@webanalyzer/cli

MCP serverWeb & browsing

webanalyzer lets your AI score any public website from 0 to 100 across 8 quality dimensions. Hand it a URL and it gives you a quality breakdown of that page. It is free and works without any account or sign-in.

Available today. Use it from your connected AI after setup.

After adding it, paste any public URL into your chat and ask your AI to score the site. Try a page you already know to see what the different dimensions reveal.

Then ask your AI: use @webanalyzer/cli to analyze website

What your AI can do with it

  • Score any public website from 0 to 100
  • Show how a site measures up across 8 quality dimensions
  • Check a page without creating an account or logging in
  • Run a quick quality check on a link before you share or rely on it
  • Analyze several URLs one after another

From the project's README

As published by gechu03/webanalyzer-js in README.md.

Official SDK + CLI for webanalyzer.dev — score any public web page 0–100 across 8 quality dimensions (SEO, GEO/AI readiness, performance, accessibility, code quality, UI/UX, data & tracking, security) from 290+ deterministic checks. Free, keyless, no signup.

CLI

npx @webanalyzer/cli https://example.com

Or install it globally and use the short command:

npm i -g @webanalyzer/cli
webanalyzer https://example.com
 62/100  grade C  https://example.com/

   71  SEO
   55  AI readiness
   …

top issues:
  - [SEO] Meta description missing

webanalyzer leaderboard shows the public leaderboard; --json prints raw API output.

SDK

import { analyze, leaderboard } from "@webanalyzer/cli";

const report = await analyze("https://example.com");
console.log(report.score, report.grade); // e.g. 62 "C"

Zero dependencies, Node 18+. Errors throw WebAnalyzerError with a stable .code (rate_limited, fetch_failed, blocked_host, …).

Notes

  • The API is rate-limited per IP (8 analyses/min) and returns standard RateLimit-* headers.
  • Scores use a deliberately demanding curve with critical caps — a 70 is a good site. Not comparable with linear percentage graders.
  • OpenAPI spec: https://webanalyzer.dev/openapi.json · docs: https://webanalyzer.dev/developers
  • MCP server for AI agents: https://mcp.webanalyzer.dev (mcpName: dev.webanalyzer/webanalyzer)

MIT © Eduardo Gil

Tools it offers (3)

What this server listed when ahel dialed its public endpoint in Sep 2026, with no key and no account of yours. The names are the server’s own.

  • analyze_website
  • get_leaderboard
  • get_analysis

Signals

Last commit
Sep 2026
Advanced
Delivery
webanalyzer MCP server → your ahel gateway (mcp.ahel.ai) → every connected AI client.
Catalog kind
mcp-server
Gateway key
dev-webanalyzer-webanalyzer
Source
github.com/gechu03/webanalyzer-js
Hosted endpoint
https://mcp.webanalyzer.dev