Virtuoso QA

Best viewed on a larger screen

This interactive demo mirrors the full Virtuoso QA desktop interface, so it needs a wider screen to be readable.

Open it on a laptop or desktop for the full walkthrough, or continue here and pinch to zoom.

Virtuoso QA

See how Virtuoso QA works

Walk through a guided demo for a consumer insurer: load your Knowledge Base and see how an AI Analyst and an AI Architect turn the EU AI Act and FCA Consumer Duty guidance into requirements and human-approved test journeys covering quote, bind, issue, and a policy document check.

The AI Autopilot then writes a deterministic Virtuoso NLP test, which runs with self-healing selectors - so every AI pricing decision is tested and traced, not just delivered.

?
Project dashboard
Overview
Tags
Webhooks
Share
Tickets

Project dashboard

From the project dashboard you can create shiny new goals - a testing scope within a project.

?
Knowledge Base beta
🔍 Search for a source
Tags
Search tags
File type
Search file types
All (0)
Project (0)
🗂️

You have no active sources

The Knowledge Base is a centralised repository of sources that our agents can use to build your requirements from.

?
Knowledge Base beta
🔍 Search for a source
Tags
Search tags
File type
Search file types
All (5)
Project (5)
SourcesTagsLast change
Quote, Bind & Issue - Consumer Motor Insurance BRDNEW
ProjectBRD_Quote_Bind_Issue.docx
008/09/2026, 09:14 PM
EU AI Act - High-Risk AI Systems (Annex III: Insurance Pricing & Underwriting)NEW
ProjectEU-AI-Act-Annex-III-Insurance.pdf
008/09/2026, 09:14 PM
FCA Consumer Duty - Fair Value Outcome Guidance (FG22/5)NEW
ProjectFCA-Consumer-Duty-FG22-5.pdf
008/09/2026, 09:14 PM
Insurance Distribution Directive (IDD) - Product GovernanceNEW
ProjectIDD-Product-Governance.pdf
008/09/2026, 09:14 PM
PRA SS1/23 - Model Risk Management PrinciplesNEW
ProjectPRA-SS1-23-MRM.pdf
008/09/2026, 09:14 PM
?
Requirements
Requirements
Requirement directives
RequirementsRiskCoverage
Quote Calculation Negative ScenariosHIGH100%
Successful Policy BindCRITICAL-
Successful Policy Issuance and Document GenerationCRITICAL100%

Consumer Duty - Fair Value Outcome

Organization-level directiveChanges apply to every project in the organisation unless overridden at the project level.
Content
Core instruction: apply FCA Consumer Duty fair-value outcome principles and EU AI Act high-risk system obligations when generating test requirements, test conditions, journeys, or test cases for any pricing or underwriting decision - using the uploaded regulatory documentation as a primary reference. For every requirement, user story, or business rule provided, the Analyst must: 1. Analyse the source material as a professional aligned to Consumer Duty and EU AI Act obligations. 2. Derive clear, testable requirements confirming customers receive fair value and that pricing/underwriting decisions remain explainable. 3. For any AI-assisted pricing or underwriting decision, require evidence of traceability: the factors considered, the rationale, and an audit trail sufficient to satisfy Annex III high-risk system obligations. 4. Structure journeys and test cases around observable customer and regulator-facing outcomes, with clear objectives, preconditions, inputs, and expected results - including that issued policy documents accurately reflect the bound price and coverage. 5. Justify key decisions with citations from the regulatory knowledge base - never inventing a citation. If none applies, the Analyst must say so explicitly. The Analyst must apply this guidance rather than merely quoting it, must analyse the underlying test basis rather than surface-level acceptance criteria, and must flag any gaps in the source documentation rather than silently filling them in.
Every AI pricing decision - tested & traced.
?
✨ New Requirements
💬 Chat history

Sources selected:

📄 Quote, Bind & Issue BRD 📄 EU AI Act (Annex III) 📄 FCA Consumer Duty (FG22/5)
AI can make mistakes!
?
✨ New Requirements
💬 Chat history

The existing requirements provide good coverage for the primary success paths. However, I identified three gaps not explicitly covered by the BRD:

  1. Quote Calculation Validation (BR-002, BR-003) - not covered by the existing quote-focused negative scenario.
  2. EU AI Act Traceability (Annex III) - automated pricing/underwriting decisions must be explainable and auditable; not currently tested.
  3. Consumer Duty Fair-Value Outcome - the bound premium, coverage, and issued policy document must match and carry the required disclosure; not addressed by existing functional tests.

I've generated three new requirements to close these gaps, applying the organisation's Consumer Duty - Fair Value Outcome directive throughout.

Requirements preview (3)

?
Requirements
Requirements
Requirement directives
All 6
RequirementsRiskCoverage
Consumer Duty Fair-Value Outcome and Policy Document AccuracyHIGH-
High-Risk AI Pricing Decision TraceabilityCRITICAL-
Quote Calculation Input ValidationMEDIUM-
Quote Calculation Negative ScenariosHIGH100%
Successful Policy BindCRITICAL-
Successful Policy Issuance and Document GenerationCRITICAL100%

Consumer Duty Fair-Value Outcome and Policy Document Accuracy

Overview
Journeys 0
Summary

The purpose of this requirement is to ensure the bound premium and coverage are presented to the customer as a fair-value outcome per FCA Consumer Duty guidance, and that the quote calculation itself completes within the 3-second threshold. It also confirms the issued policy document accurately reflects what was quoted and bound, and carries the required fair-value outcome disclosure.

Description

Fair-Value Outcome and Policy Document Accuracy Validation

This requirement validates that the bound premium and coverage are presented to the customer as a fair-value outcome per FCA Consumer Duty guidance, and that the issued policy document accurately reflects what was quoted and bound.

?
🧭 New structured journeys
💬 Chat history

Requirement: Consumer Duty Fair-Value Outcome and Policy Document Accuracy

AI can make mistakes!
?
🧭 New structured journeys
💬 Chat history

I have designed five test journeys to validate the fair-value outcome and policy document accuracy requirement, as specified in this requirement.

These journeys cover:

  1. Response time for quote calculation.
  2. The full quote, bind, and issue flow end to end.
  3. Existence of an audit trail linking the issued policy document back to the pricing decision that produced it.

I've created corresponding data tables to drive these tests and a new environment for the base URL. Please review the proposed journeys.

Journey structures preview (5)

Consumer DutyFair ValuePolicy Document
Goal
Fair-Value Outcome and Policy Document Accuracy
Data table
Quote - Bind - Issue Data
Environment
Consumer Insurance Portal (UAT)
Journey
Quote to Bind to Issue: Consumer Motor Policy
Journey summary
This journey generates a quote, binds the policy, issues the policy document, and asserts the document accurately reflects the bound premium and coverage within the 3-second quote-calculation threshold.
Checkpoints

Checkpoint 1: Navigate to New Quote Form

Assuming the user is already logged in, go to the base URL, open the Quotes section, and click "New quote" to start a new quote.

    Checkpoint 2: Quote, Bind, and Issue

    Enter the applicant's date of birth and coverage amount, then start timing. Click Get quote, wait for the premium to calculate, and stop timing once it's finished. Assert that quote calculation completed in under 3 seconds, then bind the policy and issue the policy document.

    ✓ 1 journey created successfully
    Goal: Fair-Value Outcome and Policy Document Accuracy
    Consumer Insurance
    ← Back to journey list
    Journey 1 (Draft) : Quote to Bind to Issue: Consumer Motor Policy
    Consumer DutyFair ValuePolicy Document

    ✨ Autopilot · created

    ✨ How can I help with this journey?

    Describe what it should do and I'll write the checkpoints and steps for you to review.

    This journey already has 2 checkpoints. Want to change those instead? Describe changes you wish to make.

    Summary
    Timeline
    Tickets
    Journey summary

    This journey generates a quote, binds the policy, issues the policy document, and asserts the document accurately reflects the bound premium and coverage within the 3-second quote-calculation threshold.

    NOT EXECUTED
    Assigned to:
    2 checkpoints
    17 steps
    0 extensions
    1 data table
    Journey plans

    There are no plans associated with this journey.

    Execution - Fair-Value Outcome and Policy Document Accuracy
    08 September 2026, 11:58 PM
    Journey 1: Quote to Bind to Issue: Consumer Motor Policy running…
    Wait 30 seconds for "Quotes"
    Click on "Quotes"
    Wait 18 seconds for button "New quote"
    Click on button "New quote"
    Write $applicantDOB in field "Date of birth"
    Summary
    Healed
    Journey side effects
    Tickets
    Execution in progress
    Execution is in progress…
    Execution - Fair-Value Outcome and Policy Document Accuracy
    08 September 2026, 11:58 PM
    Journey 1: Quote to Bind to Issue: Consumer Motor Policy 00:01:05
    Wait 30 seconds for "Quotes"
    Click on "Quotes"
    Wait 18 seconds for button "New quote" 🩹 healed
    Click on button "New quote" 🩹 healed
    Write $applicantDOB in field "Date of birth"
    Click on "Get quote"
    Assert performance measuring "Quote Calculation" is less than 3000
    Summary
    Healed
    Journey side effects
    Tickets
    Summary report
    1 of 1 executions in 00:01:05
    1 passed
    Launched by:
    AT
    ⌘ Environment: UAT
    Congratulations! All your
    journeys are doing great!