Match every paragraph. Measure what made it to the page.
A free Chrome extension that checks every paragraph of your document — Word, ODT, RTF, Markdown, HTML, or plain text — against the live page, flagging anything altered, missing, or duplicated. It all runs in your browser, never on a server.
No upload · No account · Ad-free · Reads files in your browser
Doc in. Page verified.
Match any document
Drop a Word .docx, ODT, RTF, Markdown, HTML, or text file — or paste text. Each paragraph is read locally and compared to the live page; nothing is uploaded.
Four clear verdicts
Every paragraph lands as found, altered, missing, or duplicated. Altered carries a coverage score; duplicated counts only page copies beyond your document's own — and ignores nav and footer echoes.
Scope, then filter
Limit the check to a paragraph range with start and end markers. Status pills filter the results, and clicking any line scrolls to and highlights it on the page — tagged with the element it landed in (a heading, list item, or cell).
Read the metadata
See the page title, meta description with character counts, the canonical, and whether the page is indexable — read live from its robots meta, right beside the results.
See the structure
H1–H6 counts plus internal and external links. Filter results by heading level, or list every link and jump straight to it on the page.
Measure the coverage
A word-count graph weighs your document against the page, and the matcher reads contiguous word-runs — so genuinely edited copy is told apart from coincidental overlap.
Drop a doc. Check every line.
From a file on your desktop — or text on your clipboard — to a paragraph-by-paragraph verdict on the page you already have open.
Your document never leaves the tab.
DocMatch reads your file right inside the browser — .docx and .odt with a bundled copy of JSZip, everything else parsed in place — and compares the text against the page already open in your tab.
DocMatch makes no network requests at all. No backend, no analytics, no trackers. Your document, the page text, and the report never leave the tab.
What's new in DocMatch
Every public release of Sukat DocMatch, dated and versioned. Updates ship through the Chrome Web Store — if auto-updates are on, you're already on the latest.
Sharper paragraph diffing. DocMatch now spots paragraphs that appear in the wrong order, pins where dropped words belong, and presents altered text more cleanly.
Added
- Out-of-order detection. Any paragraph that appears on the page in a different order than the document is now flagged. Out-of-order paragraphs are marked in blue, and a pin shows the slot where each one belongs. A new "out of order" filter isolates them. Repeated blocks — such as a call-to-action at both the top and bottom of a page — stay in order as long as one copy sits in the right place.
- Deletion markers. When an altered paragraph drops words that are in the document, a red pin marks the exact spot on the page where they belong, labelled with the missing text.
- Dropped words on the result. Altered results now list the words that are in the document but never made it to the page.
Changed
- Added text shows as a red strikethrough. Words that appear on the page but not in the document are struck through in red, replacing the earlier highlight fill.
- Cleaner altered paragraphs. The difference is carried by the red strikethroughs (additions) and the pins (deletions), so altered text is no longer covered by a block highlight.
- Stricter altered-versus-missing matching. A paragraph is now compared against the single page block it best matches, rather than words scattered across the whole page. A paragraph whose words are spread across unrelated sections is now correctly reported as missing instead of altered.
First public release on the Chrome Web Store. Add a document — or paste text — and DocMatch checks every paragraph against the live page, without anything leaving your browser.
Added
- Six document formats. Reads Word
.docx, OpenDocument.odt, Rich Text.rtf, Markdown, HTML, and plain text — or paste text directly..docxand.odtare unzipped locally with a bundled copy of JSZip; no file ever leaves the browser. - Paragraph matching. Compares each paragraph to the page after normalizing smart quotes, dashes, ellipses, non-breaking spaces, and case. Matches on contiguous word-runs, so edited copy is told apart from coincidental word overlap.
- Four verdicts. Tags every paragraph found, altered, missing, or duplicated. Altered shows the share of words present (capped at 99%); duplicated counts page copies beyond the document's own count, and ignores nav/footer chrome and hidden clones so they aren't false-flagged.
- Coverage filters. Summary pills — all, found, altered, missing, duplicated — double as filters.
- Click to locate. Click any found or altered result to scroll to and highlight it on the live page; duplicated rows step through each occurrence.
- Element tags. Each located paragraph is labelled with the element it matched — heading, list item, table cell, blockquote — so you can tell body copy from a heading at a glance.
- Scope markers. "Only check between" start and end fields scope the comparison to an inclusive paragraph range, with an optional toggle to exclude the marker lines themselves.
- Page metadata. Title and meta description with character counts, canonical match, and an indexable / noindex read from the page's robots meta.
- Page structure. H1–H6 counts plus internal and external link lists; filter results by heading level, or list and jump to any link. A word-count graph compares the document to the page.
- Reads hidden text. Checks all body text, including collapsed accordions and hidden sections. (Image alt text isn't checked.)
- Side panel mode. Opens in Chrome's side panel with a sticky summary, re-checks as you move between tabs, and remembers your loaded document and last result while the browser stays open.
Want a heads-up when DocMatch updates? Drop a line and Bernard will tag you on the next release post.
Common questions
Does it cost anything?
No. Free forever, no account required, no premium tier, no ads — the same model as the rest of Sukat.
What files can it read?
Word .docx, OpenDocument .odt, Rich Text .rtf, Markdown (.md), HTML, and plain text (.txt) — or you can paste text directly. Files are read locally (.docx and .odt via a bundled copy of JSZip) and never leave your browser. Legacy .doc and PDF aren't supported; save or export to one of the formats above.
What does "altered" mean?
A paragraph DocMatch found on the page, but not word-for-word. It needs a real anchor — three or more words matching in a row — plus at least 70% of the paragraph's significant words present, after normalizing quotes, dashes, and spacing. The badge shows that coverage, capped at 99%.
What counts as "duplicated"?
The same text standing on its own as a separate page element more times than it appears in your document. Repeats inside site chrome — a heading echoed in the nav or footer, or a hidden responsive copy — are ignored, so they aren't flagged.
Does it check page metadata and structure too?
Yes. Beside the paragraph check, DocMatch shows the page title and meta description with character counts, the canonical, and whether the page is indexable, plus H1–H6 counts, internal and external links, and a word-count comparison between your document and the page.
Can I check just part of a document?
Yes. The "Only check between" fields set a start and end marker so DocMatch compares only the paragraphs in that range — with an optional toggle to exclude the marker lines themselves. Handy when a page renders one section of a longer document.
Does it check hidden or collapsed content?
Yes. DocMatch reads all body text on the page, including collapsed accordions and hidden sections, so copy that only appears after a click still gets matched. Image alt text isn't checked.
Will it slow my browser down?
No. The extension does nothing until you add a document and click Check. It doesn't run on every page, doesn't track you, and doesn't sit in the background — it only works when you ask it to.
What pages does it work on?
Any http(s) page. Browser and internal pages (chrome://, settings) and the Chrome Web Store are off-limits, because extensions aren't allowed to read those.
What's the difference between Sukat, Inspector, Crawler, and DocMatch?
Sukat (this site) compresses images to an exact KB/MB target. Inspector audits the images on a page. Crawler audits the crawl layer — sitemap, robots.txt, llms.txt, AI bots. DocMatch checks that a document's text actually made it onto the page. Four tools, one privacy model: everything runs in your browser.
Is my data really safe?
Yes. DocMatch makes no network requests at all — there's no backend, no analytics, and no trackers. Your document and the page text are compared locally and never transmitted. Full disclosure here.
Firefox / Safari / Edge support?
Chrome and Chromium-based browsers (Edge, Brave, Arc, Vivaldi) work today. Firefox support is on the roadmap. Safari requires significant rework due to Apple's extension model and isn't planned yet.
The full privacy policy
Last updated June 7, 2026. This covers the DocMatch extension; the Sukat website has its own website privacy policy.
Sukat DocMatch runs entirely inside your browser. It makes no network requests of any kind. Your documents and the pages you check are read locally on your device and are never uploaded, stored on a server, shared, or sold. There is no account, no advertising, and no tracking or analytics in the extension.
What DocMatch does
DocMatch compares the text of a document you provide — by uploading a file or pasting text — against the content of the web page open in your active tab, then reports which paragraphs are found, altered, missing, or duplicated. Every step of this happens locally in your browser.
Information DocMatch accesses
Your document. When you upload a .docx, .odt, .rtf, .md, .html, or .txt file — or paste text — DocMatch reads it in your browser to extract its paragraphs. Word and OpenDocument files are unzipped locally using a copy of the JSZip library bundled inside the extension; every other format is parsed in place. Your file is never uploaded or copied off your device.
The current page's content. When you run a check, DocMatch reads the text of the page in your active tab so it can compare it against your document. It also reads that page's title, meta description, canonical link, robots directive, headings, and links to populate the metadata and structure panels. Page content is read only when you run a check — including the automatic re-checks that happen when you switch tabs while the side panel is open — and is used solely to produce your on-screen report. It is never transmitted anywhere.
DocMatch does not access your browsing history, cookies, passwords, form data, or any page other than the one you are actively checking.
Information DocMatch stores
So you don't lose your place during a browsing session, DocMatch saves your loaded document and most recent result in your browser's session storage. This data:
- stays on your device,
- is never synced to an account or sent off your device, and
- is cleared automatically when you close your browser.
DocMatch sets no cookies and keeps no long-term profile of you or your activity.
What DocMatch does not do
- No servers and no network requests of any kind. There is no backend for the extension to send anything to.
- No analytics, telemetry, tracking, or fingerprinting.
- No advertising, and no selling or sharing of any data.
- No remote code. All code and assets — including the JSZip library and the fonts — are bundled inside the extension package.
- No account, sign-in, or email collection.
Permissions, and why each is needed
DocMatch requests the minimum permissions required to do its job:
scripting- Reads the text of your active tab when you run a check, and scrolls to and highlights a matched paragraph on the page. It runs only on the tab you are checking, and only when you act.
storage- Remembers your loaded document and last result in session storage during your browsing session, so reopening the panel doesn't lose your place. Cleared when you close the browser.
sidePanel- Lets you open DocMatch in Chrome's side panel instead of the toolbar popup.
- Host access —
http://*/*andhttps://*/* - Chrome requires this so the extension can read the page in your active tab when you check it. DocMatch does not run automatically on websites and never reads a page except on your action.
Pages DocMatch will not read
Browser and internal pages (such as chrome:// addresses and your settings) and the Chrome Web Store are off-limits — browser extensions are not permitted to read those, and DocMatch does not attempt to.
Data sharing
None. Because DocMatch makes no network requests, there is nothing to share. DocMatch does not sell your data and does not transfer it to third parties. In the terms of the Chrome Web Store developer program, your data is not sold to third parties; it is not used or transferred for any purpose unrelated to the extension's single purpose; and it is not used or transferred to determine creditworthiness or for lending purposes.
Children's privacy
DocMatch is a general-purpose utility and is not directed at children. It does not knowingly collect personal information from anyone, including children.
Changes to this policy
If this policy changes, the "Last updated" date above will change with it, and the current version will always be available on this page.
Contact
Questions about privacy or the extension? Reach the developer, Bernard Brillo, via the Sukat contact page.
Ready to check the page?
Free, no account, ad-free. Add DocMatch to Chrome and check your first page in seconds.
Add to Chrome — Free →Or pair it with Sukat Inspector and Crawler — DocMatch checks the words, Inspector checks the images, Crawler checks the crawl layer. One privacy model end to end.