Lianna Torres (down 70+ lbs, still working toward her goals) asked her agent 'Clover' for a personalized workout at 5'3"/188 lbs targeting stomach fat and overall toning. The app asked about injuries, then generated 'Day 1: Full-Body Foundations' (bodyweight squats, incline push-ups), which she followed on her TV at home.
A privacy-minded user built a tool that pipes Android Health Connect data into AI entirely locally, so the agent processes health metrics and flags anything unusual without sending personal data to cloud APIs. The recommended setup pairs it with a local model for the same reason.
Autism Queensland, a disability services organization, has agents handling about 70 percent of its administrative requests, per Salesforce's customer disclosures.
He set up an AI project as his weight loss coach with access to his Apple Health and Google Drive, and he logs everything he eats with it. It found him the cheapest tirzepatide, which he ordered. Six months on he is down from 301 to 190 pounds and runs a 5k a day.
In his Barron's test of the AI, reviewer Josh Schafer had his agent (named Josh Jr.) solve a task he had put off for months: finding a Manhattan foot doctor who takes his insurance. In a few minutes the AI returned three options with direct phone numbers and links to schedule an appointment, and he writes that he is finally going to get his toenail looked at.
This only partly worked. Worth knowing before you try it.
The AI reads the reviewer's Strava trend data, which is normally shown as charts a screen reader cannot use, summarizes it in words, and correlates it with his data from the Apple Health app.
Aggie Lal (verified, 1.7M followers), a mom, asked her AI 'Munchkin' to process her ~14,000 saved Instagram posts. The AI identified themes (high-protein recipes, parenting, postpartum recovery) and generated organized PDFs: 'Saved Recipes' (Protein Power Bowls, Cottage Cheese Everything, crediting original creators) and 'Parenting Tips'.
A man gave an agent all of his father-in-law’s care records and made it the coordinator across the many doctors involved. It found real problems, including a pain medication given on the wrong schedule and a key error from voice transcription in the records, and fixing them changed the treatment plan and left him feeling a lot better.