What is the point of referencing scripts if it doesn't read them?

When I’m clicking the plus icon or using @scriptname in the composer message input, the impression I got was that it actually would force the system to re-read that file (or acknowledge that it already has the code in it’s memory) then answer.

Am I using this incorrectly? I just had it say it needed me to share the code in a script that I had already mentioned in the message and added with the + icon.

Do we need to always copy paste the entire script each time for it to actually read it?

Is it hallucinating?

1 Like

have this same problem often. i suspect it is a combo off it being lazy and possibly by design to increase usage on the ai model’s company’s desire for more profit… Going to make a post about it now.

Hey, we are currently tracking down a bug where @ing a file in the Composer does not actually allow the AI to see it!

While we work on this, the simplest workaround is to start a new Composer session if this happens, as that should enable the AI to see the file.