How to Run a Free SEO Page Audit Using Minifetch

The Minifetch /extract/url-metadata API endpoint returns every major on-page technical SEO signal in a single structured JSON response. This tutorial walks you through reading that response and turning it into a clear technical SEO checklist. Minifetch is free to use when you sign up and claim your free starter credits, and pay-as-you-go after that. That's pay-per-audit SEO, without a monthly subscription plan.

Check back later and we'll show you how to use the Minifetch API to turn this checklist into an analysis you can import to your AI Agents and automation workflows. To be notified, fill out our feedback form or join the Discord.

Building an AI agent? This tutorial is also available as a skill your AI agent can load directly: minifetch.com/skills/seo-page-audit/SKILL.md

  1. Get access

    Before you start, pick how you'll authenticate with Minifetch:

    API key — Sign up on the dashboard, grab free starter credits, and create a key. Recommended for most builders.

    USDC via x402 — Pay directly from a crypto wallet, no account needed.

  2. Check the URL first

    Free

    Before spending a credit, run a quick preflight check to confirm the URL is fetchable. This is free and tells you whether the site's robots.txt allows Minifetch to fetch it.

    From the homepage, enter your URL in the search box and click "Check URL".

    Or with the minifetch api client:

    await client.preflightCheck("https://yoursite.com/your-page");
    Tip: The checkAndExtractUrlMetadata() method in the minifetch-api client runs the preflight automatically before every paid fetch — so you only pay when a result is likely.

    Or from your cli:

    curl "https://minifetch.com/api/v1/free/preflight/url-check?url=https://yoursite.com/your-page"
    If you own the site and want to allow Minifetch access, see: How to unblock Minifetch from your site.

  3. Fetch the metadata

    Free to start, then $0.002 per URL

    From the homepage, just click "Go Fetch!". Metadata extraction is the default option.

    Or with the minifetch api client:

    const result = await client.checkAndExtractUrlMetadata(
  "https://yoursite.com/your-page",
  { verbosity: "full" }
);

    Or from your cli:

    curl "https://minifetch.com/api/v1/extract/url-metadata?url=https://yoursite.com/your-page&verbosity=full" \
  -H "Authorization: Bearer YOUR_API_KEY"

    Call the metadata endpoint with the verbosity=full option to get every signal, including favicons, response headers, hreflang tags, the full heading hierarchy, and image metadata.

    When to skip verbosity=full: If you only need the core SEO fields (title, description, canonical, og tags), leave it off. The default response is lighter and cheaper in an LLM context.

  4. Read the response as an SEO checklist

    The response maps directly to the signals search engines and social platforms care about. Here's what to check in each section:

    Core on-page signals

    FieldWhat to check
    titlePresent? Between 50–60 characters? Contains your primary keyword?
    descriptionPresent? Between 120–160 characters? Compelling and keyword-relevant?
    canonicalPresent? Points to itself? (A canonical pointing elsewhere signals a duplicate content issue.)
    langSet correctly for the page's target market?
    hreflangIf the site is multilingual: are all regions covered, with no missing reciprocal tags?
    viewportPresent? Google uses this as a mobile-friendliness signal.
    charsetSet? (Typically utf-8.)

    Structured data (rich results)

    FieldWhat to check
    jsonldPresent? Structured data unlocks rich results in Google Search (stars, prices, FAQs, breadcrumbs).
    jsonld[].@typeMatches the page's intent? Common types: Article, Product, BreadcrumbList, FAQPage.

    Social / Open Graph

    FieldWhat to check
    og:titlePresent? Can be the same as title or slightly different for social context.
    og:descriptionPresent?
    og:imagePresent? Recommended size is 1200×630px for best display across platforms.
    og:typeSet? (e.g. website, article, product)
    twitter:cardPresent? summary_large_image gets the best display on X/Twitter.
    twitter:titlePresent?
    twitter:imagePresent?

    Content structure

    FieldWhat to check
    headings.h1Exactly one H1? Contains the primary keyword? Multiple H1s are an SEO anti-pattern.
    headings.h2Present and logically structured?
    images[].altNon-empty on every image? Missing alt text is both an SEO flag and an accessibility issue.
    images[].width / heightSet on every image? Missing dimensions cause layout shift, which hurts Core Web Vitals scores.

    Technical signals (from response headers)

    FieldWhat to check
    responseStatusCodeShould be 200. Anything else is a problem to investigate.
    responseHeaders['x-robots-tag']Not set to noindex? A noindex header silently removes a page from search results.
    responseHeaders['cache-control']Set with a reasonable TTL? Helps crawl efficiency.

  5. Re-audit after changes

    Every response includes a minifetchCache object that tells you when the cached result expires: 

    "minifetchCache": {
  "hit": "false",
  "cachedAt": "2026-04-22T10:00:00.000Z",
  "expiresAt": "2026-04-22T10:02:00.000Z"
}

    If hit is true, you received a cached result and the page was not re-fetched. After deploying a fix, wait until after the expiresAt timestamp before re-auditing — fetching earlier returns the same cached data and charges a credit unnecessarily. The cache window is typically 2 minutes.