Improvement to inline entity creation in Rich Text (Cmd/Ctrl+E)

Inline entity creation in Rich Text is very useful, but it has some limitations that prevent it from being much more useful.

Inline entity creation in a Rich Text is not available if:

  • more than a single line of text is selected
  • the selected text contains an “@”-user reference

There is no way to hook inline entity creation to customize its fields; this must be done afterwards via a normal Rule, and it’s somewhat complicated due to the need to look up the creation Reference, which is also not immediately available.

How could this process be improved?

  • Add a new type of Automation trigger for “Entity pre-creation”, giving the possibility for a script to customize all data being used to create a new entity before it is actually created. The supplied entity Name (markdown) could then be parsed into multiple fields, such as entity Name, Assignees, Due Date, entity Description, etc.

  • The “entity creation context” (i.e. the Rich Text where the entity is being created from) is also key information that allows such a script to perform important functions such as linking the new entity to a specific Story, Feature, etc.

I’m thinking if it would be possible to use “Full add via forms” it would be a good solution too.
Especially for creating multiple things I’d love to be able to use a form to setup the fields that I want.

I do like idea of a rule but can’t think of a use case which won’t be covered by full add

The use case I have in mind is the creation of new Tasks from within a Rich Text of Meeting Notes, as the meeting is ongoing and Notes are being recorded.

Our call admins (who are typing in the notes in real time) do not have time to spare for opening a dialog to enter all the fields for a new Task, but they can type in the task name, assignee, due date and description directly into the Rich Text, then select it and Ctrl+E to create the task.

Or they could, but the current limitations of Ctrl+E do not allow for this to work :cry:

My hunch would be that this could be solved with ai command when it’s able to call fibery api’s.
Something like

Create task based on the input and map fields appropriately