On Mon, Jan 12, 2026 at 4:44 PM Thomas Passin <[email protected]> wrote:

> I have been experimenting with using chatbots during software
> development.  It's hard to communicate code when it's split up into nodes
> in a tree in typical Leo fashion, since you can only paste linear text into
> the chatbot's input box.
>
> I have found that chatbots, or at least CoPilot (essentially a variety of
> ChatGPT), can OCR screen images very effectively. When it can see an image
> of both the tree and a  node body, it can put the text into the context of
> the project. You can use a highlighting tool to draw attention to parts of
> the test you want to focus on.  You can draw a text box around the part of
> the tree to group the parts you are interested.  And you can show body text
> of a second node by opening a Freewin window.
>

Very interesting! Thanks for this report.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS2fg1bmbqXxFY64NdJJsDyake%2Bm%3Dj4Nuo8szpwXMSk6Pg%40mail.gmail.com.

Reply via email to