Emacspeak discussion list

[Tim Cross <theophilusx AT gmail.com>]

Hi Folks,

for the past two days, I have been trying out a little experiment and I
thought it might be something others may find interesting. Not sure if
it is useful, but will let you decide on that.

I decided it was time for me to find a little project where I could
experiment with using AI in software development. I wasn't interested in
the newish 'vibe programming' trend, but was a little intrigued with the
idea of using AI agents to assist in development and maintenance of
software.

So I decided to use emacspeak as my test project. I installed some LLMS
and the opencode CLI program and started to play around a bit. The first
thing I did was tell my coding agent to update emacspeak and get rid of
the warnings when you compile the program. As you would know, especially
if you run development branch of emacs like I do, due to changes in
emacs, there are a number of warnings concerning obsolete functions and
moving away from them to the new alternatives. I asked my agent to fix
all of these, which it did and so far, I've not encountered any errors.

Then this morning I decided to set it a bigger challenge. I asked it to
write a speech server backend to let emacspeak use speech dispatcher as
the speech server. This it has done and it is basically working - in
fact, I'm using it right now. It does have some issues, but they seem
pretty minor and it doesn't do voice locking, which will need further
work. However, I'm pretty impressed with the job it has done. It is
important to note I have not written a single line of code in getting
this to work. 

If you are interested in checking out what I have done (or more
accurately, what I told my AI agent to do!), you can find it at
https://github.com/theophilusx/emacspeak/tree/agentic-maint

Please keep in mind that this is just a proof of concept and a bit of
learning experimentation. There are bound to be some problems and
possibly difficult bugs to track down, but could be a good glimpse into
our future under our AI overlords!

Enjoy. Use at your own risk.