Hands-on prompt engineering and LLM evaluation. Built around a diagnose-then-ship loop: forensic transcript critique ▎ that names the exact failure mode, paired with a fixer that produces complete, deployment-ready rewrites — not ▎ vague advice.
Paste a draft and, in about two minutes, three independent reviewers score it and show exactly what to fix — before you publish. Each reviewer scores on 4 axes (up to 20 points total): readability, originality, engagement, and structure. You get a prioritized fix list with specific lines to change and why — not vague feedback. Works for blog posts, articles, newsletters, and any text you want a second opinion on. Runs in your browser, nothing to install. First 3 runs free.
Takes any piece of writing — an email, a Slack message, a report, a doc, a cover letter — and tightens it up: cuts filler, kills unnecessary hedging ("I just think," "sort of," "kind of"), removes throat-clearing openers, swaps vague words for concrete ones, varies sentence rhythm, and nudges passive voice to active where it's just habit rather than a deliberate choice.
Paste any Japanese text and instantly detect AI-generated phrasing. Scores your writing 0-100 for "AI-ness," highlights every detected pattern with a specific fix suggestion, and auto-generates a cleaned version. Scans 37 patterns across 6 categories: stock phrases, over-emphasis, syntax patterns, too-polished words, sentence-ending habits, and AI cliches. 100% offline, rule-based, no API. Single HTML file that works forever.
Helps users turn rough prompts into clear, specific, reusable prompts in different LLMs through concise critique.
Turn your pasted research proposal into clear, natural academic prose while preserving your real evidence, methods, limits, and required claims.