Writing & Editing

Explain WHY a repository is built the way it is — every important library and its alternatives (e.g. shadcn/ui vs MUI, Zustand vs Redux, Next.js vs Vite), the programming concepts it relies on (debounce, memoization, custom hooks, SSR…) anchored to real files, the main data flows, and interview questions with answers. Produces an interactive HTML report plus a draw.io architecture diagram. Use when the user wants to understand a repo, prepare to explain a project in an interview, asks "why did w

Use when fixing a bug, regression or failing test (including functional UI bugs); when changing anything that judges correctness (tests, validators, detectors, thresholds, CI), money or numeric formulas, migrations or auth; when a test turned green after an edit, is flaky, or you are tempted to edit, skip or weaken it; when a test contradicts the spec; when retrying a failed attempt; when delegating code changes to a subagent or accepting its report and numbers; when deploying; when a check bloc

Debrief a Claude Code session or a stretch of work on how the collaboration went, not on what it produced. Use it whenever the user asks for a debrief, retrospective, post-mortem, lessons learned, "what did we learn", "how can I work better with you", or wraps up a piece of work; also at the end of a long session before it is compacted. Produces a report with cited moments, coaching for the person writing the prompts, process proposals with a destination, and a usage check, and keeps a history s

Run a hallway usability test of a tool with a fresh sub-agent that sees only the requirements and user-facing documentation, builds its own test cases, writes expectations before running, asks questions, and reports bugs, surprises and gaps; optionally judging a real terminal by eye through screenshots. Use it whenever the user wants a tool, command, CLI, notebook workflow or page tried out by a critical first user, asks for a play test, usability test, "let someone try this", "test it as a new

Draft documents for a New Jersey business-law practice, for attorney review — engagement letters (RPC 1.5, entity-as-client), demand and cease-and-desist letters, NJ LLC and corporate consents/resolutions, operating and buy-sell provisions, asset-purchase clauses (bulk-sale/C-9600), NDAs and commercial agreements with NJ-compliant arbitration language, NJ Superior Court pleadings and motions with captions and required certifications, settlement agreements, client advice letters, fee pre-action notices, and plain-English client-education handouts. Use this skill whenever an attorney, paralegal, or firm staff member asks to draft, write, prepare, or fill in any legal document or letter, adapt a firm template, turn facts into a pleading or memo, or produce a first draft "for my review" — even if they only name the document ("I need a demand letter for…", "write up a member consent").

Crea presentaciones de hola Swift (la comunidad de Swift en español) como archivo .pptx editable en Keynote, PowerPoint o Google Slides, con el estilo de la marca, el de una invitación de keynote de Apple con el "hola" en tubo 3D, titulares grandes en Inter Display, fondos claro, oscuro o de color y muy poco texto. Úsala siempre que alguien de hola Swift pida una presentación, unas diapositivas, un deck, unas slides, la charla mensual, la plantilla para un ponente invitado, las pantallas de espera o el programa del encuentro en persona, un recap de una charla, una propuesta para patrocinadores o cualquier presentación con la marca hola Swift, aunque no diga "pptx" ni nombre el skill.

Build a truthful, ATS-friendly resume, cover letter and interview prep sheet from zero. Interviews the person with a short question set (8 quick, about 10 standard, more for complex cases; or reads files they send first), saves a fact sheet, drafts a master resume, tailors it and a cover letter to real job postings, builds PDFs and runs checks. Use whenever someone wants a resume, CV, cover letter or job-application package made, rewritten, improved or tailored, even if they only say "help me apply for jobs", "my resume is bad", "I'm a student with no experience" or paste a job link.

Système de création de présentations et de slides : tokens de design, catalogue de quinze mises en page, structure narrative et conventions de rédaction des titres. Appliquer à tout output de type deck, pitch, présentation client, keynote, support de formation ou rapport visuel. Déclencher dès que la demande implique des slides, un deck, une présentation PowerPoint, Google Slides, ou tout support visuel multi-pages. Prend la priorité sur une skill de génération de fichiers (pptx) pour les choix visuels et stylistiques : celle-ci gère la technique, slides gère les tokens, la structure narrative et les mises en page.

Charlotte is a Senior Ghostwriter who writes content published under someone else's name. Activate Charlotte for ANY ghostwriting task - even without "ghostwriter" in the request. If someone says "write in my voice," "write as me," "I have ideas but can't write," "write my LinkedIn post," "write my newsletter as me," "write my book," "write my keynote," "as a [job title] write a post about," "thought leadership for me," "speech for me," or "I'm an executive and want to build my personal brand but don't have time to write" - Charlotte is the skill. Also activate when someone wants to publish content in their own name that Claude will write. Charlotte has no fixed voice - she writes in whoever's voice she is supporting. She always asks for writing samples or existing content examples before producing anything. She does not invent a voice for the client.

Turn one long asset - a YouTube video, podcast, newsletter, blog post, transcript or client call - into a week of LinkedIn posts. Use when the user says "repurpose this", "turn this into posts", "I have a video/newsletter/ transcript", or pastes a long piece of content and wants it on LinkedIn.

Self-edit or review Georgian (ka) text before it is published - fact freeze, natural rewrite, adversarial copy desk, ka_lint gate, and recording corrections with ka_learn. Load after drafting, or when asked to check or fix someone's Georgian.

Act as a meticulous second reviewer on a document before it goes to a client (proposal, quote, invoice, email, report). Find every mistake a client would catch, quoting the exact line, and rank flags by damage (costs money, costs trust, looks sloppy). Use when the user asks to proofread, double-check, or review a client-facing document before sending.

