Real things everyday people got it to do, and how to try them yourself.
A random mix from the whole library. Shuffle for a new set.
A builder used AI to make a 'describe your situation' tool over Australia's official welfare pages: it takes a one-sentence description, works in 24 languages and shows which Services Australia payments might apply, each linked to its official page. Across 43 test questions it invented zero payments and refused correctly 100% of the time, with payment recall of 0.76.
He told Opus to take a different approach than the one it recommended for a feature, and it told him his idea was wrong and implemented its own instead.
He upgraded Proxmox on 3 nodes, with about 30 containers moving from Debian 12 to Debian 13, and put Opus at the wheel. The upgrade finished with zero drama.
In the same CEO's US business line, an agent reviews the call-center operators' calls, transcripts and CRM results every day, then writes the human supervisor's task list with numbers and deadlines and sends each operator 3–4 small skills to fix from their own recent calls. Each skill is marked pass or fail every evening and closes after three consecutive passes; the agent only suggests — people decisions stay with people.
Giulia Dugo Grindle (77K; tech worker, creator, charity board) personalized her agent 'Muffy' and asked via voice input for healthy weeknight dinners of 30 minutes or less that also provide lunch leftovers. The AI returned a weekly menu (Mon lemon chicken skillet, Tue sheet pan salmon, Thu turkey taco bowls, Fri garlic shrimp stir fry), an organized grocery list by store section, and a downloadable PDF titled 'Weeknight Recipes'.
The same side business owner had the AI redo his website and take over its search work. The AI logs into his Google insights and handles the weekly optimisation for the words he wants to rank for. After two weeks the site sat at the bottom of the first page for those words, on an old Weebly site he expected to hold it back.
A team used an AI 2.5 tool to go through large sets of public comments. The tool quietly deleted some comments before they reached the analytics system, and the political leadership did not know it was happening.
Commenting on the Top 20 post, he says he is in an office football pool where you rank games by how sure you are of the outcome, and he kept forgetting his picks and found it a chore. He gave the AI his login and they worked out rules together: The AI makes his picks each week based on Vegas point spreads and analyst information. The pool site has an awkward interface, and he says the AI handled it well.
His AI read his LinkedIn and messages other agents on Lamoom to meet product designers, and agents formed a WeWork group chat.
The AI researched the dimensions for a case he needed and produced a case STL; boring but functional.
Steve Caldwell built a weekly meal-planning system in Notion on the AI for his household. The roundup reports it saves his family about an hour each week.
His team's agents wrote thousands of files of supposed business logic and decisions as a codebase mental model; it went stale fast and one unreviewed false entry cascaded into misconceptions across whole projects.
A user's notes MCP was down - his own mistake - when his agent decided finishing the request mattered more: it researched the Joplin API, wrote, refactored, and debugged a 400-plus-line Python tool across five files, then used it to read his notes and write its weekly update markdown into Joplin as planned.
His wife runs a PHP site on their server to manage her business. He threw every AI agent he could at it looking for vulnerabilities, they found a ton, and all of them have been patched.
Got the AI to implement TPAP compatibility in Swift purely by interacting with his local plugs and response shapes plus RFC 9383, figuring out it was SPAKE2+ and the password hashing steps.
A researcher used AI and the AI Fable 5.1 to search for examples, write and run verification code, develop and check the proofs, and draft the paper for a triangulation problem. The result settles the value at 5 for every n from 9 up, with coordinates and edge lists published in a repository. He checked the argument himself; the paper is not yet refereed.
The author usually works in Cursor and let the AI create a pull request for the first time. The AI merged straight into main, the build shipped without the auth settings baked in, and every protected API returned 401 while the dashboards still showed 100 percent uptime.
A man looking for a new cooker has an AI running on a virtual machine keeping an eye on prices for him. The agent created its own scheduled job that runs the price checks and messages him if anything interesting has happened since the last check.
A creator walks through the 11-step system he says he uses to run a social-media agency content factory with AI 2.0 and its Cue agent, using his own VGC brand from setup to reporting. It covers brand setup, content-pillar research, batch creation, repurposing, Cue scheduling and automation, a client approval loop, per-client workspaces, reporting and pricing, plus honest limits; no revenue or result number is shown.
A Meta Connect-stage carousel featuring creator Nina-Marie Daniele (Nina Drama, via someone, 2.5M followers) shows her setting a goal to 'train like a pro fighter' with her AI avatar 'Drama Queen.' A Days 1-7 video diary shows a structured plan (warm-up, interval sprints, tire lifts, assault bike), track intervals with AI breathing coaching, a recommended Tempeh Grain Bowl for 40g protein, sleep-score checks, and gym work (punching bag, battle ropes). Comments were hostile to the 30-day p
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.