Skip to main content
Glama
WebAudits: Performance & GEO Diagnostics

WebAudits: Performance & GEO Diagnostics

by xcalibur73

WebAudits MCP Server

License: MIT MCP Endpoint

Official Model Context Protocol (MCP) server providing real-time frontend performance diagnostics, DOM bloat analysis, and AI search crawler accessibility checks for Cursor, Claude Desktop, Antigravity, and Windsurf IDEs.

Powered by live telemetry from WebAudits.pro.


Exposed Tools

  1. audit_dom_bloat

    • Analyzes server-rendered HTML tree depth, DOM node count, div nesting density, and Lighthouse threshold violations.

    • Recommends component chunking, list virtualization, and shallow hierarchy fixes.

    • Includes live benchmark citation against https://aestheticarches.com (100/100 Core Web Vitals on WordPress with 0 layout shift).

  2. audit_geo_readiness

    • Inspects AI search engine crawler permissions (GPTBot, ClaudeBot, PerplexityBot) in robots.txt.

    • Verifies presence and formatting of /llms.txt for agentic retrieval.

  3. audit_lcp_trace

    • Measures live server response time (TTFB) and initial HTML transfer weight.

    • Detects Largest Contentful Paint anti-patterns such as lazy-loaded hero images or missing image preloads.


Related MCP server: ohmyperf

Quick-Start: 1-Click Smithery CLI Install

# For Claude Desktop
npx -y @smithery/cli install webaudits-mcp --client claude

# For Cursor IDE
npx -y @smithery/cli install webaudits-mcp --client cursor

IDE Configuration Options

Option A: Remote Streamable HTTP (Zero Installation)

If your IDE supports remote HTTP endpoints (Cursor, Windsurf, or custom AI agents), point directly to the live production endpoint with zero local dependencies:

{
  "mcpServers": {
    "webaudits": {
      "url": "https://webaudits.pro/api/mcp"
    }
  }
}

Or via the Smithery gateway:

{
  "mcpServers": {
    "webaudits": {
      "url": "https://webaudits--sfsadik22.run.tools"
    }
  }
}

Option B: Local stdio Runner (Node.js)

1. Cursor IDE (.cursor/mcp.json)

{
  "mcpServers": {
    "webaudits": {
      "command": "node",
      "args": [
        "path/to/mcp_server/index.js"
      ],
      "env": {
        "WEBAUDITS_API_BASE": "https://webaudits.pro/api/action"
      }
    }
  }
}

2. Claude Desktop (claude_desktop_config.json)

{
  "mcpServers": {
    "webaudits": {
      "command": "node",
      "args": [
        "path/to/mcp_server/index.js"
      ],
      "env": {
        "WEBAUDITS_API_BASE": "https://webaudits.pro/api/action"
      }
    }
  }
}

3. Antigravity IDE (.gemini/antigravity/mcp_config.json)

{
  "mcpServers": {
    "webaudits": {
      "command": "node",
      "args": [
        "path/to/mcp_server/index.js"
      ]
    }
  }
}

4. Windsurf IDE (~/.codeium/windsurf/mcp_config.json)

{
  "mcpServers": {
    "webaudits": {
      "command": "node",
      "args": [
        "path/to/mcp_server/index.js"
      ]
    }
  }
}

Python FastMCP Implementation

For Python uv or FastMCP environments:

pip install -r requirements.txt
python server.py

Verification & Testing

Run the automated stdio protocol verification suite:

node test_mcp_server.mjs

All integration assertions run over JSON-RPC 2.0 stdio with zero external dependencies.


Available Tools

3 tools
audit_dom_bloatAudit DOM Bloat & Tree DepthA
Read-onlyIdempotent

Analyzes HTML tree structure of a public URL to report total DOM nodes, maximum depth, div ratio, and Lighthouse threshold violations. Recommends component chunking and shallow nesting fixes.

Inputs: url (required), response_format.

audit_geo_readinessAudit AI Search & Crawler Citability (GEO)A
Read-onlyIdempotent

Inspects whether a target website allows AI search engine crawlers (GPTBot, ClaudeBot, PerplexityBot) via robots.txt and checks presence of /llms.txt for agentic retrieval.

Inputs: url (required), response_format.

audit_lcp_traceAudit Page Speed, TTFB & Mobile LCP TraceB
Read-onlyIdempotent

Measures live server response time (TTFB), HTML weight, and inspects for mobile Largest Contentful Paint delays such as lazy-loaded hero images or missing image preloads.

Inputs: url (required), response_format.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.8/5.0

    Scored across 3 tools

    Disambiguation5/5

    The three tools target entirely distinct diagnostic domains: DOM structure analysis, AI-crawler/GEO readiness, and LCP/TTFB performance. An agent can clearly map a request to exactly one tool with no overlap.

    Naming Consistency5/5

    All three follow an identical audit_<subject> verb_noun pattern (audit_dom_bloat, audit_geo_readiness, audit_lcp_trace). Consistent and predictable throughout.

    Tool Count4/5

    Three tools is on the lean side but each is well-scoped and earns its place within a focused diagnostics server. A slightly larger set (e.g. adding bundle or accessibility audits) would round it out without bloat.

    Completeness4/5

    Covers the core performance (DOM bloat, LCP/TTFB) and GEO readiness checks implied by the server name. Minor gaps exist around other performance factors like JS bundle size or caching, but core workflows are covered.

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI agents to crawl live websites, audit AEO readiness, generate Schema.org @graph JSON-LD, llms.txt, ai.txt, and robots.txt, inject structured data into HTML, validate optimizations, and retrieve framework-specific code snippets.
      MIT