Open AI Conversation with Assist Control does not work properly with custom intents that have the area slot about core

[balloob]

Yes, you're right that you need to put the args in the description or else the LLM won't know. We should update the docs because intent script doesn't allow specifying fields. Our longer term goal is to just allow scripts to be exposed, and do the same thing.

[home-assistant]

Hey there @balloob, mind taking a look at this issue as it has been labeled with an integration (openai_conversation) you are listed as a code owner for? Thanks!

Code owner commands

Code owners of openai_conversation can trigger bot actions by commenting:

• @home-assistant close Closes the issue.
• @home-assistant rename Awesome new title Renames the issue.
• @home-assistant reopen Reopen the issue.
• @home-assistant unassign openai_conversation Removes the current integration label and assignees on the issue, add the integration domain after the command.
• @home-assistant add-label needs-more-information Add a label (needs-more-information, problem in dependency, problem in custom component) to the issue.
• @home-assistant remove-label needs-more-information Remove a label (needs-more-information, problem in dependency, problem in custom component) on the issue.

_{^{(message by CodeOwnersMention)}}

---

openai_conversation documentation
openai_conversation source
_{^{(message by IssueLinks)}}

[nikito]

Quick addition, I tried this with an intent script that just had a single slot with no area and ran into a smiliar error. In this example the name was not passed to the tool at all.

[nikito]

I have an update, I managed to get this to work! Turns out you need to add a very explicit, more detailed description that explains what arguments are required for this function. Here is the description I added to the intent script to get this working:

description: "Used to show a camera by name in the present area. This function requires two arguments: 'area' and 'name'. The 'area' argument is the present area, and the 'name' argument is the requested camera to display."

Perhaps we can add some additional documentation for this if other's run into this issue? :)

Just to add, I originally tried just adding a description saying "Used to show a camera by name" but that wasn't enough, I had to explicitly explain the arguments. This is true for some other intents I have as well that have the area slot among others. Once I added descriptions clearly explaining the arguments expected and where they should come from they all started working perfectly. :)