Web Page Metadata Extractor

Get the title, meta description, Open Graph tags, language, canonical URL and link and image counts of a list of web pages: one clean row per URL. Use it for SEO audits, link preview checks, content inventories or as a fast tool for AI agents that need to know what a page is about without reading all of it.

What it does

• Takes a list of page URLs.
• Loads each page with a plain HTTP request (no browser, no login, no cookies) and follows redirects.
• Returns one row per URL, in the same order as the input, with the input url and:
  • title, description and language declared by the page;
  • canonicalUrl, always as an absolute URL;
  • openGraph tags used for link previews in social networks and chat apps (title, description, image, type, siteName);
  • internalLinks and externalLinks: unique links to the same site (ignoring www.) and to other sites;
  • images: number of <img> tags;
  • finalUrl after redirects and the HTTP statusCode.
• A page that fails (HTTP error, timeout, not an HTML page) gets a row with an error message and the run continues with the other URLs.
• It does not collect personal data: no authors, emails or phone numbers.

Pages that build their content with JavaScript after loading may return fewer tags than a browser shows. Most sites put these tags in the initial HTML, because search engines and link previews need them there.

Input example

{
    "urls": [
        "https://www.python.org/",
        "https://docs.python.org/3/",
        "https://example.com/"
    ],
    "maxConcurrency": 5
}

Field	Required	Description
urls	Yes	Absolute http(s) URLs of the pages to analyze
maxConcurrency	No	Pages loaded in parallel, 1–20. Default 5

Output example

{
    "url": "https://docs.python.org/3/",
    "finalUrl": "https://docs.python.org/3/",
    "statusCode": 200,
    "title": "3.14.8 Documentation",
    "description": "The official Python documentation.",
    "language": "en",
    "canonicalUrl": "https://docs.python.org/3/index.html",
    "openGraph": {
        "title": "Python 3.14 documentation",
        "description": "The official Python documentation.",
        "image": "https://docs.python.org/3/_static/og-image.png",
        "type": "website",
        "siteName": "Python documentation"
    },
    "internalLinks": 45,
    "externalLinks": 13,
    "images": 3,
    "error": null
}

A failed URL looks like this:

{
    "url": "https://www.python.org/missing",
    "finalUrl": "https://www.python.org/missing",
    "statusCode": 404,
    "title": null,
    "description": null,
    "language": null,
    "canonicalUrl": null,
    "openGraph": null,
    "internalLinks": null,
    "externalLinks": null,
    "images": null,
    "error": "The page answered with HTTP 404"
}

Any field the page does not declare is null. Export the dataset as JSON, CSV, Excel or HTML, or read the Overview and Open Graph views in the Console.

Pricing

This Actor uses pay-per-event pricing. You pay only for pages analyzed successfully, and platform usage is included in the price. See the Pricing tab for the current price per event.

Event	When it is charged
page-analyzed	Once per URL that returns a row without error

URLs that fail are not charged. If you set a maximum charge for the run, the Actor stops cleanly when it is reached and keeps every row produced so far.

Free plan limits

Users on the Apify Free plan can analyze up to 100 URLs per run. The Actor analyzes the first 100, saves them and finishes with a clear message. Paid plans have no per-run limit.

Use cases

• SEO audits: find pages with a missing title, description or canonical URL.
• Link previews: check that every page has an Open Graph title, description and image before a campaign.
• Content inventories: list the title and language of every URL in a sitemap or a spreadsheet.
• AI agents: give an agent a quick, cheap summary of what a list of pages is about before deciding which ones to read in full.
• Monitoring: schedule a weekly run to catch broken pages (4xx and 5xx) and changed titles.

Comparison with alternatives

Option	Good for	Trade-off
This Actor	Hundreds of URLs at once, from the Console, the API, a schedule or an AI agent, with a typed dataset	Reads the initial HTML only; no JavaScript rendering
Social network link preview debuggers	Checking how one URL looks in one network	One URL at a time, no export
Browser-based SEO crawlers	Full site crawls with rendering and dozens of checks	Heavier, slower and priced for full audits
Your own script	Full control	You host, run and maintain it, including retries and error handling

You are responsible for having the right to access the pages you analyze. The Actor stores nothing outside the run's dataset.