14 real, live products built with Cursor, picked from our catalogue: why each one matters and what you can copy.
Cursor is an AI code editor: a version of VS Code with chat, autocomplete and agents built in, so you can ask for a feature in plain English and watch it edit your files. MissumAI lists 36 live products whose own sources say Cursor was part of the build. These are the 14 we think teach the most, grouped by what you can learn from them.
How we picked: every product is live, has a first-hand account of Cursor being used (a founder interview, podcast or the founder's own post), and has something worth copying: revenue anyone can check, a very fast build, or a builder who started from zero. We left out entries where Cursor was a short stop on the way to another tool, and apps we already covered in the best apps built with Claude Code (Snag, ToneAdapt and Payout use both). Each card shows the product's current rank on the main leaderboard, which is rebuilt every day, and every fact links to its source.
Post your content to every social platform at once, from one dashboard.
Jack Friks says that less than two years before his podcast interview he didn't really know how to code, beyond basic HTML and CSS.[2]
“Cursor helps me a ton.”
Jack Friks on The Bootstrapped Founder[2]He made Post Bridge for himself because the other posting tools were too expensive and had too many features, so he kept it as simple as possible. That turned out to be what other people wanted too.[2] TrustMRR now shows it close to $50,000 in monthly recurring revenue.[1]
All-in-one AI local SEO software that helps agencies rank clients on Google Maps and local search.
SEO agency owner Jacky Chou set himself a public goal: download Cursor for the first time and build a $10K MRR SaaS in 30 days. He live-streamed every day and posted a YouTube video every day, and three weeks in, two days after launch, LocalRank was at $2K MRR.[4] TrustMRR now shows over $19,000 in MRR.[3]
Generates SEO-optimised WordPress sites from a short intake form, for agencies to resell.
Bennett Black wrote that he vibe coded the tool with Cursor, Claude, n8n and Supabase, and that ten weeks later it made $1.6K a month.[6] His pitch is a story about a customer: a user built a roofing company's site in five minutes and sold it for $99 a month.[6] TrustMRR now shows over $14,000 in MRR from 411 active subscriptions.[5]
AI speech-to-text that works in Gmail, Docs, Slack and thousands of other apps.
Kourosh built a speech-to-text Chrome extension as a fun tool for his dad, going from idea to a very basic MVP in two weeks. When people downloaded it without any marketing, he took it on as a side project. After that he discovered Cursor, which he says made him about 10x more productive, and built the Windows app in about three months.[8] Today he pays for both Cursor and Codex, and uses Cursor when he wants Anthropic's Opus models.[8]
A startup leaderboard where every revenue number is verified straight from Stripe.
Marc Lou saw Pieter Levels post about fake MRR screenshots, opened Cursor the next morning and asked it to build a place where founders verify their revenue with a read-only Stripe key. Six hours later he had a leaderboard, and he launched it on October 30, 2025, two days after that post.[9] He had no plan to make money at first and added a few ad spots on the side.[9] TrustMRR's figures come from its own listing on its own platform, so we label them self-reported.
Self-reported$37,636 in the last 30 days and $404,213 all-time, per its own Stripe-connected TrustMRR listing (self-listed on the founder's platform)source
An iPhone scanner that identifies Pokémon cards, shows their market value and tracks your collection.
Kyle Fowler is a self-taught iOS developer. His first app, the sports-card app Cardstock, took about six months from idea to completion. For Scanémon he tried having Cursor's agent generate basically everything, says it did insanely well, and made the MVP in a day.[10] The idea came from the same place as the first one: his own collecting hobby.[10]
Amazon product research for private-label sellers: sales history, trends, keywords and suppliers in one report.
Hassam ran two Amazon brands next to a corporate job and had no audience. So he wrote to LegacyX, an Amazon seller coaching program he had bought two years earlier, offered to build them a better product research tool, and said "give me 48 hours".[11] He spent the first four hours mapping their systems and workflows, hours five to twelve building the core features with Cursor, and the next stretch testing and fixing bugs. He traded equity for that distribution and says he was at about $21.8K MRR 90 days after launching with the partner.[11]
A Christian app blocker that locks distracting apps until you pray.
Mau says he "literally just" used Cursor and ChatGPT, with Claude Sonnet as his daily model inside Cursor. Coding took three days; getting through App Store review took over a week.[12] For onboarding he laid out the best onboarding screens from apps he liked and from competitors in one Canva file, and picked from those.[12]
Self-reportedOver $150K a month at its peak, including Android, before Apple removed it from the App Store, per co-founder Mau Baron (Oct 2026); earlier $21K in sales in December 2025 and about 58,000 downloads. Current revenue not disclosedsource
A Hebrew-language AI calorie tracker for iPhone, built for the Israeli market.
