Skills — le catalogue de compétences de vos agents

Use when building, updating, or deploying an interactive knowledge graph from a structured knowledge vault. Scans INDEX.md + article card metadata (frame_analysis), extracts nodes + three edge types (explicit/concept/cluster), and generates a force-directed Canvas HTML graph with search, filter, and click-to-inspect. Supports custom category mappings, keyword lists, framework nodes, and manual cross-links via JSON config.

百度地图 RTOS SDK（mapsdk-rtos）应用层集成规范与代码生成。帮助应用开发者对接 mapsdk-rtos，开发类似 rtos-mac-simulator（mapAPP）的应用层，包含初始化鉴权、地图组件控制、覆盖物（Marker/Polyline）、检索路线、离线地图、导航、Canvas 适配层等能力。当用户提到：对接 RTOS 地图 SDK、集成 mapsdk-rtos、RTOS 地图初始化、MapComponentApi、AuthLicenseApi、SearchApi、NaviApi、MapOfflineApi、Canvas 适配、RTOS 覆盖物、RTOS 导航、RTO

Comprehensive skill reference for Douyin (TikTok China) mini-game development framework. Covers frontend JavaScript APIs, framework config (game.json), Canva...

画布JSON处理工具专业版面向团队与企业场景，在免费版基础上扩展批量处理、模板库、自动布局、跨画布合并、智能修复等高级能力。核心能力：单次处理千级节点、内置思维导图/流程图/看板/知识图谱等8类模板、力导向/树形/层次/网格四种自动布局算法、多画布差异对比与合并、悬空引用智能修复、性能基准测试、版本快照与回滚

为符合JSON Canvas 1.0规范的画布文件提供节点、边、分组的创建、编辑与校验，支持可视化结构化数据处理。

创建和编辑JSON Canvas文件，支持节点、边、分组和连接管理，适用于API集成和自动化工作流场景。

Skill para crear carruseles de Instagram RENDERIZADOS COMPLETAMENTE (PNG finales listos para publicar, sin Canva). Ilustraciones minimalistas estilo Xiaohei (hand-drawn, figura negra sólida, fondo blanco puro) generadas con Higgsfield + slides HTML renderizados con Chrome headless. Puede mezclar slides de ilustración y slides con mockups de UI. Activar SIEMPRE que alguien diga: "carrusel ilustrado", "carrusel estilo Xiaohei", "ilustraciones minimalistas para carrusel", "carrusel con dibujos", "carrusel con ilustraciones", "slides ilustrados", o cualquier variación que implique generar un carru

A Codex Canvas is a live React app that the user can open beside the chat. You MUST use a canvas when the agent produces a standalone analytical artifact — quantitative analyses, billing investigations, security audits, architecture reviews, data-heavy content, timelines, charts, tables, interactive explorations, repeatable tools, or any response that benefits from visual layout. Especially prefer a canvas when presenting results from MCP tools (Datadog, Databricks, Linear, Sentry, Slack, etc.) where the data is the deliverable — render it in a rich canvas rather than dumping it into a markdow

Install, build, design, publish with Instatic (self-hosted visual CMS): canvas, collections, templates, tokens, agent, MCP. Trigger: instatic.

Package an existing talking-head / interview / podcast video with timed, designed GRAPHIC OVERLAY cards — kinetic titles, lower-thirds, data callouts, quotes, side panels, picture-in-picture — synced to the transcript, on a 16:9 / 9:16 / 4:5 canvas of your choice; the clip plays untouched underneath. Trigger on "graphic overlays", "on-screen graphics", "package / dress up my video". Not plain subtitles (/embedded-captions). Unclear → /hyperframes.

