How to use Codex to build flight-time widgets
Codex turned Figma travel designs into offline Fly Mode and time-zone widgets, with human privacy choices, regression review and TestFlight release preparation.
AddedTrip research, itinerary changes and booking details. Check live prices, rules and reservations with the provider before relying on an agent’s result.
51 use casesShowing 25–48
Preparing search…
Codex turned Figma travel designs into offline Fly Mode and time-zone widgets, with human privacy choices, regression review and TestFlight release preparation.
AddedDr. Eva-Maria Hempe tested Codex against an actual travel requirement: reaching Frankfurt by 10:30 on June 29. A comparison of flights, trains and driving needed to include the journeys to stations or airports and the final leg to the meeting location.
AddedYuta used a first overseas trip to Korea to try a travel workflow built around local files. Before leaving, Codex organized an AI brain containing an index, goals, current status, next actions, open questions, entities, outputs, logs, attachments and raw material. It also produced a PDF guide that could be consulted from a phone.
AddedAmanda returned from a twelve-day cruise through Amsterdam, Scotland and Iceland with a practical use for Codex: putting the shore plans somewhere she and her travelling companion could actually use them. Their research already existed in conversations, documents and a table of ports, dates and priorities.
AddedCodex repaired MatchDay’s conflicting airport memory and empty flight handoffs, using deletion, isolation and evidence gates before a human-authorized release.
AddedCodex used Finite’s live plan to add Paris nights within fixed-flight and budget limits; a person approved one revision with a receipt, without buying travel.
AddedTravel Weekly reporter Robert Silk describes having Perplexity Comet complete a flight booking in mid-October. He first used Perplexity's conversational search to explore flights, then selected the flight he wanted and asked Comet to book it through Southwest's website.
AddedOle Reissmann asked ChatGPT Agent and Perplexity Comet, in German, to find and book the cheapest train from Hamburg to Copenhagen for a September trip. He used the same practical objective to compare price and browser execution.
AddedMrityunjay Nigam compared Comet and OpenAI's Atlas on a flight-booking task from Bangalore. The brief asked for the cheapest direct one-way option to one of four specified destinations during the week of November 30, with coupon checks and booking-site comparisons before checkout.
AddedPat Simmons gave his Dot, László, a voice request for an October 15 to 18 trip to Montreal. He wanted a relatively inexpensive central stay, a cheap direct flight, an aisle seat and a carry on, plus activities that matched his interests.
AddedDenis Shiryaev reported an Alaska Airlines delay that put an Amsterdam connection at risk. Dots used its own computer to chat with Alaska, Delta and KLM concurrently, found another route and, according to the author, rebooked without an extra charge after he approved.
AddedKarl Weinmeister used Gemini while planning a ten-day anniversary trip with his wife to Positano, Naples and Rome, then wrote about the completed journey. The most useful work went beyond listing attractions: it examined the connections and constraints that could disrupt the trip.
AddedGemini Spark revised a Hershey family plan using connected context and grandparents’ availability, shared the document, and stopped before an Airbnb purchase.
AddedKeiichiro Sato completed a two-night, three-day trip to Guangzhou after repeated cancellations disrupted plans for a direct flight. He flew to Hong Kong instead, then used the high-speed train from West Kowloon to Guangzhou South.
AddedKuber Mehta gave Manus a family travel brief covering Singapore and Malaysia, with departure from New Delhi. The party included three adults and a child, and the requested trip split time across both countries in June with flexible dates.
AddedPon tested Manus and Claude Code on a detailed comparison of three Kyushu hot spring inns. The task covered twelve criteria per property, including rooms, private baths, meals, transfers, children and pets. Each cell needed a source, and unknown information could remain unknown.
AddedManus joined a supermarket stop, high-speed trains and toddler timing into a researched itinerary; Yingying Zhang supplied corrections and handled a CAPTCHA.
AddedManus compared UK-to-Las Vegas DevLearn travel packages, then published a navigable guide after Carolyn Shepherd’s review; the account establishes no booking.
AddedManus researched a father-and-son Japan trip from JFK, recovered after a high-load error, and produced Markdown and PDF plans; no purchases are established.
AddedGenevieve Horchler tested Manus against 11 hotel booking engines using the same room request and September dates. The task was to navigate hotel websites, select the stay and room, and see how far the agent could progress through different booking flows.
AddedEjaaz reported asking Muse for compensation after a seven-hour flight delay. The author says a $250 credit appeared in the Delta account within five minutes, that Muse found and rebooked a new flight, and that it replied to the support email. These are firsthand reported outcomes, with the attached screenshots clarifying which parts are visible.
AddedOpenClaw’s Alex proposed dinner options and booked after approval, then handled reported multileg flights; the source leaves flight payment checkpoints unclear.
AddedEeshita Pande had her OpenClaw agent, EeClaw, find and book Osteria Vibrato, a new Italian restaurant in London. The choice was grounded in dining preferences held in onFabric, including her interest in Italian, Indian and Japanese food.
AddedNick Larkins tried OpenClaw on a dinner reservation after hearing about another agent's restaurant call. His own test began with an ordinary online booking attempt, which failed because the restaurant showed no availability.
Added