Probed the AI’s hidden device capabilities with a self-scan, confirming HomeKit control, geofences,
The self-scan ran 2026-09-23 and confirmed capabilities present on the device that Meta's public documentation doesn't mention.
A random mix from the whole library. Shuffle for a new set.
The self-scan ran 2026-09-23 and confirmed capabilities present on the device that Meta's public documentation doesn't mention.
His home WiFi kept acting up, so he asked the AI to diagnose it. The AI ran a script that monitored everything for a week and pinned the problem on his Google Nest Wifi, and Reddit posts backed that up.
He told the AI to scan his home network for his Samsung Frame TV and replace the picture in its art mode. The AI found the TV and swapped the picture.
A user built a custom the AI skill that remote-starts their car, reports EV battery level and monitors charging status from any chat. They published the skill because nobody else had made one for it.
No app fit what his father needed, so the AI wrote a bare bones dialer in about an hour. It only shows accept and reject buttons, and he says it is perfect.
Days earlier he set up Pi-hole on a 50 euro mini PC and it removed ads across his whole home, with Tailscale so he could carry it in his pocket. In this second part, all built in one Sunday afternoon with AI’s help, he added an app that knows when he is home without asking for his location, his own VPN using his home IP, and DNS privacy so his internet provider cannot see every site he visits.
His LLM agent reviews his robot vacuum's images and either sends the vacuum out or suggests moving things that would foul it up; works great.
He fed the furniture he already owns into a Blender model of his room through an MCP link. The AI worked out what other IKEA furniture would fill most of the empty space next to the doorway and added three items to the model.
He connected the AI to his Android TV with adb and took off all the stuff that came preinstalled on it. He says the TV is now left exactly how he wants it.
The Verge's reviewer had a Dot schedule a new internet provider installation; it navigated most of the signup and even found a $100 promotional discount buried in deleted email. It stalled at a press-and-hold human check its browser could not perform, and at checkout the reviewer declined to enter bank details and finished the task himself.
ESPM professor Wilson Silva had his AI agent research ten auto repair shops, send them photos and videos of his car, request quotes and rank the options while he worked. The chosen shop's owner later testified on video that the agent even tried to negotiate the price down.
MindStudio tested an AI monitor that watches Autotrader listings and flags cars priced well under market. Scheduled checks deliver deal alerts to Telegram, and a continuity setting stops it re-alerting on the same listing.
The AI diagnosed the user's failing waterfall pump, selected an efficient replacement, designed a 3D-printable adapter for it, and separately saved $1,100 a year moving car insurance from AAA to Progressive.
An engineer gave the AI his house drawings, the original solar plans and a Powerwall expansion permit (mostly PDFs plus DXF files) after a contractor was slow to quote, and the bot produced a complete 17-page B-size permit drawing package in under five minutes, including a roof layout and one-line wiring diagrams drawn from scratch. He checked every line himself and planned to submit once a licensed engineer stamped the structural parts.
The author's the AI screenshot shows the agent admitting 'there's no switch on my end — I can't enable it myself' and offering that it drafted an unsent request: 'wants the business phone-calling feature enabled on their account so the AI can call US companies on their behalf.'
He had black mold in his home and no idea what to buy. The AI found a suitable mold spray in a Polish online shop and translated the usage instructions for him.
The open-source 'burner' project gives the AI a real spare Android phone (Android 11+). The agent uses the phone's apps exactly like a human would — opening them, tapping, typing, scrolling — and sends the user screenshots, extending its reach to marketplaces and shopping apps with no website or API.
Spent an hour with AI migrating a Home Assistant setup to Tailscale, updating sensors and Pi-hole, and putting it all in one documented git compose file.
In the same week as his treadmill buy, a SpaceX launch engineer says his AI got several quotes for new car tires, paid for them, and scheduled the install. No prices or shop names were shown.
A buyer ran an eight-bot the AI crew for his car purchase: over nine days a SCOUT bot watched 1,412 listings, a VALUER bot built a price curve, and a HAGGLE bot sent 11 messages to the dealer. The dealer came down $4,200.
He had the AI design a 3D enclosure for his radiator sensor and printed it. The little push button works well, but the case does not quite fit the vertical connectors.
The AI in the car, with Gmail and Home Assistant connectors, calculated car charging from home power metrics and calendar, and operated the garage door, lights and air conditioners from the car.
His young daughter lost the small cap from a water bottle she likes a lot. He asked the AI to design a replacement, printed it in 3D, and says it works perfectly.
He is a physician, not a coder. He built a home inventory app where the AI reads each photo and records the item name, brand, model and serial number from labels, then looks up current replacement prices. Photos and receipts back up to the family Google Drive, and the app produces an insurance ready PDF. The whole house cost about $6 in AI charges.
He runs Home Assistant on an old PC and used AI to make a menu bar app for his Mac. A keyboard shortcut now sets his lights, though he says it may not work as well for homes with lots of different smart devices.
A couple found their mailbox on the ground with the post rotted through. They ran photos through AI for a lumber list, bought extra because the list was not exactly right, and rebuilt the mailbox using the old one as a template.
He needed quotes for a seasonal tire swap, with the winter set already on rims, and did not want a day of phone tag. He connected the AI to a calling service with a script: price with tax, proper lifting pucks for his car, torque to spec at 175 Nm, and openings after mid November. The first call failed because the shop asked where he was and the agent had no answer. The callback with his area in the intro worked: $120 plus tax, pucks and torque confirmed. Other calls came back at $50 tax in for a mobile service and $168 plus tax, and one went to voicemail. Every call produced a full transcript. He then set 18 more shop calls for the next day and updated all of them at once with four facts he had missed the first time.
He is beta testing an AI agent, and it renewed his car registration for him. He says how much time it saved is debatable, but the renewal got done.
He remembered some smart plugs more than ten years old that use Bluetooth, and found Home Assistant could talk to them by pretending to be Apple HomeKit, but some functions were missing. He asked the AI to reverse engineer the plugs, and five minutes later they were fully revived with fully local control, with even more controls than HomeKit offered back then.
Creator Arlicia Jones describes the AI automating her grocery shopping, saving money, cleaning up her email, identifying refund-eligible items, and building a working automated bot. She promotes the service with a referral code (ND6YC6).
Every idea is a real post from someone who tried it, summarized in our words, with a link to the original. App names are swapped for plain words.