Overview
We are excited to announce a powerful new addition to our Voice AI Agents! You can now create Custom Actions that allow your AI agents to interact with external systems during conversations through Webhook integrations.
What’s New?
- Webhook Integration: Create custom POST Webhook calls to external APIs with support for authentication, headers, and dynamic parameters collected during the conversation.
- Conversation Triggers: Define specific conversation scenarios that trigger your Custom Actions, enabling seamless execution of external processes.
- Dynamic Parameter Collection: Automatically extract information (strings, numeric values, emails, phone numbers, and more) from conversations to populate API calls.
- Real-time Testing: Test your webhook configurations directly in the interface before saving, ensuring integrations work correctly with your external systems.
What Does This Mean?
- Live Data Fetching: Previously, data could only be extracted after calls ended; now, agents can fetch and use data mid-call.
- Enhanced Flexibility: Integrate nearly any external service or database, opening possibilities for dynamic, context-aware conversations.
- Improved Automation: Automate tasks—like CRM lookups or order status checks—in real time without manual intervention.
- Cross-System Updates: If you’re using other systems, Voice AI can now post updates during calls, syncing statuses in restaurant software, medical systems, field service platforms, and more.
Use Cases
- Trigger Zaps & Webhooks: Launch Zapier workflows or custom POST API calls to update contacts, payment status, or order details during the call.
- Update External Systems: Push live status changes to restaurant or medical software, field service apps, or any third-party CRM.
What’s Ahead?
- Expanded Endpoint Support: We currently support POST endpoints; soon, agents will consume API responses directly and leverage returned data within the call.



Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article