Verify personal ID through AI chat

Workflow purpose

This guide demonstrates how to use Proto's platform to verify a person's identity against a national ID registry or similar authoritative source, as a shared building block other workflows can call before releasing sensitive information. Where "Verify banking KYC through AI chat" is specific to financial onboarding, this is the more general version used across government services, healthcare, and any flow that needs to confirm who it's talking to before continuing.

Who can benefit from this guide:

Government agencies confirming identity before releasing citizen services
Healthcare providers verifying a patient before sharing medical information
Any AI agent that currently asks for the same identity details in several separate flows
Operations teams wanting one consistent verification step to reuse everywhere

Verify personal ID

This workflow checks an ID number against an authoritative registry and returns a simple verified flag other flows can branch on. See how to verify personal ID through AI chat.

Just starting with the platform?

For initial setup of your AI agent – including workspace setup, agent training, and channel deployment – please start here. This workflow guide assumes your AI agent is set up and ready for more advanced configurations.

1. Create a trigger

Purpose: Make this callable from any flow that needs an identity check.

Navigate to the Actions tab in the AI agent settings.
Click “+ Add Trigger” and select the Message Received trigger type, or configure this as a Jump to trigger/action target called from other flows.
Name the trigger “Personal ID verification”.

2. Survey – ID number

Purpose: Get the one piece of information the registry lookup actually needs.

Add a Survey action titled ID confirmation.
Field id_number – Text – Please confirm your national ID number. – Required: On.

3. Send API request – registry lookup

Purpose: Check against the authoritative source rather than trusting the number alone.

Add a Send API request action to the national ID registry with {id_number}. Map the response to registry_match.

4. Branch on the match

Purpose: Return a clear result the calling flow can act on.

Add a Branch action. Condition: registry_match["found"] == true.
If true: Set chat variable – verified = true – then jump back to the calling flow.
If false: Send message – “We couldn’t verify that ID number. Please check it and try again, or reply 1 to speak with an agent.”

5. Test the flow

Purpose: Confirm the flag is set correctly before other flows rely on it.

Test with a valid ID number and confirm verified is set to true and the calling flow continues correctly.
Test with an invalid number and confirm the calling flow doesn’t proceed past the verification gate.
Test the escalation path to confirm a person can still get help if the registry lookup keeps failing.

550K