Launching today
Verbito
Generate expert AI prompts — then score them 0–100
8 followers
Generate expert AI prompts — then score them 0–100
8 followers
Verbito turns a rough goal into a production-quality AI prompt in seconds, then scores it 0-100 across six dimensions (clarity, context, specificity, constraints, format, reusability) so you see exactly why it's good. Includes Prompt Doctor (fix and score existing prompts), 17 free niche tools, a prompt library, and a prompt-engineering course. Fully bilingual EN/AR with true RTL. Free tier: 2 generations/day, no card required.
Hey Product Hunt 👋
I'm Abdalla, solo founder of Verbito. I built this because I kept noticing the same pattern in my own work: the AI wasn't the bottleneck, my prompts were. I'd re-prompt 4-5 times to get something usable, then eventually write down "the good version" and reuse it- basically manual prompt engineering with extra steps.
Verbito automates that: describe what you want, get an expert prompt back with a score and a breakdown of why it scored that way, so you're not just getting an output, you're learning the pattern.
A few things I'd genuinely love your take on:
1. Does the 0-100 score feel useful, or is it a gimmick?
2. Which of the 17 niche tools would you actually use weekly vs. just try once?
3. If you prompt in Arabic; is the RTL/bilingual experience actually good?
I'll be in the comments all day. Thanks for taking a look 🙏
Scored my messy prompt 72 and broke down exactly why in a way that finally clicked for me. The bilingual RTL support is a nice touch too, feels like it was actually built for both languages rather than bolted on.
Love the scoring breakdown, super helpful for seeing where a prompt falls apart. One thing that would push it further for me is letting you compare two prompts side by side with their scores highlighted, so you can A/B test rewrites and actually feel the difference instead of guessing which version reads better.
The scoring breakdown is genuinely useful. I pasted in a prompt I thought was solid and clarity came back at 42, which stung but pointed me straight to the issue.