Build a bill-splitting app with browser testing
A reviewer had Replit Agent 3 build a working bill-splitting app in 8 minutes, with it testing the UI in-browser and flagging failures; it was slow and costs could add up.
Done withReplit Agent
- What they did
- The reviewer described a bill-splitting app to Replit's Agent 3 and skipped the option to generate a UI design first, going straight to building. The agent produced a working version, then ran browser-based tests that simulated user interactions. It flagged failures and explained how it planned to fix them.
- How it went
- A working app arrived in 8 minutes, with core functionality working and no obviously broken buttons or dead links. Most of the time went to testing and refinement, so it was slower than Lovable, which produces simpler apps.
- Worth knowing
- On the free plan, the reviewer used up the entire credit allocation in a single iteration, so heavy agent use could get expensive.
Try it yourself with Replit Agent
Build me a simple web app that [describe a small everyday tool, e.g. splits a restaurant bill among friends, including tax and tip]. When it's built, open it in the browser, test the main flows yourself, and fix any failures you find. Stop when the core features work end to end, and give me a short list of anything you couldn't fix.
Want help getting this working? Join the Vibe Coding Academy ↗ Run by Alex Finn, who also runs this site.
Source: superblocks.com · Undated
Related use cases
Build a searchable map of funded companies
A software engineer had Replit Agent build a searchable map of Wefunder brick-and-mortar companies in about 3 hours for about $22.
Build a Pokemon-style dog collecting app
A user had Replit Agent build a catch-em-all dog app whose rarity comes from scraped American Kennel Club regional population data.
Publish four Android apps to Google Play built with Replit
A user built four apps with Replit and published them to Google Play, with Replit writing the GitHub workflow YAML files for builds.
Build a C compiler with a team of parallel agents
Anthropic's engineering write-up on tasking Opus 4.6 agent teams with building a C compiler.
Run sandboxed Claude Code fleet sharing files via MCP
A developer runs a fleet of sandboxed Claude Code instances that share files through MCP tools on a virtual filesystem, with an orchestrator handling AWS storage and audit logs.
Build a local knowledge app for agent outputs with Codex
The author built Mianotes using four Codex agents across 30+ sessions to centralize scattered files agents produced, and the agents documented the product inside it.
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.