We're launching Rephrase — what's your biggest prompt frustration?

Hey everyone! 👋

I'm building Rephrase — a native macOS app that improves your AI prompts in one click, right where you're working.

The idea came from a simple frustration: every time I wanted a good result from ChatGPT, Claude, or Midjourney, I'd spend 5-10 minutes rewriting the same prompt. Copy to browser, tweak, copy back, repeat. It felt like the dumbest bottleneck in my workflow.

So I built a tool that sits in the background and works via a global hotkey (Cmd+Shift+I) — in any app. It auto-detects whether you're writing a text prompt, a code task, or a video description, and rewrites it using official best practices from Anthropic, OpenAI, Midjourney, and Sora docs.

Quick example:

• You type: "find info about how sleep affects productivity"

• Rephrase turns it into a structured research prompt with scope, evidence standards, and output format — in 2 seconds.

Before launching, I'd love to hear from this community:

What's your #1 frustration with prompting AI tools?

Is it the iteration cycle? Getting different quality across models? Not knowing the "right" syntax for image/video generators?

Would love to learn what you struggle with most — it directly shapes what we build next.

🔗 rephrase-it.com