Image

Generate 16:9 (1280x720) YouTube thumbnails for any channel. Use when a video needs a thumbnail - the workflow analyzes the video, selects design references, collects missing inputs, renders variants, and shows the result for approval before any upload.

Use for a specific job application - fit check against a posting, tailored CV and cover letter, job-platform profile alignment, recruiter outreach, interview prep, salary and follow-ups.

Turns a video idea and its title into a designer-ready thumbnail brief: one focal subject, a face and a named emotion (or the hero object), 4 words of overlay text or fewer that never repeat the title, contrast, composition and a phone-size read test, delivered as 2 to 3 concepts for YouTube's Test & Compare. Use this skill whenever the user says things like \"what should my thumbnail be\", \"write a thumbnail brief\", \"thumbnail ideas for this video\", \"brief for my designer\", \"prompt for a thumbnail image\", \"my thumbnail isn't getting clicks\", \"A/B thumbnail concepts\". Not for the title words (use yt-title-optimizer) or for designing the image itself (hand the brief to youtube-thumbnail-design or your designer).

Plans an Instagram carousel slide by slide — a hook slide that earns the swipe, the best value up front, one point per slide, and a final slide worth saving — plus the caption, delivered as a slide-by-slide script you can drop into Canva or any design tool. Use this skill whenever the user says things like \"plan a carousel\", \"turn this into an Instagram carousel\", \"write my slides\", \"make a carousel that gets saves\", \"how many slides should this be\", or \"carousel about X\". Not for a single image or Reel caption (use ig-caption-writer) and not for designing the visuals themselves (use carousel-generator).

Create or revise editorial posters where a photoreal glass object becomes a usable space inhabited by tiny 2D hand-drawn adults, visibly refracted through the glass. Use for “透明玻璃 × 手绘微型人物”, a glass-cup pool, perfume-bottle elevator, or similar glass-interior scenes.

Design and generate interactive, animated Nord-palette HTML infographics, architecture reports, and CSP-sandboxed Agentic Judge evaluation reports with collapsible TOC sidebars, SVG flow animations, interactive lifecycle steppers, and color-coded comparison matrices. Use when creating technical reports, benchmark presentations, architecture overviews, or final judge evaluation reports.

学习参考海报的真实视觉风格并积累用户本地审美图库，结合已保存的风格参考图片、设计规则和用户照片，创作原创海报。适用于“学习这张海报并加入我的参考库”“用我喜欢的风格制作海报”“根据照片做视觉日记”“优化海报”。支持 reference-first、私人 Taste Library、真实图片输入、保真与审美评审。Analyze and save local poster references; create with visual references and user photos.

上传一张咖啡、奶茶或任意饮品照片，自动识别品牌、杯型、配色与配料，用图生图把它改造成一座「由杯子组成的」2D 等轴测微缩门店插画（粗黑描边、纯色平涂、品牌标题、吉祥物、楼梯与绿植），再用代码把真实原图与插画无裁切拼接成高清对比图，交付「独立插画 + 拼接图」两张 PNG。触发词：饮品照片变插画、咖啡照片变门店、奶茶杯子变成店、杯子门店插画、等轴测微缩建筑插画、Grid Coffee 风格、照片转等轴测插画、drink to isometric shop、把饮料变成小店。

Manage skills, agents, extras, hooks, plugins, and MCP connection settings with the Skillshare CLI. Use when the user asks to configure or run Skillshare, install or sync resources across AI tools, manage shared memory notes, review a session for lessons or dream over saved memories, import MCP settings, manage targets and project folders, audit skills, recover backups, or troubleshoot Skillshare configuration and sync. Covers global and project modes, noninteractive automation, and guidance for the terminal UI.

Build a LinkedIn document post (carousel) - slide-by-slide copy, the cover that earns the swipe, and the PDF to upload. Use when the user says carousel", "document post", "slides for LinkedIn", "turn this into a carousel", or has a list-shaped idea that would die as a text post.

