Writing & Editing

Handles anything to do with a resume, CV or cover letter, in any project, without a slash command. Use whenever the user wants to create, write, build, update, modify, edit, tailor, shorten, fix or review a resume or CV; tailor it to a job description, JD, job posting, job link or role; add a job, skill, certification or achievement to their resume; write a cover letter; see what a job wants that they lack; set up or import their resume; or prep for a role by being grilled on a JD.

Answer in ASD-STE100 Simplified Technical English (STE) style, with short sentences and one clear meaning each, so replies are easy to digest and cannot be misread. Use when the user asks for STE, STE100, ASD-STE100, Simplified Technical English, or controlled language, or asks to rewrite or check text in that style.

Write a short, well-structured recap as the final message of every completed task (a feature, fix, refactor, docs change, investigation, release step). Use at the end of any turn that changed files, ran builds or tests, or finished multi-step work. Not for one-line answers or pure conversation.

>- Write or rewrite GitHub PR descriptions in a scannable ADHD-friendly format: outcome-first lead, two-line before/after, short why-safe list, residual one-liner, tight test plan. Default for all Nakama PR bodies (agents and humans). Use when writing, improving, or rewriting a PR description, opening a PR via ce-commit-push-pr, or when the user mentions ADHD PR / adhd-pr-description. Prefer this shape over long narrative PR prose.

Take a raw product/startup idea through a rigorous, multi-agent product lifecycle — Discovery (problem, assumptions, opportunity, project brief) → Validation (market size, competitors, feasibility, personas, go/no-go, 6-dimension benchmark score) → PRD (requirements, user stories, MoSCoW, UX flows, full PRD) → Coding Context (architecture, SQL schema, task breakdown, CLAUDE.md/.cursorrules, CODING_CONTEXT.md). Use this skill whenever the user wants to validate a startup or SaaS idea, stress-test a product concept, size a market, score an idea, write a PRD from an idea, or turn an idea into architecture docs and a build plan for an AI coding assistant — even if they only ask for one phase (e.g. "is this idea any good?", "write me a PRD for X", "give me a CLAUDE.md and task list to start building Y"), and even if they don't say "LaunchPad".

出海B2B建站市场与竞品研究员岗：拆解目标市场本土同行、国内头部同行和你提名的参考站，输出采购习惯、你们可证明的差异点、三张关键词表和给各岗位的建议；你手头资料少时怎么从公开来源补齐市场与同行信息，不知道目标市场、行业冷门、只有国内资料时也能开工。涉及竞品分析、看同行网站、这个市场买家喜欢什么、关键词怎么选、拆参考站时使用。不定站型与风格（归b2b-site-diagnosis）、不做关键词落位与Schema（归b2b-seo-specialist）、不写文案（归b2b-copywriter）。

Research a topic and write a sourced 35-45 s Shorts script with a fact ledger. Use before any visual work.

>- High-performance 1080x1920 (9:16 vertical) Instagram Reel rendering engine combining headless Chrome (Puppeteer) and FFmpeg concat demuxer. Renders organic ink write-on animations, turbulence displacement wiggle, and crisp typography.

The moment DNS moves and for the thirty days after — verifies the cutover (DNS propagation, certificate, http and www redirects, the noindex, robots and sitemap on the live host, every redirect the audit mapped, forms, analytics, the staging copy, the old host, security headers), diffs production against the pre-launch baseline, and schedules the watch. Use when asked did the launch work, is the site live, what should I check after launch, the domain just switched, or when a site went live in the last month and something looks wrong. It is read-only and exits 0 whatever it finds.

- Fix the offer before fixing the copy. A stronger offer with average words outsells great words on a weak offer every time.

- State assumptions out loud before writing code. If the request has two meanings, present both and ask instead of picking silently.

Write git commit messages in Conventional Commits format. Use whenever creating a commit, when the user asks for a commit message, and at the end of every task that changed files (to propose the commit message).

Write an Architecture Decision Record for a non-trivial technical choice. Use when the user says "should we use X or Y", "document why we chose", "write an ADR", or when a change introduces a new service, datastore, vendor, pattern, or cross-cutting convention that future engineers will question.

Create well-scoped git commits with Conventional Commit messages. Use this whenever the user asks to commit, save work, "ship this", or finishes a change and wants it recorded in git, even if they do not say "commit". Also use when writing or rewriting a commit message.

Зберігає короткий підсумок виконаної задачі в окремий Markdown-файл у папці summaries/ в корені проєкту. Використовуй, коли користувач просить зберегти, записати чи зафіксувати підсумок, резюме, звіт, нотатки або результат роботи ("збережи підсумок", "запиши резюме відповіді", "зафіксуй, що зробили", "save a summary", "write this down"), навіть якщо він не каже слово "файл" чи "markdown".

Before sharing, exporting or publishing a bot template, skill, routine or config, scan it and redact API keys, tokens, passwords, card and bank numbers, internal URLs, personal data and customer data. Returns a scrubbed copy plus a findings report. Draft-only; never shares or publishes anything.

Edit or draft academic prose so it reads as precise human scholarship. Covers journal and conference papers in any field, theses, rebuttals, referee response letters, grant proposals, and Chinese academic text. Removes AI stock phrases, invented contrasts (not X but Y, 不是……而是), cue colons, meta-narrative, defensive hedging, figurative language, em dashes, repeated sentence skeletons, translationese, and disconnected flow. States each claim at the strength its evidence or proven result supports, and keeps every number, equation, citation key, label, formal statement, and genuine qualifier unchanged and verifies this with a bundled script. Use whenever the user asks to polish, proofread, tighten, rewrite, humanize, 润色, 改写, 去AI味, or make a manuscript read more naturally, and whenever the assistant drafts new manuscript text.

Strip AI-essay voice from a document that will be shared outside this conversation — a memo, spec, brief, proposal, SOW, report, or any deliverable going to a supervisor, agency, client, or contractor. Use when asked to \"clean this up\", \"make this professional\", \"take the Claude out of it\", \"de-AI this\", \"make it ready to send\", or when a draft is about to be handed off and has not had a register pass. Also run it unprompted as the last step when drafting any external-facing document. Not for chat replies, personal notes, or internal scratch files — conversational voice is fine there and flattening it is a loss. Not for fiction or other creative writing, where the voice is the point; see the fiction note in the body.

Kick off a new Azure DevOps work item — parse the pasted ADO ticket, cut a <prefix>/<id> branch from the base branch, write a context memory, do codebase discovery, then a root cause analysis (bugs) or a feature layout with targeted elaboration questions (stories/features), and stop at a checkpoint for approval. Use this whenever the user pastes an ADO bug, user story, feature, task or PBI (text containing an ID, Work Item Type, Title, Repro Steps, Acceptance Criteria, or a dev.azure.com link), or says things like "new ticket", "starting 12345", "pick up this bug", "here's the next story", "kick off work on", even if they don't ask for a branch or memory explicitly.

Put work on the correct branch before the first edit. Finds the base branch, starts a well-named branch from the current remote tip, and sets up one worktree per agent for parallel work. Use at the start of any change to a tracked file, such as 'start on issue 42', 'fix this typo', 'set up branches for these three tickets', or before you start sub-agents that edit code. Not for opening or merging a PR (github-pr-merge) or for history rewrites and other risky git commands (git-safety).

Approval gate for git commands that destroy work or skip checks: force-push, --no-verify, signing bypass, reset --hard, clean -f, discarding uncommitted edits, branch or tag deletion, rewriting pushed commits and push --mirror. Each one needs the user's explicit approval at that moment; an earlier approval never carries over. Also covers a secrets check before a commit and what to do when a secret is already committed. Use before any of these commands, or when asked to 'force push', 'skip the hooks', 'throw away my changes' or 'delete that branch'. Not for normal branching (git-branching) or the PR flow (github-pr-merge).

Create or top up a project's documentation set: list what exists, let the user choose which document types to add, collect one shared fact sheet from the code, write the documents (in parallel when there are several), then check them against each other and against the code. Use when asked to 'bootstrap the docs', 'start the documentation', 'we have no README or ARCHITECTURE', or 'fill in the missing docs'. Not for the per-commit update of existing documents (documentation-sync), and not for gener8v pipeline documents (gener8v:brownfield and gener8v:planning make those).

Write or improve Realms of Code realm guides with practical triggers, concrete actions, evidence and fantasy flavor. Use when authoring the charter itself, not when applying its roles to ordinary software work.

>- Use when planning or implementing personalization or A/B testing with Croct: creating an experience (audience, slots, content, priority, scheduling, drafts), adding an experiment (A/B/n, primary goal ID, traffic allocation, variants, cross-device consistency), wiring the code it needs (slots, `croct.track('goalCompleted', {goalId})`, `orderPlaced`, `leadGenerated`, `userSignedUp`), server-side feature-flag experiments, redirect experiments, personalization recipes (UTM campaigns, ABM, search terms, location, free shipping, interests, blog readers, leads, customers, organic intent, HubSpot forms), or reading the experience, experiment, application, or workspace dashboards (probability to be best, potential loss, "Experiment in progress", "Recommended winner", "No clear winner", "Track your goal"). Phrasings: "personalize the hero for LinkedIn visitors", "A/B test this CTA", "which variant won?", "why does my experiment show no conversions?".