Rewrite AI-sounding prose so it reads as considered human writing — cut the tells (em-dash pile-ups, "it's not X, it's Y" antitheses, inflated vocabulary, uniform rhythm, empty uplift) while preserving every fact, claim, and the author's own voice. Use when the user says "make this not sound like AI", "de-AI this", "humanize/humanise this", "this reads like ChatGPT", "remove the AI tells", "it sounds robotic/generic", or shares a draft (post, email, essay, docs, cover letter) to sharpen into a more human voice. Ships a stdlib scanner that flags the mechanical tells.
日本語の概要は準備中です。原文の説明を表示しています。
promptmechanic/ai-toolbox☆ 142026年8月4日 更新
Critically review a resume/CV against ATS (applicant tracking system) parsing rules, recruiter skim patterns, and keyword alignment with a target job description. Use when the user asks to "review my resume/CV", "check my resume against this job ad", "will this pass ATS", or shares a resume file (.docx/.pdf/.md) for feedback. Provides scored, prioritised, critical feedback — not a rewrite.
日本語の概要は準備中です。原文の説明を表示しています。
promptmechanic/ai-toolbox☆ 142026年8月4日 更新
End-of-week reflection report. Mines your Claude Code transcripts from the last 7 days to produce a self-contained HTML file with (1) the topics/areas you explored, (2) distilled key learnings with live-searched further-reading links, and (3) a token/command/skill/ctx-stack usage summary. Use when the user says "week in review", "weekly review", "what did I learn this week", "my week", or runs /week-in-review.
日本語の概要は準備中です。原文の説明を表示しています。
promptmechanic/ai-toolbox☆ 142026年8月4日 更新