The CallRecordingInsightTextSelectionAction API is experimental and may change in the future. It has been marked as @deprecated to remind the developer that it is experimental. It is not "deprecated", just experimental. Please don't use it in production versions of your app. We do not guarantee backward compatibility, or that the API will remain stable.
import {type CallRecordingInsightTextSelectionAction} from "attio/client"
Any file with a named export of callRecordingInsightTextSelectionAction will register a call recording insights text selection action.

Example

TypeScript
import {type CallRecordingInsightTextSelectionAction, showDialog} from "attio/client"
import {ProcessCallInsightDialog} from "./your-code"

export const callRecordingInsightTextSelectionAction: CallRecordingInsightTextSelectionAction =
  {
    id: "process-call-insights",
    label: "Process insights",
    onTrigger: async ({ markdown, text }: {
      markdown: string
      text: string
    }) => {
      await showDialog({
          title: "Process call insights",
          Dialog: ({hideDialog}: {hideDialog: () => void}) => {
            return (
              <ProcessCallInsightDialog
                markdown={markdown}
                text={text}
                onDone={hideDialog}
              />
            )
          },
        })
    },
  }

Arguments

id
string
required
The unique identifier for this call recording insight selection action.It is only used internally; never shown to the user.
label
string
required
A human-readable label of the call recording insight selection action that will be shown to the user when they make a text selection on a call recording insights.
icon
string
An AttioIcon to display beside the label.
If no icon prop is provided, it will default to your app’s icon that you set up in the Developer dashboard.
onTrigger
async (selection: {text: string; markdown: string}) => Promise<void>
The function to run when the action is triggered. You’ll likely want to show a dialog or run a server function here.The function will be given an object containing:
  • markdown – a markdown respresentation of the text selected (e.g. may include bullet points)
  • text – the plain text of the text selected