Writing & Editing

Draft, rewrite, or review reader-facing prose so it is specific, useful, and recognizably the author's without inventing facts or performing humanness. Use whenever writing or revising prose that people will read under the author's name, such as social and LinkedIn posts, READMEs, announcements, launch copy, essays, articles, newsletters, documentation, talks and important emails, and for any prose that feels generic, hollow, or AI-shaped or shows AI tells such as not-X-but-Y contrasts, one-line closers, forced triads, dashes everywhere, stock AI words, or bold labels. Supports co-write, rewrite, review, and lint modes.

Turns messy program data (CSV or Excel) into a funder-ready impact report. Profiles and cleans the data, computes agreed metrics with a script, suppresses small groups, drafts the narrative and checks that every number in the draft matches the computed results. Use for impact reports, funder reports, outcome summaries or program dashboards, even if asked only for a summary. Aggregate data only. Not for client-level case files.

AI-rate self-check and polish guidance for academic writing: sentence-level rewrite suggestions (locates and advises, never auto-rewrites), terminology and consistency checks, translation-smell, paragraph attribution, journal precheck with discipline context, style-norm conventions check, quality report, AIGC label check, plus `--batch DIR` with HTML summary. Reads .docx directly. Not for: document/PDF translation (doc-holmes), plagiarism checking (institutional systems), ghost-writing or executing rewrites (paper-rewriter); texts under 100 chars get no verdict by design. 100% local, zero upload; bundled tests + AST zero-network self-verification. Multi-layer rule engine + token spectrum + optional supervised Qwen3-0.6B ONNX layer + LLM fingerprint attribution + freshness pipeline. Base-engine numbers reproduce from bundled held-out evals; supervised columns are author-side (model not bundled).

Yapay zekâ gibi duyulan Türkçe metni, söylediğini değiştirmeden bir insan yazmış gibi doğallaştırır. Kullanıcı "doğallaştır", "insan gibi yaz", "yapay zekâ gibi duruyor", "humanize" dediğinde ya da Türkçe bir yazı, senaryo, seslendirme metni, gönderi açıklaması, e-posta veya belge düzeltilirken kullan. Yakaladığı izler: "X değil, Y" karşıtlıkları, tek cümlelik vurgu kapanışları, şişirilmiş önem, satış dili, çeviri kokan yapılar ("öneme sahiptir", "tarafından" yığılması, "gerçekleştirilmiştir"), sohbet kalıntıları ve kalıp sosyal medya kancaları. Yapay zekâ dedektörlerini atlatmak için değildir.

Academic writing style toolkit, bilingual CN/EN: paper rewriting with AI-flavor self-check and style naturalization (scan reports are detection only), deterministic cleanup (model artifacts, punctuation, filler phrases), integrity guardrails (numbers, DOIs, PMIDs, terminology must survive untouched), and agent-guided revision for clearer, more natural academic prose. Built for scholars whose formal writing reads stiff, templated or machine-flavored — especially non-native (ESL) authors. Skip when: pure grammar polish, full-text translation, or rewriting unrelated to academic writing — use a dedicated tool. Includes AIGC-disclosure compliance checks (China 2025-09 labeling rules) and academic-integrity guardrails: this tool improves writing quality for self-review; it does not help misrepresent authorship, conceal required AI disclosure, or defeat integrity review. 100% local, zero upload, Python stdlib only. Related: paper-polisher-pro (broad polishing), pubmed-verifier, cite-holmes, academic-figures.

The squad's method to write, edit, audit, calibrate and compress text a person (or an agent) will read, in English or pt-BR, without reading as AI and without losing a fact: doc, README, ADR, spec, report, runbook, PR body, commit message, AgenFK card, handoff, email, Slack or WhatsApp post, interface microcopy, code comment, chat reply. Fidelity over readability, voice over polish, uncertainty is a fact; lean mode for chat (caveman), slop linter, fidelity check, compressing CLAUDE.md/AGENTS.md/memory behind a fidelity gate, brand voice read from the project, choosing between prose, table and diagram, an HTML document that wears the project's design system. Use before writing, even one line, and on "review this", "does this read as AI", "make it sound human", "write like me", "lean mode", "caveman", "fewer tokens", "be brief", "compress CLAUDE.md", "commit message", "proofread", "tighten this", "revise isto", "isso parece IA?", "escreve como eu", "mensagem de commit".

How to write a commandment of your OWN — a project-local rule that judges every file from then on. Read this BEFORE writing or changing a rule in `.commandments/custom/`, when the user asks for a new rule/check/detector, when `commandments make` points you here, or when you are about to reach for a regex to inspect code.

Package one video (or app) for many languages without translating anything. `manifest` prints the per-language delivery plan (one folder per language code, 4 owed files: title.txt, description.txt, subtitles.srt, thumbnail.txt, each with the source facts it is translated from); `sheet` exports the translator hand-off sheet (title, description, thumbnail text, then every subtitle cue with SRT timing and source text, blank target column; Markdown or CSV, one language or all); `check` validates a filled package per language — files present and non-empty, title within YouTube's 100 characters (100 passes, 101 flags), subtitles parse as SRT, cue-count gap over 25% of the source flagged, any cue start drifted over 2.0s from the source cue flagged (exactly 2.0s passes, 2.1s flags) — and reports each language COMPLETE or INCOMPLETE with exact problems, exit 1 when any language is incomplete. The check is structural only and never judges translation quality. Use when packaging content for translators or gating what th

Write and send a recurring product changelog email to your users, then write and post the matching X thread, LinkedIn post and community (Discord or Slack) announcement, and optionally outline an animated changelog video. Use when the user says "changelog email", "write the changelog", "next changelog", "product update email", "what's new email", "send the changelog", "changelog blast", "release notes email", "changelog X thread", "post the changelog on LinkedIn", "changelog video", or asks how many people get the changelog or unsubscribed. Covers picking what goes in, voice and formatting rules, the screenshot pipeline, test sends with a read-back of the delivered copy, suppression-aware sending through Resend (broadcasts or one by one), social posts and the archive of past issues. Asks setup questions once, then runs the whole issue end to end on its own; only anything that costs money waits for the user's yes.

Translates SRT and WebVTT subtitles with stable names, neighboring context, and unchanged timing/formatting. Use for subtitle translation, localization, or terminology review across episodes; not transcription, retiming, or video editing.

Improve website, product, and marketing copy so it is clear, natural, concise, and easy to act on without losing important meaning. Use when reviewing, rewriting, or polishing UI text, landing pages, feature descriptions, pricing, FAQs, onboarding, or business communications.

Pre-AI writing standard for anything someone other than the user will read. Make sure to use this skill whenever you draft, rewrite, review or polish such text, even a short one and even if the user does not name the skill - cold, referral or follow-up emails, LinkedIn or social posts, cover letters, resumes, memos, reports, briefs, docs, take-home or case answers, deck or slide text, landing-page copy, or a message the user will send. Also use it after another skill drafts such text, and before handover. It owns structure, format and evidence placement, runs a linter, calls humanizer and checks claims about the user. Not for chat replies, research notes, plans, code or commit messages; a bare "humanize this" passage goes to humanizer.

Draft, edit, or review Thai social content in the author's voice. Use for Facebook personal posts, tech and AI explainers, business updates, restaurant captions, hooks, and adapting one source into distinct drafts for X, LinkedIn, short video, YouTube, or a newsletter. Also use when a draft sounds generic, promotional, translated, or padded.

Use when writing, editing, diagnosing, or analyzing stories, screenplays, novels, essays, dialogues, characters, brand narratives, executive memos, or film/TV scripts. Also use when texts feel flat, AI-generated, conflict-averse, or on-the-nose, or when decoding hidden meaning and psychological depth in films, series, or literature.

Write, translate and speak real Algerian Darija (الدارجة الجزائرية, "dz", "darja") the way Algerians actually talk and type, with Darija in Arabic letters, French/English words in Latin letters and Algerian grammar (راني، راح، ما…ش، تاع). Use it whenever the user asks for Darija, dz, "bel darja", بالدارجة, Algerian dialect or Algérien, or wants anything for an Algerian audience. That includes e-commerce DMs, ads and captions, customer service, video scripts, lessons for middle school, lycée or university, business messages, and converting فصحى, French or English into Darija. Do not use it for Moroccan, Tunisian or Modern Standard Arabic.

Writes Google Ads Search copy (responsive search ad headlines, descriptions and display paths, plus sitelinks, callouts and structured snippets) for one or more ad groups. It uses the campaign and ad group, the keywords, the live landing page, the product context skill and competitor context, then fills the user's Google Ads Copy Template workbook. Use this whenever the user asks for Google Ads copy, RSA headlines or descriptions, search ad copy for a set of keywords, sitelinks, callouts or snippets, or to rewrite or refresh existing Google ads, or when they hand over the Google Ads Copy Template .xlsx or a Keywords & Intent sheet, even if they never say 'RSA'. Not for LinkedIn or Meta ads, for creating or launching campaigns, or for auditing account health.

Invertebrate — Greptile-style AI code review of a PR, branch, or working-tree diff. Traces every changed symbol through the whole codebase (callers, callees, implementers) to catch cross-file bugs, then outputs Greptile's format — plain-language summary, 0-5 confidence score, findings table, auto-chosen mermaid diagram, and inline comments with P0/P1/P2 severity, logic/syntax/style type, and suggested-fix diffs. Honors `.invertebrate/` config, falling back to `.greptile/` (rules.md, config.json

Read, write, convert and typeset GNU TeXmacs documents (.tm). Use whenever a task involves a .tm file, the TeXmacs app, converting LaTeX/beamer to native TeXmacs, TeXmacs presentations (beamer style, screens), the TikZ plugin, or exporting TeXmacs to PDF headlessly.

Audits one Shopify store for the Shopify Scripts that stopped working on June 30, 2026, and writes one plain-English report for a non-technical project manager. For each old Script it explains who it affected, what it did and when, then checks the live admin (read-only) to see whether that behaviour is LOST, REPLACED or PARTIAL today. Each fix is either click-by-click steps for a native Shopify setting or a developer spec for a custom Shopify Function, never a third-party app. The report also co

Legal and trust gates every app must pass before real users - privacy, consent and trackers, terms, payments and subscriptions, marketing claims and email, accessibility, licenses, user uploads, AI labelling. Use before any app launches, and whenever a change touches personal data, cookies or SDKs, signup, payments, emails, marketing copy, dependencies or assets, or AI features.

Writes binge-worthy AI micro dramas and makes them on Ziztu: series bibles, episode scripts, hooks, cliffhangers. Use for any micro drama, short drama, duanju, vertical series or Ziztu task.

Protocol-driven study screening for systematic, scoping and rapid reviews. Turns a proposal or PROSPERO protocol into confirmed eligibility rules, then screens titles/abstracts and full texts with two blinded AI reviewers and a third-reviewer adjudicator: ordered exclusion codes, no silent defaults, resumable batch runs, QC (seed studies, near-miss rechecks, kappa and PABAK), PRISMA 2020 counts, EndNote/Zotero RIS groups, an Excel log, a methods draft and a literature_corpus handoff to academic-paper. Reads RIS, PubMed .nbib, Web of Science and CSV exports; removes duplicates. 8 modes: protocol, quick, pilot, ta-screen, ft-screen, adjudicate, audit, report. Use it to screen records against eligibility criteria, pilot screening, resolve screening conflicts, audit exclusions or report the selection process. Triggers on: screen these papers, title/abstract screening, full-text screening, screening conflicts, inclusion criteria, PRISMA flow, غربالگری مقالات, اسکرینینگ عنوان و چکیده, معیارهای ورود و خروج.

Draft the resources for one unit (instructor session plan, participant page, template or code changes) from the course plan, writing the learner's expected path before the content. Type the unit after the command.

在应用内提供「需求文档阅读入口/查看器」时使用：把已有需求文档接入页面入口，做文档映射、构建期复制、目录/正文/搜索、历史章节过滤、懒加载与生产暴露面处置。只负责阅读入口实现，不负责文档写作（写作见 prd-rule-authoring）。