Build a Next.js directory app in one shot
Poster had Manus build a Next.js directory app and then QA and fix bugs; the result was structurally promising but needed human correction.
Done withManus
- What they did
- The poster gave Manus a detailed prompt to build a simple Next.js and Tailwind directory app with milestones, seed script, search, admin dashboard, and Docker. Manus planned the work, checked in at each milestone, and finished in about 40 minutes with quick human replies, but when asked to start the dev server and then run its own QA and bug-fixing pass, it still could not get the app running.
- How it went
- The structure looked promising, but the delivered app had simple React runtime errors, incomplete UI, and functionality the poster judged only okay-ish. They spent 1,643 credits total, about $16.43, and ended up relying on GPT-4o to spot at least one obvious issue.
- Worth knowing
- The main catch was cost versus reliability: the initial build, bug-fix pass, and project archive together used 1,643 credits even though the app never reached a working state.
Try it yourself with Manus
Build a Next.js directory app for [topic, e.g. local coffee shops] with listings, search, category filters, and a detail page for each entry, seeded with [number] sample entries. When it's built, test it yourself, fix any bugs you find, and give me a list of what you tested and any known issues remaining. Done when the app runs and the main flows work. Ask me before connecting any paid services or deploying it.
Want help getting this working? Join the Vibe Coding Academy ↗ Run by Alex Finn, who also runs this site.
Source: Reddit · Undated
Related use cases
Compile journalist lists and search apartment listings
A reporter tested Manus on building a reporter list, searching NYC rentals and finding award candidates; results were thin at first, and a three-hour search yielded only three full profiles.
Generate slide decks from titles and a branded template
A reviewer had Manus find content for slide decks and produce a monthly marketing report in an uploaded PPTX template; editing tools were limited.
Produce competitor research reports and build a landing page
A team ran a multi-source competitor research report and a working landing page build through Manus, logging output and fixes needed.
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.
Diagnose Kubernetes alerts with MCP servers
An n8n agent triggered by Alertmanager queries Kubernetes, Grafana and other MCP servers and posts a diagnostic report with root-cause hypotheses in a Mattermost thread.
Scan every git commit for secrets and injection flaws
A free tool runs 11 Claude Code CLI agents on each commit to catch secrets, injection flaws, and bad migrations before code lands.
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.