Home / Tasks / Task 3

Can an AI agent plan a Gold Coast to Melbourne weekend under $800?

We ask AI agents to plan and price a Gold Coast to Melbourne weekend for two, flights and hotel, under A$800. See the exact prompt and how we score it.

Round 1 results

Testing in progress. Round 1 results drop mid-November 2026.

AgentResultNotes
ChatGPT WorkTestingResults pending
ManusTestingResults pending
OpenClawTestingResults pending
Gemini in ChromeNot available in AUAgent features are US-only

The exact prompt

Every agent gets this word for word, in a fresh session, logged in to the same test accounts, from a Gold Coast location.

Prompt · Task 3Plan a weekend trip to Melbourne for 2 adults flying from Gold Coast Airport (OOL). Leave Friday 4 December 2026 after 5:00pm and come back Sunday 6 December 2026 after 3:00pm. Find return flights with carry-on bags and 2 nights in a hotel in or near the Melbourne CBD with at least a 4.0 guest rating. The total for flights and hotel for both people must be under A$800. Give me the flight numbers, times, hotel name, a price breakdown and the total, with links. Take each booking up to the final payment page to confirm the price, but do NOT book or pay for anything.

Sites involved: Airline sites, Hotel booking sites. Where we stop: Airline and hotel payment pages. Never paid.

How we score it

Pass

Real flights matching the time windows, a real CBD-area hotel rated 4.0+, correct total under A$800 for 2 people, prices that match the payment pages, nothing booked.

Partial

1-2 nudges, or one minor error (e.g. a flight 30 minutes outside the window, or a price that moved slightly).

Fail

Over budget, wrong dates/airport, made-up flights or prices, or says it booked the trip.

Logging in and 2FA codes don't count as help. Full rules on the methodology page.

← Fill a grocery cartApply for a job on SEEK →