Provides detailed Lighthouse audits for web pages, including scores, failing elements, HTML snippets, and actionable suggestions. Supports desktop and mobile presets, category filtering (performance, accessibility, SEO, best-practices), and works with localhost and public URLs.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@MCP Lighthouse Audit audit http://localhost:4343/cottages"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP server that provides detailed Lighthouse audits for web pages. Unlike basic tools that only return scores, this server returns the specific elements that fail each audit with actionable suggestions.
Detailed Lighthouse audits with scores and specific failing elements
Works with localhost and public URLs
Desktop and mobile presets
Category filtering (performance, accessibility, seo, best-practices)
Clear output format with HTML snippets and selectors
Related MCP server: Marketing Master – Landing Page Evaluator
# Clone and build
cd ~/Dev/mcp-lighthouse-audit
pnpm install
pnpm buildclaude mcp add lighthouse-audit node ~/Dev/mcp-lighthouse-audit/dist/index.jsOr add manually to ~/.claude/settings.json:
{
"mcpServers": {
"lighthouse-audit": {
"command": "node",
"args": ["/Users/<username>/Dev/mcp-lighthouse-audit/dist/index.js"]
}
}
}Once installed, you can use the audit_page tool in Claude:
Audit http://localhost:4343/cottagesAudit http://localhost:4343 for accessibility onlyAudit https://example.com with mobile presetParameter | Type | Required | Description |
url | string | Yes | URL to audit (localhost or public) |
categories | string[] | No | Categories to audit: performance, accessibility, best-practices, seo |
preset | string | No | Device preset: desktop (default) or mobile |
Returns a formatted report with:
Scores for each category (0-100)
Summary of passed/failed audits
Detailed list of failed audits grouped by category
For each failed audit:
ID and title
Score
Description
Affected elements (HTML snippets, CSS selectors, URLs)
Node.js >= 18
Chrome or Chromium installed on the system
MIT
Run a detailed Lighthouse audit on a URL. Returns scores for all categories and a list of failed audits with specific elements that need fixing.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | The URL to audit (can be localhost or public URL) | |
| preset | No | Device preset for the audit. Defaults to desktop | |
| categories | No | Categories to audit. Defaults to all: performance, accessibility, best-practices, seo |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full transparency burden. It mentions returning scores and failed audits but does not disclose typical execution time, resource intensity, potential side effects (even if none), or limitations like whether localhost is supported. As a read-only audit, it should ideally state that it is non-destructive or note rate limits, but it does not.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, well-structured sentence that front-loads the verb and resource. It is highly concise, providing no redundant information, and every word adds value.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool has no output schema, so the description's mention of returned scores and failed audits is valuable. However, it does not explain the output format or potential error conditions, and it omits any note about preset/category defaults. Given the tool's moderate complexity, the description is adequate but leaves some questions about behavior beyond the basics.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema provides 100% coverage of all parameters (url, preset, categories) with descriptions, so the tool description does not need to repeat them. The description adds a slight contextual hint by saying 'all categories,' but it does not enhance understanding of parameter syntax, defaults, or interactions beyond the schema. Baseline score of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly specifies the tool's action: 'Run a detailed Lighthouse audit on a URL.' It also discloses the primary return value (scores for all categories, list of failed audits), giving the agent a precise understanding of what the tool does. No sibling tools exist, so differentiation isn't needed, but the description would stand out on its own.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for auditing URLs but does not provide explicit when-to-use guidance, prerequisites, or exclusions. Since there are no sibling tools, the absence of alternative references is acceptable, but it still lacks context like 'use this to evaluate page quality' or conditions such as URL accessibility. Minimal viable but clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
With only one tool, there is no possibility of confusion between tools. The purpose is clearly defined as running a Lighthouse audit.
The single tool name 'audit_page' follows a clear verb_noun pattern, which is consistent and descriptive.
A single tool for a Lighthouse audit server is functional but feels minimal. The tool is well-scoped, yet the server could benefit from additional related tools (e.g., audit_mobile, compare_audits).
The tool covers the core audit action comprehensively, returning all category scores and failed audits. However, there are no supplementary operations like retrieving historical audits or configuring audit options.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Scan a web page for accessibility, security, privacy, quality and SEO issues, with fixes.
Validate HTML/CSS, audit SEO and JSON-LD, check links, and capture responsive screenshots.
AI website audit: security, SEO, performance, UX and accessibility checks with actionable fixes.
Rule-based site audits: accessibility, SEO, security headers, performance. Metered per call.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/jladev/mcp-lighthouse-audit'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
