Following the implementation in v4.48 by @pierre-gilles, it would be interesting to be able to retrieve the AI’s response to use it later in a scene. (See: Message écrit par une IA qui décrit le mouvement détecté d’une caméra - #29 par pierre-gilles)
For example, initially, to play the response on Sonos or elsewhere, or to send it directly to another person.
In a second phase, it could be explored to provide a response model to the AI that would allow retrieving multiple pieces of information.
And that would also allow sending this value to an MQTT device (with the action ‹ control a device ›). 
I’m in favor!
I’ve just developed this feature 
Tested just now at home,
1 Like
It’s merged and it’ll be included in the next Gladys release!
3 Likes
This new feature is available in Gladys Assistant 4.58: