Stop copy-pasting prompts from a doc. Research Writing Assistant routes your academic writing task to the right expert prompt automatically — translation (ZH↔EN, LaTeX & Word), polishing, compression, expansion, de-AI, figure generation, caption writing, experiment analysis, mock review, and chart recommendation. Built from workflows used at MSRA, ByteDance Seed, and top CS PhD programs.
逐句告訴你「這篇貼文為什麼沒人看」，並用四條鐵律改寫而不會把你的語氣改成 AI 腔。第一條就是「用刪的不用加的」——改完通常比原文短。市面上的社群 AI 都在把文案改得更豐富：堆形容詞、加 emoji、把口語訂正成書面語，結果所有人的貼文都長一樣。這支反過來做，而且絕不發明數據或客戶案例。
I’ve created and shaped social content with nearly 200M views across YouTube, TikTok, and Instagram, and the pattern is clear: people do not connect with text that sounds polished but empty. They respond to voice, opinion, rhythm, and specific details. This Skill applies that editorial judgment to AI drafts by spotting obvious AI tells, repeated language habits, and stiff structure, then rewriting with stronger voice while keeping the original meaning intact.
I run an AI-focused Medium blog and built this pipeline from real posts I kept rewriting by hand. It does multi-source research with a checkpoint you approve, an outline checkpoint before drafting, a mandatory humanizer pass (kills em-dashes and AI-tell phrasing), a readability/SEO quality check, plagiarism-prep chunking, and a publish-ready export bundle. You control the pace: allow once, allow always, edit, or send it back for a redo at every step.
Turn your pasted research proposal into clear, natural academic prose while preserving your real evidence, methods, limits, and required claims.
Turn your research into bilingual social drafts with traceable evidence. Includes a Python source-reference checker and reusable writing workflow. Requires your own AI client for writing; no model credits, web search, or auto-posting included. 將研究整理成中英文社群草稿與來源對照。包含 Python 引用檢查器與寫作流程；寫作使用買家自己的 AI 客戶端，不含模型額度、網路搜尋或自動發布。
AIの下書きが「AIっぽい」まま公開していませんか？本セットは、あなたの文体ルールを一度登録すれば、①下書きがあなたの声で返ってくる文体合わせ（my-tone）②読者レンズで見出し・1行目を4タイプ×5案＝20案出し、上位3つを理由つきで選ぶ（hook-maker）③あなたが決めた「出さないことリスト」を最優先に、危ない箇所は勝手に直さず引用して差し戻す公開前チェック（self-review）のSkill3本で、書く工程まるごとを整えます。穴埋め式｛｛ ｝｝＝あなたの情報は手元で埋める設計。導入はフォルダに置くだけの3ステップ・README付き。※同店のAI Writing Style Checkerは「書いた後の検査ツール」、本品は「書く工程まるごと」の設定型3本セットです。
This agent helps Python teams using the Anthropic or OpenAI SDK safely cut LLM spend without risking regressions. With zero code changes, it maps call sites and runs your test scenarios to rank token waste from missed prompt caching or heavy model tiers. It tests fixes in isolated git worktrees, enforcing a strict "do-no-harm" gate that reverts any patch failing to preserve behavior. It delivers a clear savings audit that converts into verified, behavior-preserving code patches upon opt-in.
Compile how you actually decide into one paste-in a file any AI can run. Proven with your own task, two drafts side by side. File always stays yours.