTypes": ["images", "javascript"],
"minSize": 1024
}
}
Access built-in guidelines and best practices:
// Get Core Web Vitals thresholds
{
"resource": {
"uri": "lighthouse://performance/core-web-vitals-thresholds"
}
}
// Access WCAG accessibility guidelines
{
"resource": {
"uri": "lighthouse://accessibility/wcag-guidelines"
}
}
// Get framework-specific optimization guides
{
"resource": {
"uri": "lighthouse://frameworks/optimization-guides"
}
}
// Analyze audit results with focused recommendations
{
"prompt": "analyze-audit-results",
"arguments": {
"auditResults": "{...lighthouse audit json...}",
"focusArea": "performance"
}
}
// Create a performance improvement plan
{
"prompt": "create-performance-plan",
"arguments": {
"currentMetrics": "{...current performance metrics...}",
"targetGoals": "Achieve 90+ performance score and sub-2s LCP",
"timeframe": "3 months"
}
}
// Compare before/after audit results
{
"prompt": "compare-audits",
"arguments": {
"beforeAudit": "{...before audit results...}",
"afterAudit": "{...after audit results...}",
"changesImplemented": "Implemented lazy loading and image optimization"
}
}
The server is built using:
npm run test:run # unit tests
npm run test:coverage # unit tests with coverage
npm run test:e2e # end-to-end tests
The end-to-end suite builds the server, launches it over stdio with a real MCP client, and
runs actual Lighthouse audits against a fixture page served on loopback. It requires Chrome
to be installed; set CHROME_PATH if it lives somewhere non-standard.
Contributions are welcome! Please read our Contributing Guide for details on:
This project is licensed under the MIT License - see the LICENSE file for details.
For security issues, please see our Security Policy.
Built with ❤️ by Daniel Sogl
MCP server that enables AI agents to perform comprehensive web audits using Google Lighthouse with 13+ tools for performance, accessibility, SEO, and security analysis.
{ "mcpServers": { "danielsogl-lighthouse-mcp-server": { "command": "npx", "args": [ "-y", "@danielsogl/[email protected]" ] } }}Tools
Noch keine Tool-Liste. Remote-Server erhalten eine nach dem ersten Health-Check.
FAQ
Kopieren Sie auf dieser Seite die stdio-Konfiguration für Cursor, Claude Desktop, Claude Code oder VS Code. Sie nutzt die Installation des Herausgebers, keinen PlugMCP-Proxy.
In diesem Eintrag ist keine erforderliche Umgebungsvariable aufgeführt.
Lighthouse läuft lokal über stdio.
Zuletzt geprüft: 2026-11-17. MCP server that enables AI agents to perform comprehensive web audits using Google Lighthouse with 13+ tools for performance, accessibility, SEO, and security analysis.