Tomer says he isn't a developer. Cursor had just come out, he wanted to see if you could really build software by prompting, and two weeks into the experiment he had a working version; about two months after the first line of code it was on the App Store.[13] His point is that the build is the easy part: he says the real difference is brand and distribution, and that getting users for half what competitors pay lets him outspend them.[13]
Research and replicate viral YouTube and short-form video formats, with built-in AI video generation.
Sam Adeniyi, a university student, started by copying ChatGPT's code into Notepad, then VS Code. A friend told him to use Cursor; he says he first thought it meant a mouse cursor, and he has used it ever since, now on the $200 a month plan.[14] His first 400 users came from Discord: he joined voice chats in a Discord server where people kept asking how to find a niche, and silently shared his screen while using his own tool until people asked what it was, then posted a waitlist.[14]
Dashboard shownOver £10,000 (about $13K-14K) in revenue in the previous four weeks and 10,000+ users about six months after launch, per the founder (Stripe dashboard shown)source
A weight-cutting app that gives wrestlers and combat-sport athletes a science-based plan to make weight.
Ethan, 19, left university to learn development on his own. He says discovering Cursor made building much faster, which let him spend his time learning sales and marketing instead.[15] He built the MVP in about a month with Cursor and ChatGPT and gave it to his wrestling club to test. When the first idea (coaches send plans to athletes) had too much friction, he redesigned it so the app writes the plan.[15]
A breathwork app with visual animations and frequency tones for calm, focus and sleep.
The Sweeney brothers did it the other way round: content first. A tweet about a breathing technique went viral, and that became the app idea.[16]
“I didn't actually write a single line of code in this entire code base”
Jack Sweeney on Starter Story[16]They built the MVP in Cursor and launched it in February 2025, and now use Claude Code and Codex as their coding agents.[16]
An offline idle RPG for mobile from a one-person studio.
Ken Law spent about 12 years in big tech, most recently as an engineering manager at Meta, and hadn't written a line of code in over four years when he decided to see if he could build an app with AI.[17] His stack is short: Cursor as the editor, ChatGPT for brainstorming and icon touch-ups, React Native and Firebase. One Reddit post about the game, which got mostly critical replies, took it from about $1,000 to about $10,000 a month, and he left Meta that month.[17]
A free multiplayer flight game in the browser, paid for by in-game ads.
Pieter Levels built a browser flight game in public and sold in-game planes, blimps and UFOs to sponsors as subscriptions. He says he still mostly used Cursor with Claude 3.5.[18]
“Without AI it would have taken me 10-100x more time”
Pieter Levels on X[18]In the same post he gives the warning every list like this needs: AI doesn't care about security holes like XSS unless you explicitly ask it to check.[18]
Cursor is the editor; the model inside changes. Mau runs Claude Sonnet in Cursor,[12] Kourosh goes to Cursor for Opus models,[8] and Levels used Claude 3.5.[18] Several builders pair it with something else: Codex for Kourosh and Mau,[8][12] and the Sweeneys moved on to Claude Code and Codex after the MVP.[16]
They build in a niche they already know. Amazon selling for Hassam,[11] wrestling weight cuts for Ethan,[15] card collecting for Kyle,[10] local SEO for Jacky Chou.[4]
Distribution was planned, not hoped for. A partner's audience,[11] daily live streams,[4] Discord voice chats,[14] a Reddit post,[17] a viral tweet before any code.[16]Cal Buddy's founder says it plainly: the brand and distribution are the real difference.[13]
Lessons for your own build
01
Build the demo for someone who already has the audience
LaunchFast's 48-hour build was a pitch to a coaching program with thousands of sellers, not a launch to nobody.[11]
02
Design first, then let Cursor match it
Ethan designs in Figma and tells Cursor to build the front end to match before he starts the back end;[15] Mau picked his onboarding from a board of the best screens he could find.[12]
03
Ask for the security check
Levels had to keep fixing XSS holes and says you have to ask the AI to check your code.[18]
04
Spread the work across tools
Mau uses Cursor every day, Codex CLI for harder tasks and ChatGPT for questions, partly so he wasn't burning money on tokens.[12]
05
Tell it the scale you're aiming for
Sam's advice: tell Cursor to build as if you'll have 100,000 users, not just yourself, and it changes how it approaches the code.[14]
