A few days ago i made a post on various platforms sharing my software ScreenGuard which received mixed reviews every where , since I AM using it for myself i decided to take in the reviews and am working towards adding necessary features and primarily polishing up things that were left behind and can have some 1st time setup hiccups.
The new features-
* Settings menu added
General tab(control notifications and sounds)
- add notification sounds(custom options too)
-toggle sounds and notifications for various causes
- dark mode(asked bby a few people so added the option)
Tracker Daemon tab
-stop background daemon
- toggle autostart of daemon on boot
- warning when daemon is off Pomodoro timer tab
- Turn on pomodoro in focus mode settings (fixes the unbaked settings of focus mode in v0.1.0 -set break between pomodoros
- no of pomodoro loops
* Daily limits can now be set instead of the fixed 8hr daily limit bug with v0.1.0
* Several changes to Daemon, including some minute layout changes to optimise resources while adding the newer features which are Stated above
Note: this is a beta release and will NOT be pushed to the APT ,COPR AND ARCH repositories for casual users, this is meant for testers and people eager to help me out find the bugs and tiny issues here and there AND YOU CAN HAVE A LOOK INTO THE CODE IN THE BETA BRANCH OF THE REPOSITORY, it hasn’t been pushed to main branch to keep it clear until bugs are fixed.
Release page: (https://github.com/adityakrishnan005-a11y/ScreenGuard/releases/tag/v0.1.1-beta.4)
_____________________________________________________________________________
CLEAR AI EXTENT
Codebase: Manual
Implementation: Manual
Debugging:Assisted
Packaging: generated(specifically for Custom aarch repository due to issues with AUR)
Documentation: Assisted


Yeah, that’s a nono from me. That’s a big security red flag.
Secondly, if goes against the stated goal of having testers find bugs. Part of finding bugs is being able to go through the code and figure out what issues are happening, in what parts of the code.
Finally, this project appears vibecoded. Here is the initial commit: https://github.com/adityakrishnan005-a11y/ScreenGuard/commit/c97cc2e6bd17b35e99ff0bb892aeed41f95cb4e0
That’s a lot of code in the initial commit. And emojis in the .md’s…
There are also some other bad patterns like binaries in the git repo :/
I can see why this project received mixed review on it’s previous pass. Please:
It’s a cool project, sure.
But what makes this project better than something I could vibecode? What makes this project better than the previously existing https://github.com/polesapart/timekpr-next , or other time keeper/time control projects? Can you guarantee that this project will stay maintained, which is a common issue with vibecoded projects.
And about the common issue u say , will it be maintained regularly? I have had only 11 stars on GITHUB honestly…and yet there u can see 2 new beta releases in the beta branch in a difference of about 7-8 days. This is because it’s a project of personal interest and i WILL have to improve it if I want to use it…there is no substitute to this in the community right now i believe and u must know that as well. Also creating the BETA branch and not pushing beta updates to repository IS the correct approach because these are under testing and i am testing it on my own device and VMS but actual devices with defferent environments will be a better test and hence this request.
Pushing to the repository (custom repository of apt and aur ) means that whenever people run sudo dnf upgrade or sudo apt upgrade they get the beta version installed
NOTE: I HAVE MENTIONED THAT THE UPDATES WILL NOT BE PUSHED TO THE CUSTOM APT AND ARCH AND COPR repositories , but you can always have a look at the codebase, just switch to beta branch , both versions are there
Ok improving the extent of llm usage, should have done it from before given my previous experience but thanks and u can go on ahead and confirm that the project is not vibe coded with the code of both the branches, main as well as beta, and the code is fully public sir which part is exactly not public i mean if that part can be cleared I will fix that as well
Currenttly changing the post body here to improve transparency regarding llm usage, i have added it everywhere else.
The initial commit btw is only the initial commit of GitHub , i had been working on this project for a long while and was NOT accustomed to GIT usage much as a result the code base is so huge. Once I brushed up on git i initialise the repository and uploaded the repo publicly. It’s not AI generated sir.
Thank you so much again for pointing out the fatal issues and yes the bug fixes commit u are saying, ok I will look into it and if there is no change in codebase then change it to readme update commit message, but I can assure that there must be some reason I committed that with the bug fixes name , probably I might have left the bug fixes in stash and forgotten to pop it …and commited only the README change by mistake. Once again Thanks a lot