More or less, but most harnesses will rely on tools such as grep to only read portions of files. For even a small sized repo, only a small portion would be tokenized and uploaded
>For even a small sized repo, only a small portion would be tokenized and uploaded
Only on a per-chat basis. Over time, it'll eventually grab the entire repo, or enough of the "secret sauce" that the rest can be reconstructed with AI.
Only on a per-chat basis. Over time, it'll eventually grab the entire repo, or enough of the "secret sauce" that the rest can be reconstructed with AI.