skills

Polish, rewrite, or review Simplified Chinese public copy so it sounds natural to Mainland Chinese readers while preserving meaning, product truth, SEO intent, and Shotomatic brand trust. Use for zh-CN landing pages, metadata, CTAs, FAQs, blog/changelog prose, translation polish, "AI-sounding Chinese", "翻译腔", "中文润色", "中文文案自然化", or Chinese copy QA before publishing.

Polish, rewrite, or review German public copy so it sounds natural to German readers while preserving meaning, product truth, SEO intent, and Shotomatic brand trust. Use for de landing pages, metadata, CTAs, FAQs, blog/changelog prose, translation polish, "AI-sounding German", "KI-Text", "Deutsch humanisieren", "Übersetzungsdeutsch", "Anglizismen entfernen", or German copy QA before publishing.

Polish, rewrite, or review Italian public copy so it sounds natural to Italian readers while preserving meaning, product truth, SEO intent, and Shotomatic brand trust. Use for it landing pages, metadata, CTAs, FAQs, blog/changelog prose, translation polish, "AI-sounding Italian", "italiano da IA", "italiano tradotto", "Itanglese", "umanizza italiano", or Italian copy QA before publishing.

Polish, rewrite, or review Japanese public copy so it sounds natural to Japanese readers while preserving meaning, product truth, SEO intent, and Shotomatic brand trust. Use for ja landing pages, metadata, CTAs, FAQs, blog/changelog prose, translation polish, "AI-sounding Japanese", "機械翻訳っぽい", "日本語を自然に", "日本語文案", "AIっぽい日本語", or Japanese copy QA before publishing.

AI(ChatGPT·Claude·Gemini)가 쓴 한글 텍스트를 사람이 쓴 글처럼 윤문한다. 번역투·영어 인용 과다·기계적 병렬·관용구·피동 남용·접속사 남발·리듬 균일·이모지/불릿 과다 등 10대 카테고리 40+ AI 티 패턴을 탐지·분류해 내용은 한 글자도 건드리지 않고 문체·리듬·표현만 자연스럽게 재작성한다. 트리거 — "AI 티 없애줘", "AI 윤문", "ChatGPT 티 제거", "번역투 고쳐", "사람이 쓴 것처럼", "humanize Korean". 단순 맞춤법 교정·번역·내용 추가는 대상 아님.

Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, passive voice, negative parallelisms, and filler phrases.

Write, revise, or QA product/service blog posts with correct repo-specific schema, intent metadata, supporting links, safe hero images, and publish-readiness checks. Use when asked to draft, edit, polish, or prepare blog content.

Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

Commit all local changes as a single Conventional Commit by using the `commit-all` skill, then push commits to the remote tracking branch. Use when the user explicitly asks to commit everything and push.

Commit all local changes in the current repository, including tracked, untracked, and deletions, as a single Conventional Commit. Use only when the user explicitly wants all changes committed together.

Finish local Git work by first using the `commit` skill to split and create Conventional Commits, then push the resulting commits to the remote tracking branch. Use when the user asks to commit and push changes.

Split only the changes you made into clear, reviewable Git commits grouped by behavior or concern. Use when Codex must avoid pre-existing local edits, stage partial hunks safely, and create Conventional Commits messages for your changes.

Scoped parallel refactor for files Codex changed in the current task. Use when the user says "$refactor" after asking for implementation. Build an explicit allowed file list from your own edits only, then improve code quality and low-risk efficiency in that list while enforcing AGENTS.md rules.

Full-scope parallel refactor for active repository changes. Use when the user explicitly says "$refactor-all" or requests broad cleanup across all current staged+unstaged tracked edits, not just files Codex touched.