Switch AI agents between regulators and financial institutions

Workflow purpose

This guide demonstrates how to use Proto's platform to hand a citizen off between a regulator's AI agent and a specific financial institution's AI agent, when the issue needs the institution's own account access rather than the regulator's oversight role. A regulator's agent can triage and record a complaint, but it usually can't see a citizen's actual account – this workflow hands the conversation to the institution's agent, with the complaint reference attached, so the citizen isn't asked to start over.

Who can benefit from this guide:

Central banks and financial regulators supervising many institutions
Financial institutions receiving referred complaints from a regulator
Consumer protection teams coordinating between regulator and institution
Anyone building "no wrong door" complaint referral pathways

Switch between a regulator and a financial institution

This workflow verifies the institution is within the regulator's supervised list, then hands the citizen off with their complaint reference attached. See how to switch AI agents between regulators and financial institutions.

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: Catch the moment a citizen’s issue needs institution-specific account access.

Navigate to the Actions tab in the AI agent settings.
Click “+ Add Trigger” and select the Message Received trigger type.
Name the trigger “Institution handoff”.
Add a description such as “Triggers when the citizen’s issue requires their specific bank or institution’s own AI agent.”

2. Survey – confirm the institution

Purpose: Get an explicit confirmation rather than guessing which institution the citizen means.

Add a Survey action titled Institution confirmation.
Field institution_name – Text – Which bank or financial institution is this about? – Required: On.

3. Send API request – verify supervision

Purpose: Confirm the institution is actually within scope before handing anything off.

Add a Send API request action to the regulator’s supervised-entities list, filtered by {institution_name}. Map the response to is_supervised.

4. AI Agent Network handoff

Purpose: Transfer the conversation with the complaint reference already attached.

Add a Branch action. Condition: is_supervised == true.
If true: AI Agent Network handoff to {institution_name}’s AI agent, passing {_transferred_context.complaint_reference} and {_transferred_context.issue_description}.
If false: Send message explaining the institution isn’t within this regulator’s supervision, and suggest the citizen contact it directly.

5. Handoff confirmation

Purpose: Tell the citizen what happened and what reference to quote going forward.

Add a Send message action: “I’m transferring you to {institution_name}’s support team with your complaint reference {complaint_reference} attached.”

6. Test the flow

Purpose: Confirm the supervision check and handoff both behave correctly.

Test with an institution known to be supervised, and confirm the handoff carries the complaint reference through correctly.
Test with an institution not on the supervised list, and confirm the citizen gets a clear explanation rather than a failed handoff.
Test with a misspelled institution name, and confirm the lookup either matches a close result or asks for clarification.

550K