No LLM scenario #482
No LLM scenario #482
Conversation
✅ Deploy Preview for fluxdocs ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
✅ Deploy Preview for fluxsocial-dev ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
| conversation = allConversations.find((c) => | ||
| subgroupLinks.find((link) => link.data.source === c.baseExpression) | ||
| ); | ||
| subgroups = [subgroup]; |
There was a problem hiding this comment.
I think this could be done in more concise way with one prolog query...
|
|
||
| resolve(); | ||
| resolve(); | ||
| } |
There was a problem hiding this comment.
This is a pretty long function with several blocks with different concerns. It would help to extract meaningful functions...
There was a problem hiding this comment.
for instance having one function to collect all the needed data like the subgroup and previous messages..
| } | ||
| } | ||
| } | ||
| if (!conversation) { |
There was a problem hiding this comment.
I would think of this function as like a method on conversation - so maybe at least pass in the conversation as paremeter so we don't have to deal with this case here too.
LLM processing if new items (messages, etc.) gets signalled through an "empty" SemanticRelationship on that item. Agents without LLM configured won't create any link. There is new UI to show to users which items are being processed and if there are any items that nobody processes. Users with an LLM config can take on the processing of those items by clicking a button.