Knockoff: a Chrome extension that hides pseudo-brand junk on Amazon
Chrome extension that filters pseudo-brand junk out of Amazon. Buy from real, established brands.
At a glance
- What is it?
- Knockoff filters trademark-squat brand names out of Amazon search results with a local content script and bundled lists. The public repo is a frozen snapshot, so what you load is what you get.
- Who is it for?
- Adopt Knockoff if you shop Amazon marketplaces regularly and want suspect brand names removed from results without sending your browsing to a server; the Chrome and Firefox store builds are the maintained path, and the repo build is for people who want to edit the lists themselves. Skip it if you need per-marketplace tuning beyond the three filter levels, or if you expect the public snapshot to receive fixes.
- Can I use it commercially?
- Check first. The repository uses a licence we do not classify automatically, so read its LICENSE file before any commercial use.
- Is it still maintained?
- Yes. The repository last received commits 62 days ago.
- What is it written in?
- Mainly JavaScript, according to GitHub's language statistics.
Answers come from the project's GitHub data, last synced on September 17, 2026, and from our analysis. They are not legal advice.
DEEP OPEN-SOURCE ANALYSIS
The problem Knockoff targets: brands that exist only to unlock Brand Registry
The README names the pattern directly: trademark-squat "brands" such as SZHLUX, HORUSDY, LATTOOK and DOZAWA, described as random strings registered at the USPTO purely to unlock Amazon Brand Registry. The goods behind them are commodity items with, in the project's framing, no company, no warranty and no reputation. The intended user is someone searching an Amazon marketplace who would rather pay more for a brand they can look up than save a few dollars on a name they cannot pronounce.
That is a narrower audience than "everyone who shops online." Knockoff does not compare prices, check seller ratings or verify reviews. It only answers one question about each product tile: does the brand at the front of the title look like a real company or like a registration placeholder? Everything downstream depends on that judgement.
How the brand pipeline resolves a verdict, first match wins
The README describes a content script that runs entirely locally, with no accounts, no tracking and no network requests. Each product tile's brand passes through an ordered pipeline where the first matching check decides the verdict. Your allowlist wins first and is never filtered. Your blocklist comes second and is always filtered. Then comes the seed list of notorious pseudo-brands in data/flagged-brands.js, which produces a flagged verdict, followed by data/chinese-major.js for established Chinese-owned brands, which yields known unless you turn on a setting that treats them as flagged. Next is the roughly 5,000 established brands split across data/known-brands.js and the bundled community allowlist in data/community-brands.js. Only then do name heuristics run, and a tile with no brand at the front of the title is classed as unbranded.
The heuristics score unknown names on the linguistic signature of squat trademarks: ALL-CAPS strings of five to nine characters, vanishing vowel ratios, unpronounceable consonant runs, un-English letter pairs, non-Latin characters and random internal capitalization. High scores become flagged, mid scores suspect. The known-brands list vetoes the heuristics, which the README justifies with ASICS, RYOBI and HOKA, real brands whose short uppercase names would otherwise look suspicious. Scoring lives in src/detector.js and is described as deliberately readable and easy to tune.
Filter levels sit on top of the verdicts. Relaxed filters only known pseudo-brands plus your blocklist. Standard, the default, adds suspect-looking names and unbranded listings. Strict is allowlist-only: anything not on a known-brands list gets filtered. Filtered items can be hidden behind a floating pill showing the count with a one-click reveal, dimmed with a fade and desaturate that restores on hover, or simply labeled. Product detail pages get a verdict chip next to the brand byline, and the README states the page is never hidden out from under you.
Installing Knockoff and running the local build
The maintained builds come from the stores: the README links an Add to Chrome entry on the Chrome Web Store and an Add to Firefox entry on Firefox Add-ons. The repository build is a different thing, and the README is explicit that it is a frozen public snapshot whose development has moved to a private repository. It makes zero calls to any Knockoff server and will not receive further updates.
To load that local build, clone the repo and point Chrome at it:
git clone https://github.com/Shpigford/knockoffThen open `chrome://extensions`, turn on Developer mode in the top right, click Load unpacked and select the repo folder. There is no build step, because the extension is plain JavaScript loaded directly from the repo. The README states it works on every Amazon marketplace with no server behind it.
Safari is a separate path, since the extension has to be wrapped in a native app. Open `safari/Knockoff/Knockoff.xcodeproj` in Xcode, run the Knockoff scheme, then enable Knockoff under Safari, Settings, Extensions. For unsigned local builds the README says to first check Allow unsigned extensions in Safari's Develop menu. The Xcode project carries its own copy of the extension files, so after editing the extension you run `scripts/sync-safari.sh` before rebuilding.
Your first real use is tuning the lists. A real brand that gets filtered belongs in data/known-brands.js; a pseudo-brand that slips through belongs in data/flagged-brands.js. Because there is no build step, edits take effect immediately in a local build. CONTRIBUTING.md has the full rundown.
The frozen snapshot trade-off and the three constants that decide it
The most consequential limitation is stated in the README itself: this repository is a frozen public snapshot. Active development has moved to a private repository, and future versions are no longer built in the open. The last push was on 2026-07-17, and the newest release listed is v0.7.2 from 2026-07-13. Nothing in the repository suggests the public copy tracks the store builds.
The practical effect is that the bundled data is the whole list. The README says there is no daily refresh, the roughly 5,000 known brands and the community allowlist ship in data/, and the selectors the scanner uses live in data/config.js with no config push. The one-click wrong-verdict button opens a prefilled GitHub issue rather than POSTing to an API.
Three constants at the top of src/content.js control all of this: BRANDS_URL, CONFIG_URL and REPORT_ENDPOINT. They ship blank, which is what keeps the build offline. Point them at your own host to re-enable on-demand list refresh, config pushes and structured reports. The README notes the server itself is not part of the repo, so anyone re-enabling those paths is writing or hosting that side themselves. Reports sent to a custom endpoint carry brand, verdict, ASIN, marketplace and extension version, with no PII according to the README.
The second limitation is scope. Knockoff judges brands, not sellers, not prices, not review authenticity. A well-known brand sold by a third-party seller with counterfeit stock is not something this pipeline addresses. The third is that the heuristics are heuristics: the known-brands veto exists precisely because name shape alone misclassifies real companies, and the same logic means a squat brand with an ordinary-looking name can pass. Pull requests against this repository will not ship to the published extension, though edits take effect immediately in a local build.
A fourth constraint is platform. The topics list mentions ios-extension and safari-extension, and the repository contains a safari/ directory with an Xcode project, but the README's install instructions cover Chrome, Firefox and Safari only. Nothing in the README describes an iOS install path, so treat the iOS topic as unverified against the documentation.
How Knockoff differs from a price tracker or a review checker
The obvious alternative is not another brand filter; it is the class of Amazon tools that attack the same shopping problem from a different angle. A price-history extension such as Keepa or CamelCamelCamel records price over time and shows you whether today's deal is real. Its data model is a time series per ASIN, and its verdict is about price. Knockoff's data model is a brand name string matched against bundled lists, and its verdict is about identity. Neither substitutes for the other: a price tracker will happily chart a pseudo-brand's price history, and Knockoff will leave an overpriced listing from a real brand untouched.
Review-analysis tools take a third approach, scoring review text and reviewer histories to estimate whether a listing's reviews are trustworthy. That is closer to Knockoff in spirit, since both try to separate genuine from manufactured, but the signal is different. Knockoff deliberately avoids review data, which is why it can run with zero network requests: the brand lists are static files in the repository. A review-analysis tool generally cannot, because review graphs change constantly. If your actual concern is fake reviews on a legitimate brand's listing, Knockoff is the wrong tool by design.
Licence, maintenance and what upgrading costs
The repository's licence is reported as NOASSERTION, meaning GitHub could not match the LICENSE file to a known licence template. The LICENSE file exists at the top level, so the terms are written down, but a tool cannot classify them. Read that file before you fork or redistribute, and do not assume a permissive licence. This is a description of the repository metadata, not legal advice.
Maintenance splits in two. The store builds are the maintained path according to the README, which says future versions are built in the private repository. The public snapshot is not updated: last push 2026-07-17, newest release v0.7.2. Upgrade cost follows from that. If you run the store build, updates arrive through the browser's extension mechanism and you do not manage anything. If you run the repo build and edit the lists, your edits live only in your copy, and there is no upstream to merge from. If you also host your own BRANDS_URL, CONFIG_URL and REPORT_ENDPOINT, you own that infrastructure and its upkeep, since the server is not in the repo.
Editorial conclusion
Adopt Knockoff if you shop Amazon marketplaces regularly and want suspect brand names removed from results without sending your browsing to a server; the Chrome and Firefox store builds are the maintained path, and the repo build is for people who want to edit the lists themselves. Skip it if you need per-marketplace tuning beyond the three filter levels, or if you expect the public snapshot to receive fixes. Before installing, open data/flagged-brands.js and data/known-brands.js to check whether the brands you buy are classified the way you expect, and confirm in src/content.js that BRANDS_URL, CONFIG_URL and REPORT_ENDPOINT are blank so the build stays offline.
Frequently asked questions
How do I use Knockoff?
Install it from the Chrome Web Store or Firefox Add-ons, then browse Amazon as usual: the extension scans product tiles and hides, dims or labels listings whose brand it flags. You can tune the result with the allowlist, the blocklist and the three filter levels (Relaxed, Standard, Strict), and every badge is clickable so you can trust a brand, block it, show an item once, or report a misclassification.
What is Knockoff?
Knockoff is a browser extension that filters pseudo-brand junk out of Amazon. According to the README, it detects trademark-squat brand names and hides, dims or labels those listings directly in the search results, running locally in a content script with no accounts and no tracking.
Is the Knockoff extension safe?
The README states that everything runs locally in a content script with no accounts, no tracking and no network requests of any kind. The public repository build ships with BRANDS_URL, CONFIG_URL and REPORT_ENDPOINT blank, so it makes zero calls to any Knockoff server and relies on the brand lists bundled in data/.
Is Knockoff for Amazon legit?
The repository is a public snapshot of a Chrome and Firefox extension that has been covered by outlets listed in the README, including Fast Company, Gizmodo, 404 Media, PC Gamer, Yahoo and Lifehacker. The README also notes that development has moved to a private repository and that the public snapshot will not receive further updates.
Is knockoff.shopping legit?
The README does not mention knockoff.shopping. It points to knockoff.co as the project homepage and to the Chrome Web Store and Firefox Add-ons listings as the places to install the extension, so there is no basis in the documentation for judging that domain.
Community notes