Metadata

The Metadata panel covers the tags that control how a page is indexed and how it appears in search. This includes the title and description a user sees in the results, plus the technical directives that decide whether the page can be indexed at all. It's the first place to look when a page ranks poorly despite good content. A single wrong tag here can suppress everything else.

AI Audit

The AI Metadata Audit button sits at the top of the panel. Until you connect a provider, it reads Unlock AI Audit.

Run it, and it reviews your metadata and returns three sections:

• What's working
• Needs improvement
• Quick wins

Below the review, it suggests SERP rewrites: alternative titles and meta descriptions. Each one has a live character count and a one-click copy. Paste a fix straight into your CMS.

Basic meta

The core card lists each tag with a pass / warning / fail status and a character count. Read it top to bottom, in the order the panel presents it. The title and description carry the most weight, since they're what a searcher actually sees.

At the very top sits Domain age: how long the domain has been registered, shown as an E-E-A-T trust signal (for example, 2y 8m, Established). Fetching it is a server-side lookup, available with a free Extension Key.

Title and description

The panel also checks keyword placement: whether the primary term appears near the front of the title, where it carries the most weight.

Canonical and URL

• URL: flagged for length and readability
• Canonical URL: the panel marks it as self-referencing (the healthy default), different, or missing. A different canonical means the page points elsewhere. That is correct for duplicates, but a problem if it's unintentional. A wrong canonical is one of the most common causes of a page silently dropping out of the index.

Language and internationalization

• Language: the declared lang attribute checked against the language actually detected on the page, plus BCP-47 validity. This catches a mismatch, for example en markup on French content.
• Alternate pages (hreflang): the locale count and sample codes, including x-default

Everything else

• Character set, Author & publisher, and Mobile optimization (the viewport tag, required for mobile-first indexing)
• Built with: the detected technology stack, including frameworks, CMS, analytics, CDN, and more
• Site icon (favicon): type, sizes, and any issues. The favicon now shows next to your result on mobile.

Expanding All meta tags lists every tag the page declares, verbatim: viewport, charset, description, keywords, robots, Open Graph, and the rest. Use it to spot a stray or conflicting directive that the summarized rows above don't call out.

Indexation

A dedicated card states plainly whether the page is indexable. It breaks the verdict into the two signals that decide it:

• Meta robots: the <meta name="robots"> directive (index/noindex, follow/nofollow)
• X-Robots-Tag: the same controls, but set as an HTTP header at the server

Seeing them separately matters. A page can look fine in its HTML while an X-Robots-Tag: noindex header quietly blocks it. That's a problem you can't spot by viewing source.

Discovery files

Checks the files that tell crawlers what exists and what to fetch: robots.txt (crawl rules), sitemaps (the list of URLs to discover), and llms.txt / llms-full.txt / ai.txt. The last group is the emerging set of files that guide AI crawlers to your most important content.

Crawlability & bot access

Shows which crawlers can actually reach the page: Googlebot, Bingbot, GPTBot, Claude-Web, PerplexityBot, and others. Each one is marked allowed, blocked, or restricted.

This is where you catch a page that's accidentally blocking AI crawlers in robots.txt. That's increasingly costly as ChatGPT, Perplexity, and Gemini send real traffic and citations.