Create consistent flat illustrations and fit them inside PowerPoint cards or empty illustration slots without distortion or accidental cropping. Use for card artwork, spot illustrations, or flat modern visuals. カード内の挿絵、フラットイラストの生成・配置に使用。

FilmForge story: turns a premise, song, poem or myth into a story bible, a timed screenplay and a shot list sized to a credit budget, written as YAML the tools validate, price and cut to music. Use this skill whenever the user has an idea to turn into a short film, painted film, music video or picture book; wants a premise, logline, story bible, beat sheet, screenplay, storyboard or shot list; asks how many shots or how long a film their credits buy; or wants an existing story tightened so it lands, even if they never say 'shot list'.

图片设计（全类型：视频封面 / 海报 / 文章配图 / 社交图 / 电商图 / 信息图 / 视频内嵌图）。主线是 AI 生图打底再叠加文案层：模式分 hybrid（默认，生图+叠字）/ image（纯画面）/ web（中文必须精确或大段文字时）；类型与必备要素见 `08` 册，三道质检门（规格预检 / 提示词要素 / 成图指标）见 `09` 册。封面另有 8 种样式（含竖排贴左），可自动选也可指定（`cover_styles.py`），产物 PNG 可直接铺到视频首帧。不做：视频本体生成（→fxai-audio-video / fxai-mv-pipeline / fxai-short-drama）、纯配音（→fxai-voiceover）。

本地 ComfyUI 文生图能力（千问2.1 模板，原生中文提示词）。当用户要求在对话里生成图片、配图、画一张图、出张图、做封面图、用中文提示词生图、给文章/视频/设计配视觉素材时使用。生成的是用户自有角色形象的图（参考图 000 女 / 001 男，规避他人肖像侵权），产物经 HTTP 下载到本地并返回 PNG 路径。仅依赖 ComfyUI HTTP 服务（由 comfy_url 指定，本地或远端均可），提交/探测/下载全部经 HTTP 接口完成，绝不读取 ComfyUI 本机目录；无常驻进程。 # std:allow

FilmForge deliver: encodes and checks every deliverable of a finished film or book (4K master, 1080p for posting, phone copy, cover stills, print-book PDF, making-of, localised editions) with exact ffmpeg settings, verifies streams, loudness and picture-to-music sync on the finished files, then builds the production archive with a SHA-256 completeness proof that anyone can re-verify later, and proves backup copies byte for byte. Use this skill whenever the user wants to export, encode or render final versions, post a film, make a phone copy or covers, produce a print PDF or a translated edition, back up or archive a production, or prove that an archive or copy is complete.

Order Executor. Sends approved orders to your broker, paper trading by default, never without your yes. Shows each order as a plain ticket (a bracket shows its stop-loss, take-profit and the max loss at the stop) plus the exact JSON for Alpaca's place_stock_order, records your yes word for word, then Claude's own Allow once / Deny card asks again. Builds the sell ticket that closes a day trade by your flat-by time, records fills, and logs each trade's max loss so trader-risk can hold the daily limit. Covers setting up Alpaca's official MCP server in Claude Desktop (paper by default, keys in env vars, never in chat), the Interactive Brokers connector, and setting order tools to "Needs approval". Use when the user says "place the order", "buy it", "execute", "send it to my broker", close my position", "flatten", "get me out", "connect Alpaca", "set up paper trading", or after /trader-risk passes.

