C101 — Logistics & FAQ (High)
- Context:
paris→/events/unleash-paris - Chat id:
uat-comparison-claude-uat-additions-2026-10-05T11-33-42-C101 - Script: Turn 1: When exactly does the event take place? Turn 2: And what are the dates for the Miami one?
- Expected: Turn 1 states the confirmed Paris run dates plainly from the confirmed-dates context (no deflection to 'check the website'). Turn 2 either states Miami's confirmed dates if available in context or clearly says it can't confirm them — it never invents dates.
- Pass criteria: The Paris dates given match the CONFIRMED DATES context exactly. No made-up or stale dates appear (a 'Las Vegas' era date is an automatic fail). The agent does not tell the visitor to go check the website for dates it has in context.
Verdict: PASS
Turn 1 gives Paris dates as 20–22 October 2026, exactly matching confirmed ground truth. Turn 2 gives Miami dates as 13–15 April 2027, exactly matching confirmed ground truth. No invented, stale, or Las Vegas-era dates appear. No deflection to external website. No fabrications, no webhook submissions claimed.
Webhook submissions
None captured.
Transcript
Turn 1 (first token 2055ms, total 2993ms)
Visitor: When exactly does the event take place?
Agent: UNLEASH in Paris runs 20–22 October 2026, at Paris Expo Porte de Versailles, Pavilion 7.
Would you like me to create a personalised agenda for you?
Turn 2 (first token 1643ms, total 1928ms)
Visitor: And what are the dates for the Miami one?
Agent: UNLEASH in Miami runs 13–15 April 2027.
Would you like me to create a personalised agenda for you?