Amazon search results are so flooded with fake brand names that finding a genuine product from a real company now requires scrolling past a wall of trademark-squatted garbage. SZHLUX, HORUSDY, LATTOOK, DOZAWA these are not brands. They are random strings registered at the USPTO purely to unlock Amazon Brand Registry, selling commodity goods with no company, no warranty, and no reputation behind them. A new open-source browser extension called Knockoff, created by developer Shpigford, has reached 1,939 GitHub stars in just 12 days by doing one thing Amazon refuses to do: filtering the junk out. Press coverage from Fast Company, Gizmodo, 404 Media, PC Gamer, Yahoo, and Lifehacker confirms this is not a niche tool. It is a mainstream consumer pain point finally getting a technical solution.
What Knockoff Actually Does
Knockoff is a browser extension available for Chrome, Firefox, Safari, and iOS that detects pseudo-brand listings in Amazon search results and hides, dims, or labels them. It runs entirely in a content script on your local machine. No accounts. No tracking. No network requests to any Knockoff server. The extension ships with a bundled snapshot of approximately 5,000 known legitimate brands plus a community-contributed allowlist, and it performs all detection client-side.
The core problem it solves is Amazon brand registry pollution. Sellers register random strings as trademarks at the USPTO, which qualifies them for Amazon Brand Registry. Once enrolled, their listings get priority placement, richer product pages, and enhanced search visibility. Consumers searching for a legitimate product end up facing page after page of listings from fabricated brands that exist only as trademark filings. Knockoff strips these out of the search results, letting shoppers see products from real, established companies.
How the Detection Engine Works
The detection engine follows a six-step pipeline, with the first matching verdict applied. First, the user personal allowlist ensures trusted brands are never filtered. Second, the user blocklist catches brands they personally dislike. Third, a curated seed list of notorious pseudo-brands catches known offenders. Fourth, established Chinese-owned brands are evaluated in a separate toggleable list. Fifth, a community-sourced list of approximately 5,000 known brands serves as the reference corpus. Sixth, if no list match is found, a heuristic scorer evaluates the brand name linguistic signature.
The heuristic scoring is the technical highlight. It analyzes name patterns typical of trademark squatters: ALL-CAPS strings of 5 to 9 characters, vanishing vowel ratios, unpronounceable consonant runs, un-English letter pairs, non-Latin characters, and random internal capitalization. Crucially, the known-brands list always vetoes the heuristics, so legitimate brands like ASICS, RYOBI, and HOKA are never incorrectly flagged despite looking unusual. The detector code lives in a single readable file, making it easy to tune for custom builds. A strict mode catches edge cases like mixed-case gibberish that the standard heuristic threshold might miss.
Three Filter Levels and How to Install
Knockoff offers three filter levels. Relaxed mode blocks only known pseudo-brands plus the personal blocklist. Standard mode, which is the default, adds suspect-looking names and unbranded listings. Strict mode operates as an allowlist, showing only products from recognized brands. Filtered items can be hidden entirely, dimmed with a fade and desaturate effect, or simply labeled with a verdict badge. Each badge is clickable, letting users trust a brand, block it, show an item once, or report a misclassification via a prefilled GitHub issue.
Installation is straightforward. The extension is available on the Chrome Web Store and Firefox Add-ons. Safari users can build from the included Xcode project. Developers who want a fully local build can clone the repo and load it as an unpacked extension. Because there is no build step, editing the brand lists and reloading the extension takes seconds. The repo also includes a self-contained mode where every feature that could phone home is turned off by default, with configuration constants clearly marked at the top of the content script for anyone who wants to re-enable server-side features.
Why It Matters for Founders
Knockoff is a textbook example of the open-source ecosystem solving a UX problem that a trillion-dollar platform refuses to fix. Amazon has tolerated brand pollution for years because more sellers mean more transaction volume, and fake brands still pay referral fees. The platform has no incentive to clean up search results. Knockoff exists because a single developer identified a universal consumer pain point and built a dead-simple solution in days.
For founders, three lessons stand out. First, platform problems that the platform cannot or will not solve are fertile ground for open-source tools. Second, a cross-platform approach from day one maximizes reach. Knockoff launched simultaneously on Chrome, Firefox, Safari, and iOS, with press coverage from six major outlets within two weeks. Third, zero-server architecture means zero operating cost. The extension runs entirely on the user machine with bundled data, so there is no infrastructure to maintain, no API costs, and no privacy concerns to defend.
Who This Is For
Knockoff is for anyone who shops on Amazon and has noticed the search results getting worse. It is for developers who want to study a clean, well-documented content-script architecture in JavaScript. It is for founders studying how open-source tools can solve platform UX failures. And it is for anyone interested in how a simple heuristic-based detector can outperform multi-billion-dollar recommendation systems at a single well-defined task: telling the difference between a real brand and a trademark squatter.