Answers what 17 advertising and editing books say about a question (Schwartz, Caples, Hopkins, Ogilvy, Sugarman, Cialdini, Heath, Bencivenga, Masters of Copywriting, Miller, Sharp, three Hormozi books, Murch, Pearlman, Block), by searching notes on 3,286 passages that were each read in full, and citing the passage ids. Use it whenever someone asks what the books say about a hook, offer, proof, awareness, urgency, pricing, guarantees, rhythm, cuts or where the eye goes, wants evidence for a creative call, wants to settle an argument about an ad, or asks what a named book or author is good for. /bb:script and /bb:edit call it for their decisions.

Viết/sửa chữ tiếng Việt người chơi thấy trong server Minecraft — lang/messages.yml, config item, GUI, NPC, announcement — và cả khi code plugin mới (message, lang file mặc định, lệnh, GUI).

What a piece of writing says and in what order, for a reader who wasn't there. Use when writing or editing anything a person will read (reports, papers, abstracts, summaries, tl;drs, captions, slides, emails, READMEs), or for an editing, proofreading or cleanup pass on a draft.

Audit an AI-generated knowledge tree (knowledge graph, concept map, mind map; 知识树、知识图谱、思维导图) against the source material it was built from, such as lecture slides or courseware exported as PNG/JPG images (课件、讲义截图). Finds wrong definitions, wrong relationships, content not in the source, overgeneralizations, vague points and important omissions, and backs every finding with evidence IDs, page numbers and quotes. Produces report.md and audit.json. Use this skill whenever the user wants to check, verify, review, fact-check or proofread a knowledge tree or study notes generated by AI against their slides, asks whether an AI summary of a course is accurate or complete, or asks "这个知识树对不对" / "对照课件检查" — even if they don't say "audit".

按 NocoBase 官网博客「AI & No-Code」栏目的真实写法，产出自然、没有 AI 腔、SEO/GEO 友好的英文或中文文章，类型包括 GitHub 开源工具盘点、替代品/竞品对比、「用 AI + NocoBase 搭建 X」教程和概念解读。凡是给 NocoBase 写博客、SEO 文章、选型文、榜单、对比文、教程或长文落地页，或要给这类文章出大纲、改写、去 AI 味、做 SEO 检查，都使用这个技能，即使用户只说「写一篇关于 X 的文章」「出个选题大纲」「这篇读起来太像 AI 了」。Use it for any NocoBase blog post, SEO article, listicle, alternatives/comparison post, tutorial or rewrite, in English or Chinese.

Met en page la newsletter que la personne a écrite (texte collé ou page Notion) en mail HTML à sa charte, sans changer un mot, puis la prépare pour son outil d'envoi : campagne Brevo en brouillon si Brevo est connecté (objet, pré-en-tête, expéditeur, liste par défaut), sinon HTML à coller dans Brevo, Mailchimp, Kit ou autre. Jamais d'envoi ni de programmation. Déclenche dès que la personne dit « ma newsletter », « mets en page ma newsletter », « passe ma newsletter dans Brevo », « fais le HTML de ma newsletter », « crée le brouillon de campagne », « prépare ma newsletter de la semaine », ou colle le texte d'une newsletter en demandant de le mettre en forme. Ne pas confondre avec Mon ADN copywriting (écrire ou relire un texte) ni Ma charte (créer la charte).

Use alongside non-commodity-content when writing or editing any page or post for a LOCAL service business: location pages, venue pages, "[service] in [city]" pages, local blog posts, service-area content. Adds the local-EEAT layer (named venues, geographic specificity, first-hand local proof, service-area signals) that the general non-commodity standard doesn't cover, plus a voice-capture and internal-linking workflow, and the nine-part City Page Anatomy with its programmatic gate that every city and service-area page (hand-written or generated) is built to. Skip for non-local content (general guides, national/global topics), use non-commodity-content alone there.

Make any prose an AI writes read like a careful human wrote it - clear, direct, specific, and persuasive. Use when drafting, editing, or reviewing emails, memos, executive summaries, reports, proposals, documentation, posts, slide text, or any writing meant for people, and whenever text sounds generic, padded, promotional, or "AI-like." Covers audience and purpose, top-down structure, plain style, and a full checklist of AI-writing patterns to remove.

Detect and avoid the telltale patterns of AI-generated (LLM, chatbot, ChatGPT-style) writing, based on Wikipedia's "Signs of AI writing" field guide by WikiProject AI Cleanup. Use this skill whenever the user asks whether text was written by AI, wants a draft checked for "AI tells", "slop", puffery or generic chatbot prose, is reviewing Wikipedia edits, articles, essays, submissions or comments for undisclosed AI use, or wants writing revised so it reads as specific and human. Also consult it before writing or editing encyclopedic, journalistic, academic or other formal prose, where inflated significance, vague attributions, invented sources or chatbot formatting would hurt the result.

Write or rewrite a README that makes a stranger star the repo within 15 seconds. Use when creating a README, when the user says the README is weak, before any launch, or when a repo gets traffic but few stars.

Write or rewrite selling copy, then humanize it so it reads like a person wrote it. Use when someone asks to make copy better, write a hook, headline or subject line, or write or rewrite a cold email, DM, follow-up, sales page, landing page, social post, video script, flyer, ad, sales script or offer. Uses notes from Cialdini, Bly, Hormozi, Harry Dry, Tyson 4D and Stefan Georgi, then runs a humanizer pass last.

Apply eight storytelling principles to business communication: predictive processing, establishing shots, word pictures, story structure, Chekhov's gun, a specific villain, sugar coating, and bookending. Use for writing or improving emails, posts, landing pages, scripts, sales narratives, and content strategy; also for discovering what existing content to edit, remove, or develop. Adapt the method to the business, audience, and request; answer direct factual questions plainly.