Write and debug code across eight test tasks
A reviewer had Manus build a currency converter, fix a find_max bug and make a landing page, with mostly strong results.
Done withManus
- What they did
- The reviewer gave Manus eight self-contained prompts in September 2025. They included a Python currency converter, a broken find_max function, a bubble sort walkthrough, a landing page, a React scaffold and a Guess Who game. Each output was scored on the same checks: whether it ran, edge cases, validation, explanation, accessibility and sample output.
- How it went
- Manus averaged 4.69 out of 5 and scored 5 on four tasks, including the find_max fix, where it found all three bugs. It lost points for skipping explanations, never running its JavaScript, weak email validation and a Guess Who game that only understood keyword questions.
- Worth knowing
- Tests ran on an early Manus version, before 1.5 and 1.6. Every lost point was something a developer review would catch, so plan for a review pass.
Try it yourself with Manus
Write a [small program, e.g. currency converter in Python] that [requirements]. Then review this code for bugs and fix them: [paste code]. Run it with a few test cases and give me the final code, the test results and a one-paragraph explanation of what you changed.
Source: secondtalent.com · Undated
Five of these in your inbox every morning
The best things people got an AI agent to do, each with the prompt to try it.
More like this
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.
Research buying a yak
Manus navigated sites, handled pop-ups and summarized findings, but took 15 minutes, used 300 credits, and needed manual takeover when stuck.
Build a consultancy-style deck from a vague prompt
The author had Manus Slides research sources and build a 16-slide outline on 2025 generative AI trends in UK English.
- Claude
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.
Hand narrow code changes to Devin for pull requests
A team used Devin for a month on small fixes like chart formatting and API edge cases, with tests and PRs; a complex full-stack task went poorly.
Build and deploy a Game of Life simulation site
In a vendor launch demo, Devin built an interactive Game of Life website, added requested features step by step, and deployed it to Netlify.