# Built ProReview with Cursor — review drills for AI-generated engineering output

**URL:** <https://forum.cursor.com/t/built-proreview-with-cursor-review-drills-for-ai-generated-engineering-output/161510>\
**Category:** Built with Cursor\
**Created:** [May 25, 2026, 10:16am UTC](https://forum.cursor.com/t/built-proreview-with-cursor-review-drills-for-ai-generated-engineering-output/161510 "2026-05-25T10:16:27Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![SHaaD94](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/shaad94/32/109608_2.png) [@SHaaD94](https://forum.cursor.com/u/SHaaD94)\
**Post date:** [May 25, 2026, 10:16am UTC](https://forum.cursor.com/t/built-proreview-with-cursor-review-drills-for-ai-generated-engineering-output/161510/1 "2026-05-25T10:16:27Z")

</div>

Hey folks, I built a small thing called ProReview, mostly with Cursor.

It’s a set of short drills where you review AI-generated engineering output before trusting it — shell scripts, kubectl commands, infra changes, prod-debug suggestions, etc.

The funny part is that I built it with AI tools, but the whole point is: don’t blindly trust AI tools.

Here is an example:

 ![image](https://us1.discourse-cdn.com/cursor1/original/3X/9/a/9aa2296c607b610db632e354c2cb454d1e53381f.png)

An interactive version:

> **[ProReview - Catch AI Before It Wrecks Production](https://www.proreview.dev/challenges/shell-parallel-worker-restart)**
>
> Practice reviewing AI-generated code, infrastructure, SQL, CI/CD, and security artifacts before they ship. Blind judgment, segment-level scoring, and expert reveals.

Would you use something like this?