Yeah absolutely, my bad. First time publishing things here and I thought it was attached to the post. https://github.com/kumucode/vigil.git

Thanks man, appreciate it!

Hi Dan, Thanks for your feedback. Much appreciated. For the first question, you click on add and past the image you're currently using on your compose so the app creates a card with the current version. It's a bit manual and tedious at first, but once it's done, it's easier to maintain. I think your idea is great to have the app just ¨find your docker-compose and do the work", but I don't know how to do it yet. I wanted to test it manually first and see how it'd work out.

Vigil tells you if the newer version of the image is a major change or not. If you set it to update your compose automatically it will notify you and create a log, it something goes wrong you can easily revert it from the dashboard. Did I get your question right? Let me know if you meant something else.

Finally, security is an absolute must! I decided to use 2FA because most people won't need to expose it to the web.They'll probably use it on LAN. However, I do have adding OIDC (OpenID Connect) in mind, since many people indeed use Authentik, Authelia (these are the ones I'm familiar with). Since this is the early version, I didn't want to make things too complex and also, I'm vibecoding it, so I'll certainly need some experts out there to help me out to implement it correctly and safely.

If you have any question, just let me know and I'll try my best to answer that.

Hi Dan, I'm also copying the answer from homelab community.

Thanks for your feedback. Much appreciated. For the first question, you click on add and past the image you’re currently using on your compose so the app creates a card with the current version. It’s a bit manual and tedious at first, but once it’s done, it’s easier to maintain. I think your idea is great to have the app just ¨find your docker-compose and do the work", but I don’t know how to do it yet. I wanted to test it manually first and see how it’d work out.

Vigil tells you if the newer version of the image is a major change or not. If you set it to update your compose automatically it will notify you and create a log, it something goes wrong you can easily revert it from the dashboard. Did I get your question right? Let me know if you meant something else.

Finally, security is an absolute must! I decided to use 2FA because most people won’t need to expose it to the web.They’ll probably use it on LAN. However, I do have adding OIDC (OpenID Connect) in mind, since many people indeed use Authentik, Authelia (these are the ones I’m familiar with). Since this is the early version, I didn’t want to make things too complex and also, I’m vibecoding it, so I’ll certainly need some experts out there to help me out to implement it correctly and safely.

If you have any question, just let me know and I’ll try my best to answer that.

Hi markko, you're right, adding a tag would certainly lower expectations and give more clarity to projects. I'll add this on the post and to the repo.

That's actually a good idea. I was thinking about getting the features "working" as intended. I'll give it a try to the "explain back" method and see how it goes. Thanks

Absolutelly vibecoding it with Cloude. I understand a bit of python and html but I'm no dev or technical professional at all. I just wanted to see if I could build something useful without much of technical expertise.

Haha, don’t worry, brother. I hardly ever get defensive, especially in the virtual world. We’re good.

I did start using Portainer and I’m continuing with it—I like it very much. Most of my apps are isolated in a dedicated LXC. I know it might sound counterproductive or even exaggerated, but I like it that way. I had a basic bash script that would text me whenever a new version of Jellyfin was available so I could check it out and update it. I started wondering how annoying it would be to copy that script for every app… so why not centralize it?

That’s when I thought, “Maybe I could use AI and see if it can build a simple app.” It was more about having fun with the tool and seeing what I could do without knowing much. Besides all the valid technical critiques, I think the end result was pretty decent and cool. It kind of allowed me to create something I could only dream of or go bankrupt paying someone to build. I’m sure there are other applications from more reputable sources that do the same, but I think the feeling of bringing an idea into reality is a really good one. Maybe now I can rethink about how to structure a new project, how to avoid some pitfalls and create something even more useful... or at least create a proof of concept so someone else can do it properly. Let's keep in touch

I'd love too if I knew how to do that. I still have lots of things to learn and do before putting my toes on these waters, but I'm glad you showed some interest even to mention an integration with other tools. Thanks for that.