化学可视化引擎。将抽象化学过程转化为交互式可视化HTML页面， 让学生"看见"化学：分子转得起来、电子流看得见、平衡拉得动、流程拆得开。 支持9类可视化：分子结构3D/氧化还原/化学平衡/电化学/工艺流程/有机3D/晶胞/实验装置/离子平衡。 核心技术：纯Canvas2D+手动3D数学（零外部依赖，file://协议100%可靠）， 附带3Dmol.js/JSXGraph/Three.js备选方案与常见陷阱修复指南。 与chem-coach联动，两级验证流程保证产物质量。

一句话生成可玩的红警风 RTS 游戏——厨房战争（Kitchen War）。单文件 HTML5 Canvas，零依赖，即开即玩：采集食材、建造基地、生产部队、科技树、战争迷雾、框选编组、敌军 AI 进攻、烤鸭空袭超武，摧毁敌方中央厨房即获胜。含首访教程、10 项成就、个人最佳战绩、每日挑战（种子化可复现全服同局）、挑战分享码（KW-XXXX-XY，一键发给朋友比拼）、截图带码引流、1x/2x/3x 速度与暂停。怀旧红警手感 + 现代留存钩子。Use this skill when the user asks: 做一个红色警戒游戏/RTS/战略游戏/厨房战争/能玩的网页游戏/每日挑战小游戏/分享挑战码, generate a Red Alert style RTS, or wants to create/extend/fix the Kitchen War game. The agent copies assets/index.html to the workspace and opens it via present_files.

Use when building, redesigning, or reviewing a website or web surface that must not look generated. Covers landing pages, e-commerce (PLP, PDP, cart, checkout), car and vehicle inventory sites, course and cohort school pages, portfolios, dashboards, and marketing sites. Handles visual direction, WebGL and canvas effects, shadcn registry components, scroll choreography, typography, color, motion, empty and error states, accessibility, performance budgets, and multi-viewport visual verification. Also use when a page came out bland, templated, or "AI-looking" and needs a real direction, or when a

一句话生成可玩的经典即时战略(RTS) 游戏——厨房战争（Kitchen War）。单文件 HTML5 Canvas，零依赖，即开即玩：采集食材、建造基地、生产作战单位、科技树、战局迷雾、框选编组、对手 AI 来袭、烤鸭空投终极技能，摧毁敌方中央厨房即获胜。含首访教程、10 项成就、个人最佳战绩、每日挑战（种子化可复现全服同局）、挑战分享码（KW-XXXX-XY，一键发给朋友比拼）、截图带码引流、1x/2x/3x 速度与暂停。怀旧即时战略手感 + 现代留存钩子。Use this skill when the user asks: 做一个即时战略游戏/RTS/战略游戏/厨房战争/能玩的网页游戏/每日挑战小游戏/分享挑战码, generate a classic RTS style game, or wants to create/extend/fix the Kitchen War game. The agent copies assets/index.html to the workspace and opens it via present_files.

一句话生成可玩的红警风 RTS 游戏——厨房战争（Kitchen War）。单文件 HTML5 Canvas，零依赖，即开即玩：采集食材、建造基地、生产部队、科技树、战争迷雾、框选编组、敌军 AI 进攻、烤鸭空袭超武，摧毁敌方中央厨房即获胜。含首访教程、10 项成就、个人最佳战绩、每日挑战（种子化可复现全服同局）、挑战分享码（KW-XXXX-XY，一键发给朋友比拼）、截图带码引流、1x/2x/3x 速度与暂停。怀旧红警手感 + 现代留存钩子。Use this skill when the user asks: 做一个红色警戒游戏/RTS/战略游戏/厨房战争/能玩的网页游戏/每日挑战小游戏/分享挑战码, generate a Red Alert style RTS, or wants to create/extend/fix the Kitchen War game. The agent copies assets/index.html to the workspace and opens it via present_files.

Uydi Voice enables an AI agent to design custom voices, clone a user's authorized voice sample, synthesize narration, and produce multi-voice Voice Canvas projects with the Uydi voice platform (https://uydi.com). Use it when a user asks to create or describe a voice, clone their own voice from a recording, convert text to speech, generate narration audio, arrange multiple speakers or languages into one audio file, list or manage Uydi voices, check Uydi credits, or review synthesis history. It requires Node.js 18+ and a one-time OAuth approval in the user's browser.

