MCP (Model Context Protocol) is an open-source standard for connecting AI applications to external systems. Using MCP, AI applications like Claude or ChatGPT can connect to data sources (e.g. local files, databases), tools (e.g. search engines, calculators) and workflows (e.g. specialized prompts)—enabling them to access key information and perform tasks.
Think of MCP like a USB-C port for AI applications. Just as USB-C provides a standardized way to connect electronic devices, MCP provides a standardized way to connect AI applications to external systems.
While existing AIs[1] have a vast knowledge base, they lack the specialized tools.
Imagine a complete assistant for your system
fetch what you need from Wiki, official repos or from AUR
run 50+ security checks before installing any AUR Pacakge
troubleshoot your system with AI assistance
and much more
While I agree that there could be some uses for this on certain environments, I must say that I am quite concerned about privacy. Although I like to play with AI’s, I use them simply for generating silly pictures, some short references from long texts or to get some info about some subject in general.
But I would never let internet based AI to access to my system.
It is meant to assist the user and not take complete access to the global system or directly make actions on the system without user commanding it to, which most of it is out of the scope of the program.
It is just installed in a python environment and make calls to the wiki and aur.
A member who has only participated in 2 Threads in 5 months one of which is new that they started to “SPAM” their content and we should not criticize? A welcoming Community does not mean free from Criticism.
We are not here to be lab rats for whatever project People may want to work on. We are here to help people with issue’s and to have a little fun doing so. Please spam your stuff in more appropriate places.
Its one thing for a Member who is active in helping others to ask people to have a look at their project its another to come here offer nothing but then expect “Community Support”.
Sir, I understand where you are coming from and I too would be skeptic of someone new. It is unfortunate that I can’t prove it now as I don’t do that from my main account.
I have assisted many times in both forum and telegram; I have interacted with you earlier as well, and that was nothing like this.
From my main account I do these things: like this appreciation post
and this project that can actually assist users, which you call “SPAM” and using the community as lab rats.
Criticism comes in two forms- Constructive and Destructive.
Aren’t we in “Community Contribution”? or am I not part of the community?
It is one thing to not like the project or it’s “AI” involvement
and another thing to take the space for people for who actually could and want to make use of it.
If you don’t want to use it then that’s your personal choice and I respect that, why to ruining the discussion for others
Honetly don’t care about the A.I part at all. My issue is and will always be that ARCH is a DO IT YOURSELF system. This usually means people want to be in control of their system. Depending on Any out side is a contradiction to that control.
AI can occasionally be useful, although I have found it to be wrong as often as right, because the training doesn’t take into account the quality of the data being used. For example, if they train it on my drivel, they have a problem.
Two things I will never do though is pay for it, and run it on my computer.
If AI generation isn’t the problem what does “outside control” mean? The Arch-Wiki is clearly a bunch of people putting together information on the “outside”. Do they have “outside control” over our systems? Or do we reject the wiki too?
PS: Can a mod maybe split off this subtopic, and leave this thread for people who are interested in OP’s app?
I don’t think they are both exclusive to each other. One is a community project designed to help teach while the other really doesn’t teach you anything. Your trying to say that because both apples and oranges are a fruit they are the same. While half right they are both a fruit they are very different in every way.
I said that because the argument was about “outside control”, but when engaged it moved to a different point of contention. I’m not chasing that goalpost for a fourth time.
Honestly, I still feel some people just don’t like AI, which is fine, people can and imho should have that discussion. On the other hand some people moved on from “if” to “how and what can it do for me”. That should be respected too and can’t be an invitation to roll out the whole “if” debate every time.
integration is planned!