I think I got it what you meant. I've only tested it on a LXC that has one application installed on it, like Jellyfin for instance. I need to run a test on a machine with several applications. I also need to test it to see how it behaves with Portainer. Regarding the option of "import YAML" it just works with one file at the time. I think I'd be nice to handle multiple compose files so users could just drop them and have them uploaded.

'll keep an eye on it and keep asking around. I'm pretty sure there are better ways to do that.

Hi comrademiao. I'm certainly not assuming anything. I'm pretty sure there might be other tools out there from reputable companies or devs. I just wanted to have a way to centralize the information about my apps and have a friendly dashboard so I could keep track of what is outdated and get a notification so I could manually update it. I'm not reinventing the wheel nor claiming anything extraordinary, but I thought maybe sharing this initiative with others would be useful somehow and would be a great opportunity to connect with other people.

I think documenting everything is key so people can clearly see the progress. I didn't know that you could merge the changes to the main branch and that it is a core skill as you mentioned. I kind of noticed exactly what you said "it may change other parts of that file that you didn’t ask for" while doing some tweaks. This AI topic has too many branches and it's a fascinating topic for debates. Thanks for taking the time to give this advice.

In this world of technology, applications there are too many areas of expertise required to make things "functional" I'm not sure if I can learn everything required to make applications at the level of the most popular ones. I'm more interested in general knowledge and putting ideas out there. I still think that getting this project to the public even if it's not that great, is still better than have it just on my computer. So, the main purpose is to hear from people what they think of the project, maybe inspire others with more experience to put their projects out there too. My expectations were pretty low about this project, but it turned out to be a great experience to engage with many people from different background just like you.

Hi Appoxo, I agree that AI can be slop, but it seems like it's getting better over time. I'm not claiming that this is a revolutionary new application, it's more like a prove of concept that I wanted to test out. What does it do? well, you basically have a centralized place to keep tracking of your docker images and get a notification when there's a new release so you can update it manually or automatically as you wish that's the main utility of it. There are a few other things but, maybe thye aren't interesting enough.

That looks really interesting. I didn't know you could do that. I think what tried to do with Vigil, is something simpler more focused on users that are beginners rather than pros. But I appreciate you sharing that

Hi ramielrowe. You made great points here. I'm definitely stepping out of my area of expertise. I also understand that when comes to LLM we must not blindly follow/accept things and having some previous knowledge on the topic you intend to work with, gives you much better results and allow you to spot inconsistencies or more importantly, mistakes. I'm aware of the "positive feedback" that is pretty evident specially on ChatGPT, that's why I try my best to challenge it. I completely understand your analogy on "cognitive debt". It's pretty similar to a reinforcing learning process on humans. If you teach people the "wrong way" and keep reinforcing that without any correction, you know the results.

Regarding the code quality, I'm pretty sure it isn't top notch, that's why I'm sharing it here so people who really understand it can point out the flaws and suggest improvements. What I've learned so far from the feedback in the comments, is that I need to improve the way I communicate my ideas and the purpose of the application. Since this is my first project, and I'm not very familiar with the dev & tech community, I'm learning the do's and don'ts along the way.

That's a great question isn't it? That's why I posted it here, so maybe I can find people interested in working on this project and help me out to clean things up, get it more organized, structured and "free of AI slop". What do you think?

😅 no worries man, your comment just reminded me a popular saying in my local community; "who watches the watchman"

Geez, it's really hard to have normal conversation nowadays on the internet. There's this constant skepticism of AI comments, and I get it. Most of the time I just feel like I'm talking to bots. However, I'm surprised that people really take their time to give their opinions here and they seem to be legit. Weirdly enough, I'm quite a positive and enthusiastic person... or at least I'm trying to practice that. There's been too much negativity around me lately, but who cares about what happens in my life, right haha? Don't need to explain myself to any one, but just felt like I needed to do.