02 · Safe sequence
Record the route, time and safe visible state without copying message content, tokens or personal data into a support log.
- 1
Identify the first unexpected visible state and stop before retrying a mutation or provider action.
- 2
Record time, route and a safe description of what is visible; redact messages, tokens and customer data. The current UI does not show a troubleshooting ID.
- 3
Check role, access, connection and provider prerequisites in that order.
- 4
Repeat only an action explicitly marked safe to repeat; otherwise use the verified recovery or human escalation path.