# UAT Scenario Checklist

## S0 Login and RBAC deny

- Preconditions:
  - Test users are available (`admin`, `supervisor`, `operator`, `analyst`)
- Steps:
  - Sign in through the gateway
  - Verify role-dependent access to main flows
  - Attempt a denied privileged action as `operator` (for example create a user)
- Expected:
  - Login succeeds
  - Allowed actions work for the selected role
  - Denied privileged action returns `403`
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - user:
  - screenshot/log:

## S1 Customer create and search

- Preconditions:
  - Operator UI or API access is available
- Steps:
  - Create a customer
  - Search the customer by name or phone
  - Open the customer card
- Expected:
  - Customer is created once
  - Search returns the same customer
  - Customer card data is consistent
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - customer_id:
  - screenshot/log:

## S2 Voice interaction lifecycle

- Preconditions:
  - Operator account is available
  - Voice queue is available
- Steps:
  - Create a voice interaction
  - Assign interaction to operator
  - Change status to `closed`
  - Open the interaction timeline
- Expected:
  - Status transitions are correct
  - Timeline contains creation, assignment, and close events
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - interaction_id:
  - screenshot/log:

## S3 Assignment and escalation to second line

- Preconditions:
  - Queue routing rule for escalation exists
- Steps:
  - Create or select an active interaction
  - Assign it to an operator
  - Escalate it to the target queue
- Expected:
  - Assignment is accepted
  - Queue is changed to the target queue
  - Status changes to `escalated`
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - interaction_id:
  - target_queue:

## S4 Routing and timeline verification

- Preconditions:
  - Queue rules are configured
  - At least one completed interaction exists
- Steps:
  - Execute a route request for the active queue
  - Verify assigned operator and SLA result
  - Re-open interaction timeline
- Expected:
  - Routing returns assignee and SLA
  - Timeline remains complete and chronological
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - queue_id:
  - route output:

## S5 Voice event intake

- Preconditions:
  - Voice adapter endpoint is reachable
- Steps:
  - Submit a voice event (`call.started`, `ivr.completed`, or `call.ended`)
  - Verify event is stored and retrievable
- Expected:
  - Voice event is accepted
  - Event ID is returned
  - Event can be listed afterwards
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - event_id:
  - call_id:

## S6 Telegram interaction lifecycle

- Preconditions:
  - Telegram adapter endpoint is reachable
- Steps:
  - Send webhook payload to Telegram adapter
  - Verify message registration
  - Verify interaction visibility in operator workflow
- Expected:
  - Message is accepted
  - Related interaction or adapter state is traceable
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - message_id:
  - screenshot/log:

## S7 KB usage in active handling

- Preconditions:
  - KB category/article created by analyst/supervisor
- Steps:
  - Search KB by keyword
  - Open article
  - Use article during interaction processing
- Expected:
  - Relevant article found
  - Operator confirms article usability
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - article_id:
  - keyword:

## S8 Supervisor realtime

- Preconditions:
  - At least two agent states are submitted
- Steps:
  - Set agent states (`READY`, `BUSY`)
  - Update queue metrics
  - Open realtime view/API
- Expected:
  - Agent totals and states are visible
  - Queue snapshot is visible
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - timestamp:
  - screenshot/log:

## S9 KPI report and export validation

- Preconditions:
  - KPI events are ingested for the test queue
- Steps:
  - Open `/reports/kpi`
  - Verify KPI payload
  - Open `/reports/export` and capture CSV output
- Expected:
  - KPI payload contains `SL`, `ASA`, `AHT`, `Abandon`, `FCR`
  - Export contains the expected header and queue rows
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - queue_id:
  - export snippet:

## O1 Backup and restore verification

- Preconditions:
  - Pilot environment contains test data
- Steps:
  - Run backup script
  - Restore into a clean target
  - Verify the restored data set
- Expected:
  - Archive is created
  - Restore succeeds
  - Core interaction data remains intact
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - archive:
  - restore check:

## O2 Load test baseline (`100 voice + 100 digital`)

- Preconditions:
  - Pilot environment is stable
- Steps:
  - Run `python scripts/load_test.py --base-url http://<gateway-host>:8080 --voice 100 --digital 100`
  - Record success and latency output
- Expected:
  - Load completes without blocking errors
  - MVP baseline target is met
- Result:
  - [ ] Pass
  - [ ] Fail
- Evidence:
  - command:
  - summary: