You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When asking a long multi-line question, and especially when asking several questions using the --chat option, and after getting answers from shell gpt, quite often it is very hard to spot where the question ended and where the answer begun.
It is visually confusing:
In this screenshot, the question ended at...? The first line of the answer is "To monitor the boot process".
Not easy to spot.
Would it be possible to add some kind of delimiter, or a line, or "****" or ">>>", or someting, to mark the beggining of the answer?
The text was updated successfully, but these errors were encountered:
bionota
pushed a commit
to mountaineerbr/shellChatGPT
that referenced
this issue
Nov 10, 2024
Inspired by Ollama project commit ollama/ollama@c2e8cba, I reviewed all our "[[ -[efs] file.ext ]]" tests to consider _not adding_ files with zero length to the request block, even though they will be recognised as possible file with payloads if the files exists.
Add a separation mark after dumping the contents of a text file/pdf/url and also after running a shell command, which may have dumped lots of text that the user will not recognise as part of hi own prompt at first, according to TheR1D/shell_gpt#638.
---
When asking a long multi-line question, and especially when asking several questions using the
--chat
option, and after getting answers from shell gpt, quite often it is very hard to spot where the question ended and where the answer begun.It is visually confusing:
In this screenshot, the question ended at...? The first line of the answer is "To monitor the boot process".
Not easy to spot.
Would it be possible to add some kind of delimiter, or a line, or "****" or ">>>", or someting, to mark the beggining of the answer?
The text was updated successfully, but these errors were encountered: