Skip to main content
Configure tasks that execute when the call completes—whether the caller hung up, the agent ended the call, or a human handoff occurred.

​
Overview

  • When it runs: After the conversation ends.
  • Typical tasks:
    • Generate and store a call summary and disposition
    • Persist transcript to CRM or data warehouse
    • Trigger downstream workflows (e.g. ticket creation)

​
Create a Post-Call Action

  1. Open Studio → Build → AI Agents → Actions tab → Post-Call.
  2. Click “Create New Action” and fill in:
    • Name: Short, descriptive identifier
    • Description: What the action does and why
  3. Configure the request:
    • Method: GET, POST, PUT, DELETE, PATCH
    • URL: Static URL
    • Query Params: Key–value pairs, variable-aware
    • Headers: Static or variable-aware (e.g. auth tokens)
    • Body: JSON request body for non-GET methods
    • Auth Profile: Optional reusable credentials profile — see Auth Profiles for details. Also supports mTLS for APIs that require certificate-based mutual authentication.
  4. Click Send to test the request and preview the response (Data and Meta tabs).
  5. Map response fields to variables:
    • Pick response paths to save into variables if you need to store references for analytics.
  6. Save.
Response previews help you build a sample structure so you can reliably map fields to variables.

​
Best Practices

  • Keep post-call operations idempotent; the same call may be retried.
  • If you need both summary and transcript, design your backend to accept a single payload to minimize network calls.
  • Store external IDs (e.g. ticket number) for cross-system traceability.