Why does the prompt demand specific examples?

Vague feedback like "make it clearer" is not actionable. The prompt forces the model to quote concrete parts of the work and explain exactly what to change, which is what makes feedback useful.

What does prioritizing by impact mean?

The prompt instructs the model to order points so the changes that matter most appear first, instead of a flat list where a typo sits next to a structural problem. That helps the recipient fix what counts.

Will it rewrite my work for me?

No. The prompt explicitly tells the model to coach rather than rewrite, so you learn and retain ownership. If you do want a rewrite, you can ask separately after reviewing the feedback.

Which tone should I choose?

Supportive suits early drafts and learners, neutral suits professional peer review, and direct suits experienced people who want candor without padding. The content of the feedback stays the same; only the framing changes.

Feedback Prompt Builder

Feedback prompt builder

Most AI feedback is too soft, too vague, or too generic to act on. This tool builds a prompt that forces the model to give the kind of feedback that actually helps: balanced, specific, prioritized by impact, and concretely actionable. You tune the subject, depth, tone, and format, then paste in your work. Everything is assembled locally in your browser.

How it works

You choose the subject being reviewed — writing, code, design, a presentation, work performance, or an essay — and set the depth, from a quick three-point pass to a deep section-by-section critique. You pick a tone (supportive, neutral, or direct) and an output format. The builder assembles a prompt with hard rules: acknowledge what works first, cite a concrete example for every point, order points by impact, attach an actionable suggestion to each improvement, and coach rather than rewrite.

Tips and notes

Match depth to the stage of the work — a quick pass is right for an early draft, while a deep critique suits something close to final. The “coach, don’t rewrite” rule keeps you in the driver’s seat; if you genuinely want a rewrite, ask for that separately after you have absorbed the feedback. The direct tone is the one most people under-use: experienced recipients usually prefer candor to padding, and the prompt keeps it constructive rather than harsh. After reviewing, you can feed the top one or two improvements back into the model and ask it to show one worked example of the fix.