Ayalon / Hachshara Motor Policy Agent
Simulation of an operational assistant guiding an agency employee through a motor policy issuance flow across a broker portal, an insurer gateway and Legacy screens — with field filling, risk control, human approval, exception handling and a full log.
This is a product simulation. There is no real integration with Ayalon, Hachshara, TIS, Amigo, BestPrice or any external insurance system. All data is mocked and masked. System labels are neutral.
How the work actually looks today
- 1Insurance agency employee starts in a broker portal.
- 2Employee authenticates and navigates to policy issuance.
- 3Workflow continues in a legacy insurer system.
- 4MILO detects a repetitive policy issuance process.
- 5MILO proposes a reusable, governed Agent Workflow.
- 6The Agent assists with safe fields and navigation.
- 7The Agent stops before sensitive or binding actions.
- 8The employee approves each high-risk step.
- 9Every event is recorded in an audit log.
- 10Legacy alerts pause the run and open an exception.
- 11Insurance history is human-only — never auto-declared.
- 12Final outcome: a policy draft is prepared — not issued autonomously.
Three mocked systems plus the governance layer
Entry point, authentication, service selection.
Navigation from broker portal into insurer production system.
Customer, vehicle, driver, history and premium screens.
MILO logs, approvals, risk policy and exception management.
12 governed steps, end to end
Run the cross-portal workflow end-to-end
Switch between the four modes to see the happy path, a legacy alert exception, a compliance block, and the authentication boundary. The Agent never issues a binding policy — it prepares a draft for human issuance.
- 1Detect broker portalLow→
- 2Authenticate user sessionCritical→
- 3Navigate to issuance serviceMedium→
- 4Detect legacy policy systemLow→
- 5Open policy draftMedium→
- 6Add customer / insuredHigh→
- 7Add vehicle detailsHigh→
- 8Handle legacy system alertsHigh→
- 9Insurance history declarationCritical→
- 10Driver detailsHigh→
- 11Additional vehicle rating dataMedium→
- 12Premium summary & draftCritical
Broker portal · Sign in
Live workflow timeline
0 eventsRisk & Governance Controls
- LowDetect portalAuto-detect
- CriticalAuthentication / OTPHuman only
- MediumNavigate to issuanceApproval
- MediumOpen policy draftApproval
- HighCustomer detailsApproval + Masking
- HighVehicle detailsPreview + Approval
- HighLegacy alertPause + Escalate
- CriticalInsurance historyHuman only
- HighDriver profileApproval
- CriticalPremium summaryPreview only
- CriticalFinal policy issuanceProhibited in simulation
Audit log
0 entries| Time | Actor | Event | Entity | Risk | Data handling | Result |
|---|---|---|---|---|---|---|
| Audit entries will appear here as the simulation runs. | ||||||
Every action is logged
The audit log above grows in real time as the agent runs. Each row records actor, event, masked entity, risk classification, data handling and result — ready for compliance review.
How MILO assembles this as a reusable workflow
- customer_id_masked
- vehicle_plate_masked
- policy_start_date
- driver_profile
- insurance_history_answers
- coverage_preferences
- consent_status
- 1detect_broker_portal
- 2wait_for_user_authentication
- 3navigate_to_issuance_service
- 4detect_legacy_policy_system
- 5open_policy_draft
- 6prepare_customer_details
- 7request_customer_approval
- 8prepare_vehicle_details
- 9request_vehicle_approval
- 10detect_and_handle_legacy_alerts
- 11request_insurance_history_human_input
- 12prepare_driver_profile
- 13request_driver_profile_approval
- 14generate_premium_summary
- 15prepare_policy_draft
- 16block_final_autonomous_issuance
- Never store password or OTP
- Never declare insurance history without human source
- Never override legacy system alerts
- Never issue final policy autonomously
- Always mask sensitive customer, vehicle and policy data
- Always log approvals and exceptions
Why this workflow matters
This use case demonstrates MILO's ability to support complex insurance operations that do not live in one clean modern system.
“MILO is not another RPA script. It is a governed operational layer above fragmented insurance systems — built to assist employees, preserve professional judgment, and create a measurable audit trail.”
Two insurance use cases, one governed agent
Best for demonstrating a structured policy workflow inside a more modern browser process.
Open Migdal simulationBest for demonstrating cross-portal, authentication-boundary, and legacy-system workflow governance.
Run this simulationTogether, these two use cases show that MILO can support both modern browser workflows and messy real-world insurance operations.