Page meta

curl -s "https://api2.globus.studio/v2/page-meta?url=https://globus.studio/"

The head of a page is where search engines and social networks make their first decision. This endpoint reads one page and returns what those systems see: title, description, canonical, robots, hreflang, Open Graph and Twitter tags, JSON-LD types, headings, images and links.

Length is part of the answer

Title and description come with character counts and a preview cut at 60 and 160 characters on a word boundary. That preview shows what would be lost, which is more useful than a boolean flag. The serp block carries both.

curl -s "https://api2.globus.studio/v2/page-meta?url=https%3A%2F%2Fglobus.studio%2F"

Content checks beyond the head

Headings are counted, with the text of the first H1. Images are counted with three separate defects: no alt attribute, an empty alt used for decoration, and missing width or height. Links are split into internal and external, and nofollow is counted.

JSON-LD blocks are validated for syntax, and the @type values are collected, including items inside @graph. Broken JSON is counted separately, because a syntax error silently removes every rich result on the page.

Issues and limits

The issues array names the problems: missing title or description, over-long text, no canonical, a noindex robots value, zero or several H1, no language, no viewport, no og:image, images without alt, images without dimensions, broken JSON-LD.

Redirects are followed, the HTML limit is 2 MB, and the answer is cached for an hour. Only the requested page is fetched; the endpoint does not crawl the site.

Related

The full parameter table, every error code and a live request are on the documentation page. It opens in the same language as this page and needs no account.

A longer walkthrough, with the checks that matter and the limits worth planning for: Auditing the head of a page without a browser.