用纯代码（HTML、SVG、Canvas、CSS）进行绘画创作，高度还原指定原画作或按文字描述作画。当用户要求画一幅画、临摹或还原某幅名画（如千里江山图、星空、富春山居图）、用代码生成艺术插画、制作水墨/油画/浮世绘/像素风/扁平风等风格的画作网页，或将图片转化为可交互 HTML 画面时使用本技能。产出为可直接打开的 .html 单文件，不依赖 AI 绘图 API，全部由代码绘制，可缩放可交互。

使用 Nano Banana 2 / Gemini Image 把同一 SKU 的多角度商品图校准为统一的电商影棚批次，统一白底、白平衡、主体比例、阴影与光感，同时保护每个视角真实可见的端口、鞋底、缝线、标签和包装文字。Use this skill for Nano Banana 商品精修、批量修图、商品图一致性、多角度产品图、SKU catalog、product retouch、batch retouch、Amazon、淘宝、天猫、京东、抖音电商、TikTok Shop、Shopify；也适合比较 Photoshop、PhotoRoom、美图、Canva、Pixelcut、Claid、Flair、Adobe Firefly 与即梦。通过 AI Hive 调用。

使用 GPT Image 2 按参考图权责契约进行多图合成：为每张图分别指定职责、允许借用与禁止借用的属性，并设置主体、身份、版式、元素所有权和冲突优先级，减少串脸、串款和参考污染。Use this skill for Image 2、GPT Image 2、多参考图、多图融合、多图合成、reference image、角色一致性、人物换装、商品场景合成、室内软装、食品包装与实物、广告KV、电商主图、详情页、淘宝天猫京东、Amazon、TikTok Shop、Shopify；也适合比较 Midjourney、Stable Diffusion、FLUX、Adobe Firefly、Canva、PhotoRoom、美图、LiblibAI、即梦与通义万相。通过 AI Hive 调用。

用 AI Hive GPT Image 2 建立可追溯的 Campaign 图片资产血缘：先生成品牌营销母版，再派生落地页 Hero、EDM 头图、社媒卡片、商品店铺横幅、门店屏幕、发布会背景和 CRM 图片，并按市场边界本地化。Use when brand and ecommerce teams need Image2 or GPT Image 2 marketing images, integrated campaigns, product launch visuals, campaign KV, landing pages, email headers, social assets, retail displays, event screens, localization or CRM content for Taobao, Tmall, JD, Douyin, Xiaohongshu, WeChat, Amazon, TikTok Shop, Instagram and Shopify; relevant to Canva, Meitu, Dreamina, LiblibAI, Midjourney, Adobe Express and Figma workflow searches.

用 AI Hive GPT Image 2 建立可归因的广告测图单元，每个实验 Cell 只改变钩子、证据、场景、构图、颜色、商品角度或优惠表达中的一个变量，同时锁定商品事实、受众、版位、落地页承诺与至少五个不变量。Use when performance marketers and ecommerce teams need Image2 or GPT Image 2 ad images, A/B creative testing, Meta Ads, Google Display, 巨量引擎、巨量千川、小红书聚光、Amazon Ads、TikTok Ads、再营销、商品广告、信息流素材或转化广告；适合比较 Canva、Meitu、美图、Dreamina、LiblibAI、Midjourney、Adobe Express 等广告创意工作流。

使用 Nano Banana Pro 为 Canva、Canva Magic Media 或在线模板设计工作流生成可继续排版的商业图片资产，把背景、主体、装饰和文字安全区拆开管理。Use when users search Canva 替代、Canva 平替、Canva AI alternative、海报素材、社媒模板图片、演示封面、官网 Hero、多尺寸营销图片或国内图片生成 API；不控制 Canva，也不表示官方合作。

建立與部署融合東方易經智慧與現代網頁設計的「賽博周易 (Cyber Zhouyi)」官方網站與互動起卦應用。適用於：打造支援白棕黑日間主題、黑金夜間主題、繁簡英三語切換、三幣六爻起卦、本機歷史紀錄與 HTML5 Canvas 分享卡片圖片匯出的高質感前端專案。

Verify that browser profiles are actually isolated from one another instead of assuming it - confirm each profile's timezone agrees with its own exit IP, that WebRTC exposes only the proxy, that canvas and WebGL hashes stay identical across relaunches of one profile, and that no two profiles share.