A son’s fix for his parents’ TV became a 35,000-star GitHub project

A developer who used Claude to strip ads from his parents’ TV has turned the method into an open-source project, CleanYourTV, which shows 35,100 GitHub stars and a banner claiming 307 TV models. A Reddit post pointed to it.

The repository, maintained by Gathin23 and described by its author as a way to debloat any Android or Google TV over Wi-Fi, is licensed under MIT. Its project page, which we read directly, is the basis for the details below.

How the TV gets cleaned without a cable

According to the README, the tool uses Android’s debug bridge and the pm disable-user command. Nothing is uninstalled, no root access is needed, and the changes can be undone with one restore script. A coding assistant can run the process, or a user can run the included tools by hand.

The project then replaces the home screen with a launcher styled like Apple TV. The README describes it as a native app of roughly 520 KB. It also lists themes, add-ons, a privacy audit tool and per-model debloat lists contributed by users.

The origin is modest. As the tech site It’s FOSS reported on October 19, the work started as an afternoon fix for the author’s parents’ Xiaomi Mi TV, which showed an ad before Netflix opened. The same article said the first run on a TV with 1 GB of memory took an afternoon, while people working from the repository as context reported about fifteen minutes.

Which TVs it covers and which it does not

The numbers on the page do not fully agree. A banner says 307 TV models, 87 brands and 28 languages, while the body text refers to more than 170 models from over 200 contributors. The headline figure of 300 models is therefore the project’s own count, and the page does not say how it was verified.

Coverage stops at Android TV and Google TV. Samsung’s Tizen and LG’s webOS are different platforms and are not supported. Fire TV gets only the debloating half, and the README warns that some TCL sets sold in China refuse any replacement launcher. Roku is also outside the scope, according to It’s FOSS.

Anyone with a supported set needs to enable Developer Options, switch on USB or wireless debugging, and connect a computer on the same network. It’s FOSS names Claude Code and the Claude desktop app among the agents people use, and says Codex and other coding agents also work. No fee is attached to the project itself.

Why disabling packages is not risk free

The project’s pitch is reversibility, but outside reports show where it can go wrong. It’s FOSS advises disabling rather than uninstalling, because removing packages can break the Home button or HDMI input switching. Order also matters: disabling the stock launcher before installing a replacement can drop the TV into a factory test app.

There are security trade-offs as well. Turning off firmware updates removes security fixes, and the debug port should be closed afterward, since an open ADB port is reachable by other devices on the network. A search of coverage of the trend also turned up user reports of installations that stopped halfway and of agents disabling packages a TV needs. Those reports are anecdotal, and we could not independently confirm them.

It’s FOSS also noted that some Claude sessions refused prompts phrased as “hack my TV”, and that sharing the model number and package list first reportedly worked better. That is a reminder that the tool is a prompt and a set of scripts around an AI agent, not a guaranteed installer.

A related project, dumbtv, targets Amazon Fire TV. According to It’s FOSS, it disables about 65 packages, including Amazon’s analytics components, and swaps in a 41 KB launcher.

Growth that is real but still unproven at scale

The speed is the notable part. It’s FOSS counted just over 1,300 stars on the earlier version of the repository on October 19. The page we read now shows 35,100 stars, 4,200 forks and 2,924 commits, and places the launch between October 17 and 19. The README does not give a date for the latest commit, so the current pace of updates cannot be judged from it.

Stars measure attention, not successful installs. What would confirm the project’s claims is a public record of per-model results: how many of the 307 listed models were tested by someone other than the author, and how many users report restoring a set cleanly. The page does not break that down, and neither the author nor TV makers have commented in the material we read.

For readers, the practical takeaway is narrower than the star count suggests. If a TV runs Android TV or Google TV, a documented, reversible path now exists, and it is free. If it is a Samsung, an LG or a Roku, nothing here applies.