Prepare, validate, build, and use Nemotron Customizer airgap image bundles for offline clusters. Use when planning airgapped deployments, editing deploy/nemotron-customizer/airgap/airgap.yaml, selecting workflow targets, grouping step execution images, baking repo overlays or wheel additions, resuming airgap runner builds, or submitting `nemotron steps run` jobs inside an airgapped environment.

编写、精简或审查中英文界面文案：按钮、标题、表单、报错、空状态、toast、确认弹窗、设置项和 i18n 源文案，以及给产品内 AI 生成文案写约束。用于“改下文案”“这个报错怎么写”“界面字太多”“去 AI slop”“审一下这些 key”。长段落的 Slack、文档和文章交给 chinese-writing；翻译交给 gemini-translate。

Reference and helper for the PopNow AI generation API (api.popnow.ai) - API keys and auth, the model catalog and input schemas, the async task lifecycle (quote, createTask, recordInfo polling, wait, retry, purge), credits and holds, error codes, rate limits, idempotency keys, webhooks with signature verification, file uploads, the asset library and the OpenAI-compatible /v1/videos endpoints. Use when writing or debugging code that integrates PopNow, answering questions about how the PopNow API works, or running a PopNow call not covered by the video, image, audio or effects skills.

Serves a real URL behind a custom display domain through a local trusted-HTTPS reverse proxy (mkcert certificate, /etc/hosts entry) with `web mask`, so demos and recordings show a clean domain with a padlock. macOS only. Use when the user wants to mask, alias, or fake a domain for a demo, recording, or screenshot.

Reduces a webpage to a structural skeleton. The web CLI detects sections and replaces content with semantic tokens (text, headings, images, CTAs, links, inputs), then the agent collapses repeated patterns and decorative wrappers and writes skeleton.html and manifest.json. Use when migrating a page to AEM Edge Delivery Services, extracting a page blueprint, or analyzing page structure for block generation.

Build, audit, improve, test, or prepare n8n community-node packages for npm and Creator Portal verification. Use for n8n node UX, credentials, node properties, community-package metadata, scanners, local node development, and verified-node releases; not for merely authoring or running an ordinary n8n workflow.

Drive an open PR to mergeable. Watch CI, fix real failures at the root, triage review comments (fix, dismiss, or ask), and resolve conflicts. Use for /babysit, "babysit this PR", or right after opening a PR.

Use before any change to how a screen looks in this repository — a new page or component, a layout, a redesign, or a one-line colour or spacing fix. Maps each frontend to the DESIGN.md or wiki page that governs its look, ranks that document above any design skill, says which design skill fits which task and which must never touch a surface with a design system, and how to show the result rendered. Loaded by the designer and design-mechanic agents; any session doing UI work can load it.

Sync the Trail Sense Priorities GitHub project (kylecorry31 project 2) with the repo's open issues - remove closed and community available issues, and rank new and misplaced issues for approval before adding or moving them. Use when asked to sync with issues or update the priority project.

Use when a task risks running out of context window, when a long-running agent session or multi-step workflow must stay coherent across many turns, when token cost or latency climbs with conversation length, or when the user asks about compaction, summarizing conversation history, truncating tool output, KV-cache or prefix caching, cache-friendly prompt ordering, context budgets, or splitting work across sub-agents. Covers compaction, observation masking, cache-aware layout, and context partitioning, both for an agent managing its own session and for code that builds agent systems. Do not use for RAG retrieval design or vector database selection.

Prepare a front-on, background-free Tranmere match programme cover from an image URL, upload it to Cloudflare R2 and link it to an existing D1 game. Use when asked to add or replace a programme image for a known fixture. Do not use for PDF ingestion or metadata-only programme edits.

Draft the GitHub release body for a new Docus version. Use when asked to write release notes, do the release notes, or prepare the GitHub release for a version. Generates emoji sections, Contributors and new docs candidates from CHANGELOG.md into a local file to edit and paste, plus Highlights on minor and major versions.

Drives a real browser through the omowright library from the js eval kernel: sites the user is already signed into, forms and clicks, JS-rendered pages, screenshots, web QA, extension popups, a human handoff for login, CAPTCHA or OTP, and a browser you own for scraping, bot-scored targets, network capture and QA traces. Use for any interactive browser task; not for a plain search or an unblocked static fetch.

Look up Y Combinator companies and batches from the public yc-oss API, a static JSON dataset refreshed daily: company profiles, batch rosters (e.g. W25, S24), companies by industry or tag, top companies, who is hiring, founder-diversity lists, and overall YC stats. Use this skill whenever the user asks about YC-backed startups, a YC batch, YC companies in a sector or tag or that are hiring, the Y Combinator portfolio, or startup and venture research that draws on YC data. Read-only.

Use when auditing an existing agent skill (SKILL.md) to find where it leans on LLM judgment for work that could be deterministic, and to emit runnable promptfoo assertions that validate the deterministic parts. Covers two axes — offloading AI steps to code, and constraining AI output format. Triggers for "/determinize-skill PATH", "make this skill more deterministic", "audit a skill for determinism", "where can this skill be deterministic".

为 MaaEnd 添加、导入或补充节点识别测试截图，并按仓库约定脱敏图片、放入 tests/MaaEndTestset、维护 tests/**/test_*.json 和运行 pnpm check/pnpm test 验证。用户提到“添加测试图片/截图”“把截图加入 ADB 或 Win32 测试集”“给节点补正例或反例”“更新 hits/box”“UID 打码后提交测试图”时使用；即使用户只说把某张游戏截图放进 MaaEnd 测试，也要用本技能避免未脱敏原图进入仓库。

Compare two versions of a page and see exactly what changed — a live URL against its cached copy, two URLs, or two markdown blobs. Section-level hunks, word- or line-level granularity, or a summary of counts. Use when the user says "what changed", "diff these", "compare this page to last time", "show me the changes", "did this doc update", or wants a changelog between two versions of the same content.

Automates Apple-platform apps (iOS, tvOS, macOS), Android devices (including Android TV and Fire TV on Fire OS), and Amazon Vega OS TV apps in Vega Virtual Devices. Use when navigating apps, taking snapshots/screenshots where supported, driving TV remotes, tapping, typing, scrolling, extracting UI info, collecting evidence, or planning agent-device CLI commands.

Set up or modernize code-quality tooling for JavaScript and TypeScript projects. Use when creating a project, adding code-health or CI quality checks, making a repository agent-ready, or consolidating dead-code, duplication, architecture, dependency, and changed-code analysis. Do not use for formatting-only, lint-rule-only, or TypeScript type-error tasks. Do not use to run an analysis of a project that is already set up; use the fallow skill for that.

Set up or modernize code-quality tooling for JavaScript and TypeScript projects. Use when creating a project, adding code-health or CI quality checks, making a repository agent-ready, or consolidating dead-code, duplication, architecture, dependency, and changed-code analysis. Do not use for formatting-only, lint-rule-only, or TypeScript type-error tasks. Do not use to run an analysis of a project that is already set up; use the fallow skill for that.