This is personal AI assistant that communicates with the user using Telegram. It's bit of a personal project that I decide to make public, so I am forced to put sensitive information such as keys somewhere else that in to the source code. Originally this was very focused on using AI to manage stuff like my timers or notes, but either the models are too expensive for me to use on daily basis, or they aren't smart enough to actually perform operations I need from them consistently, without me having the check their work. So while the AI is still part of this bot, it's taking a back seat until it gets to a point where it can handle more complex tasks without needing babysitter.
NOTE: I'm slowly getting this project to a state where anybody could run their own version, it currently only needs TELEGRAM_KEY, OPENAI_KEY and VALKEY_URI passed as environment variables, I used aiven to host my Valkey instance, but it should be possible to run one on local host.
Currently, the AI assistant can create reminders, create and manage files and I allowed for it to be able to see send images.
- add google calendar integration
- implement retrival mode
- novelAI image generation (maybe even text if I'm bored)
toggle_retrival- doesn't work yettoggle_debug- Turns debug mode on allowing you to see what functions and tools where used and what their results where. Great to see if the AI actually did what it should have or if it just made stuff up.clear_thread- Deletes current thread and creates new one, basically deleting current chat history from the AI memory. Important as right now the history grows until it reaches maximum tokens which can be expensive.
get_current_time- return current UT datetimecalculate_seconds- calculates seconds, hopefully the AI can use this instead of doing math itselfadd_reminder- adds reminder, that will send message to the user when it expiresget_reminders- returns list off all reminders and their indexesremove_reminders- accepts array of indexes of reminders that are to be removedcalculate- Uses wolfram alpha to calculate expressions. This is way to avoid the AI having to do math itself or use code interpreter to do it.
Looks like function output should be formatted in the style of json or at least provide context as a see better results when I do.
I can only guess this is because the AI doesn't remember what functions it called so when it calls for example get_current_time and gets only time like 10:20:34
it just remembers some numbers as result and doesn't know the context of them but if you return {"current_time":"10:16:19"} it will now know that what it got is current time.
Example of this: The AI really struggles to use get_current_timewith set reminder function to create reminders for specific times not in specific time.
After some more testing the problem could have been with the fact that I was returning time that had seconds as float making the AI not recognize it necessarily as time.
The AI has the ability to recognize errors and try again, but it seems this is very strict as it will only recognize outputs that specifically say that the function call failed. When your function returns result that is clearly incorrect it doesn't seem to recognize it as an error and will pretend the funtion did what it should have.
They provide nice speciality to them allowing them to perform specific tasks better than general assistant can. But the question is if they are needed or cheaper that just using GP-4, as this model seems to just be able to handle most task. Unfortunately because of the cost I can't test with the more advanced model.
