Support
Is it free?
Yes. MetaPeek is free: every feature, for everyone. There is no payment, no paid tier, no licence key and no account, and no email address is asked for or collected. There is nothing to buy, so there is nothing to refund.
It says "This page cannot be inspected"
Chrome does not let any extension read its own pages — the new tab page, settings, chrome:// addresses, the Chrome Web Store — or another extension's pages. Open an ordinary web page and try again.
Pages opened from a file on your computer (file://) also need a switch turning on: go to chrome://extensions, open MetaPeek's details, and turn on "Allow access to file URLs".
It says "Could not read this page"
The page refused the script, which happens on a few sites that lock down scripting for everything, and on PDFs. Nothing can be done about those from an extension.
The numbers do not match another tool
- Pixel widths for the title and description are an estimate from letter shapes, calibrated so that about 60 characters fills the title bar and about 155 fills the description. Google's real rendering will differ slightly.
- Word count is of rendered text, the way Google reads it. Anything hidden with
display:noneorvisibility:hiddenis left out — including a page whose whole<body>is hidden, which counts as nothing at all. Text insidescriptorstyletags is left out. Text that is merely scrolled or clipped out of view — a collapsed accordion that keeps its height at zero, an off-screen element — is counted, because Google counts it too: a clipped-panel test page of six words, four of them clipped, reports six. Text inside an open shadow root is counted as well, since Google renders it; text inside a closed shadow root is not, because nothing outside can read it, Google’s renderer included. Tools that count everything in the HTML will give a bigger number. - Readability (Flesch reading ease and grade) is an English formula and is only shown on English pages with at least 30 words.
- Links counts anchors that have an
href. An<a>with nohrefis a placeholder, not a link, and is counted separately.
What MetaPeek does not look at
Added 28 September 2026 after an independent review asked for the scope to be published rather than implied. None of these is a fault; they are the edges of what a page-reading tool can honestly claim.
- The top frame only. Content inside an
<iframe>is a different document and is not read. - The markup, not the HTTP headers. A
noindexsent as anX-Robots-Tagheader, or a canonical sent as aLinkheader, will not appear in the report. Only the tags written in the page are read. - Not
robots.txt. MetaPeek never fetches anything, so it cannot know what your robots file says. - Not Google’s opinion. The report describes what is on the page. It cannot tell you whether Google has indexed it, how it ranks, or what it will choose as the canonical — Google makes those decisions with information no extension can see.
- hreflang is not followed. The tags are listed and their codes checked, but the pages they point at are not fetched, so a missing return link on the other end cannot be detected from here.
- Closed shadow roots. Content in a closed shadow root is unreadable to everything outside it, Google’s renderer included, so it is not counted.
It flagged a canonical that I think is right
"Canonical points at a different URL" compares the canonical tag with the address in your address bar, including any ?utm_ or tracking parameters. If the canonical is the same page without those parameters, the warning is telling you what it found, not that something is broken.