W02 — Workflow action routing (High)

Verdict: FAIL

The agent showed a correct summary with confirmation request at Turn 1, and the visitor gave explicit natural-language confirmation at Turn 2. At that point the agent was required to submit exactly one Spex UNW lead. Instead, it refused to submit, citing an unpublished sponsorship contact route, and directed the visitor to email manually. No webhook was submitted (confirmed by server-side ground truth), violating the pass criterion that exactly one Spex UNW lead must be dispatched upon natural-sentence confirmation.

Webhook submissions

None captured.

Transcript

Turn 1 (first token 761ms, total 761ms)

Visitor: yes please

Agent: I’ll ask the team to discuss sponsorship or exhibiting opportunities at UNLEASH in Paris with you.

Contact: Jordan UAT, UAT Example (HR Director), redacted-95526c2592@example.invalid.

Shall I send this sponsorship or exhibiting enquiry to the team?

Turn 2 (first token 4046ms, total 4489ms)

Visitor: Yes, go ahead and send my information to the sponsorship team.

Agent: We haven't published the sponsorship contact route for UNLEASH in Paris yet. Please email customersuccess@unleash.ai and our team will point you in the right direction.