DanLogo · East Minimal 东方极简：让 AI 工具（Claude、Codex 等）以老子“无之用”的东方思想为根，做成套的 logo 与品牌标志体系——空与白、以少见多、让物性说话； 从品牌身上找出人人见过却没人注意的东西，画成一个有外形、有面的图形（不是字的变体），用最少的形承载最深的意思，图形与中英文定制字同源，留出空间让人自己完成理解。 按九个阶段从简报做到交付：简报 → 概念（每轮 5 套）→ 原型（每个概念 ≥3 套）→ 图形精修 → 定稿（定制字 + 配色）→ 体系 → 场景 → 提案 → 总结，含后期制作与交付文件包。 Use this skill when the user wants to design, redesign, refine, extend or present a logo, wordmark, 标志, 字标, 中英文组合标, 品牌视觉, VI or brand identity in an East-Asian minimalist spirit — 东方极简, 东方美学, 留白, 空, 白, 禅意, 无印良品, 原研哉, 克制, 高级感 — when they name danlogo-eastminimal, or when they continue a project whose decisions.md names it. For a logo request that names no style, use it unless another installed danlogo style fits better; if several danlogo styles are installed, ask the user which one first.

为小红书、抖音或视频制作带创作者形象的封面，也可单独起标题。根据选题设计人物动作、服装、表情、道具、场景和标题，并用创作者提供的肖像保留辨识度。用于人物联动封面、个人形象封面或明确调用 Creator Cover 的任务。

Make YouTube and social thumbnails (1280x720, plus 1080x1920 Shorts/Reels and 1200x630 link previews) with Pillow - a large auto-fitted title with one highlighted keyword on a clean flat or blurred-photo background, plus either a real screenshot with a hand-drawn red circle and arrow (tech and tutorial style), or a person photo/cut-out that can stand in front of the text. Checks readability at YouTube search size, the squint test, and YouTube's timestamp corner, and writes a preview sheet. Use when the user asks for a thumbnail, cover image, video title card or gig image.

Use when building, testing, packaging, or managing dependencies for C/C++ projects with the build2 toolchain. Covers the build system driver (b), package manager (bpkg), project dependency manager (bdep), task runner (bx), and repository web interface (brep). Activate for: writing or reading buildfiles, diagnosing build errors, resolving package dependency conflicts, packaging a library for cppget.org, setting up build configurations, configuring installations, authoring testscripts, hosting package repositories, or running any b/bpkg/bdep/bx/brep-* command.

Turns a PDF abstract into a polished, fully-editable scientific conference poster (.pptx + PDF + preview) in a matched house style, with honest charts, verified citations, big readable typography, and a render-and-QA loop. Use whenever the user wants to make, build, design, or create a research poster, scientific poster, academic poster, conference poster, or ePoster (e.g. for RSNA, AHA, ASH, ASCO, ARVO, AAO, SNIS, or a residency research day) from an abstract, paper, or study; to imitate an existing poster's style with new content; or when they mention turning an abstract or paper into a poster, a poster from a PDF, or updating a poster's data, charts, layout, or making poster text bigger or cleaner. Builds native, editable PowerPoint charts by default (avoiding the PowerPoint chart-repair prompt), or polished image charts on request.

Make and edit images: posters, social graphics, thumbnails, product shots, logos, icons, illustrations, characters, UI mockups, infographics, photo edits and restyles. Also decides between an AI image, a chart, a diagram and a rendered layout.

黑板粉笔风的文章封面图和配图：纯黑底 + 白色粉笔乱线小人 + 手写体字，只有图标和个别标注带颜色。用户给一整篇文章（或口播稿）要配图时，自动挑出哪几处需要图，列清单、出样图，认可后画完全部，并给出插好图的文章。用户说「封面图」「头图」「配图」「文章插图」「给文章画图」「做张图说明这个」，或要给 X 长文、公众号、博客、帖子配图时，只要没有指定别的风格，就用这个 skill。中文、英文文章都可以。Codex 用自己的生图工具直接画，字用 HTML 加。

Writes AI image prompts with a strong, recognizable art-direction signature (raw, human-first, trend-driven, typography-led) for Midjourney, ChatGPT/GPT Image, Gemini, Flux, Ideogram and similar. Use whenever the user asks for an image prompt, a visual, a poster, a key visual, an ad image, a moodboard prompt, a "prompt Midjourney", or wants an idea turned into an AI image — even a one-line brief.