设计与三维建模

设计、实现并重构端口与适配器系统，具备清晰的领域边界、依赖倒置，以及跨 TypeScript、Java、Kotlin 和 Go 服务的可测试用例编排。适用于引入或重构为端口与适配器架构，或领域逻辑已与 I/O 纠缠的情况。

设计、实现并重构端口与适配器系统，具备清晰的领域边界、依赖倒置，以及跨 TypeScript、Java、Kotlin 和 Go 服务的可测试用例编排。适用于引入或重构为端口与适配器架构，或领域逻辑已与 I/O 纠缠的情况。

HeyGen最佳实践——AI数字人视频创作API。

创建默认启用自动扩缩、CloudWatch告警和标记的SageMaker端点（实时、实时缩容至零或异步）。每当即将创建SageMaker端点、编写调用`create_endpoint`的部署代码，或在镜像URI和IAM角色已知后完成部署时，使用此技能。提供用于实时端点的deploy.py、用于通过推理组件缩容至零实例的实时端点的deploy_ic.py，以及用于异步端点（同样缩容至零）的deploy_async.py。这是SageMaker部署工作流的最后一步。切勿在没有这些默认设置的情况下生成裸`create_endpoint`调用——没有自动扩缩或告警的端点只是演示，不是部署。

Pick the right serving container for a SageMaker model deployment and find its current image URI. Use this skill whenever about to deploy a model to a SageMaker endpoint and an image URI needs to be chosen — including when the user says "deploy this LLM", "host this HuggingFace model", "serve this fine-tuned model", "deploy this embedding model", "host a reranker", "serve a sentence-transformers model", or when about to hardcode any container URI in deployment code. HuggingFace-curated Deep Learning Containers are ALWAYS preferred: HuggingFace vLLM (LLMs and generative rerankers), HuggingFace vLLM-Omni (multimodal), TEI (embeddings/cross-encoder rerankers), HF Inference Toolkit (other transformers). Generic images (AWS vLLM, DJL-LMI, SGLang) are used only when no HuggingFace image is compatible — never merely because they carry a newer version. Never hardcode a container URI from memory and never default to TGI. Prevents stale-image failures and wrong-region URIs.

在 Honeycomb 中追踪 Paperclip 运行、请求和完整 CI 套件。用于查找追踪、诊断延迟、设计易读且可折叠的瀑布图、配置自愿启用的 OpenTelemetry，或补充缺失的 span。

构建一个自包含的HTML文件——无需构建步骤、无依赖、无CDN——用于解释、规划、报告、评审、绘制图表、比较或制作供人阅读、分享或试用的原型。当交付物包含颜色、布局、图表、表格或交互，或用户要求提供制品、说明文档、计划文稿、图表或设计变体时使用。不适用于Markdown形式的笔记、仿照已知网站外观的品牌匹配，或生成式画布艺术。

Build and publish a Gradio demo on Hugging Face Spaces for a user-provided LoRA. Use when someone asks to create, generate, ship, or publish a Space, demo, Gradio app, or playground for a LoRA — including LoRAs for Qwen-Image, Qwen-Image-Edit, LTX-Video, Wan, FLUX, SDXL, or other diffusion base models. Also triggers when someone describes a LoRA they trained or hosts on the Hub and wants to share it. Covers picking the right base pipeline and `diffusers` inference recipe, designing a UI tailored to the LoRA's task and inputs (Union/multi-task control, edit, video, image, etc.), respecting model-card recommendations (trigger words, steps, guidance, LoRA scale, example inputs), and shipping to ZeroGPU hardware as a private Space by default.

Trains and fine-tunes vision models for object detection (D-FINE, RT-DETR v2, DETR, YOLOS), image classification (timm models — MobileNetV3, MobileViT, ResNet, ViT/DINOv3 — plus any Transformers classifier), and SAM/SAM2 segmentation using Hugging Face Transformers on Hugging Face Jobs cloud GPUs. Covers COCO-format dataset preparation, Albumentations augmentation, mAP/mAR evaluation, accuracy metrics, SAM segmentation with bbox/point prompts, DiceCE loss, hardware selection, cost estimation, Trackio monitoring, and Hub persistence. Use when users mention training object detection, image classification, SAM, SAM2, segmentation, image matting, DETR, D-FINE, RT-DETR, ViT, timm, MobileNet, ResNet, bounding box models, or fine-tuning vision models on Hugging Face Jobs.

生成可检验假设：从观察中提出假设、设计实验、探索竞争性解释、推导预测并提出机制，用于跨领域科学探究。

Create, edit, design and export Apple Numbers, Keynote and Pages files. Verified and Arabic-safe.

Detect AI-generated images, videos, and audio with identifAI's deepfake detection tools.

Turn a self-contained HTML design into an editable Adobe InDesign (.idml) file. Up to 2 pages free.

49 audio tools: trim, convert, normalize, remove vocals, split stems, transcribe, and more.

Human know-how for AI agents: rules, reasons and checks for design, code, product and writing.

Shares Docker build layers and heavy images between cloud sessions through the owner's ghcr.io. Use when a repo's image build is slow on cloud, when a session pulls or pushes ghcr.io images, or when a heavy tool image should be prebuilt once.

通过提升分辨率、锐度和清晰度来改善图像质量，尤其是截图。非常适合为演示文稿、文档或社交媒体帖子准备图像。

Generate and edit images via OpenAI's GPT Image 2.5 API. Use for requests to create, draw, edit, extend, combine, or remove backgrounds from images. Outputs PNG, JPEG, or WebP files to disk and requires OPENAI_API_KEY.

图像 URL → 供机器人使用的纯文本 OCR。免费试用后为预付 TEST。https://ocr.kymac.co

Process images in TS/JS with Bun's Bun.Image API: resize, crop, rotate, compress, convert (JPEG/PNG/WebP/HEIC/AVIF), thumbnails, LQIP/ThumbHash placeholders, EXIF. Replaces sharp on Bun.

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output should be a bitmap asset rather than repo-native code or vector. Do not use when the task is better handled by editing existing SVG/vector/code-native assets, extending an established icon or logo system, or building the visual directly in HTML/CSS/canvas.

当用户要求通过 OpenAI Image API 生成或编辑图像时使用（例如：生成图像、编辑/修复/蒙版、移除或替换背景、透明背景、产品图、概念艺术、封面或批量变体）；运行捆绑的 CLI（`scripts/image_gen.py`），实时调用需要 `OPENAI_API_KEY`。

Design, critique, polish or overhaul frontend UI: sites, landing pages, dashboards, components, forms, onboarding, empty states. Covers UX review, hierarchy, accessibility, typography, color, motion, design systems. Not backend.

擅长打造真正能带来工作和客户的个人作品集——不只是展示作品，而是创造令人难忘的体验。涵盖开发者作品集、设计师作品集、创意作品集，以及能将访客转化为机会的作品集。适用场景：作品集、个人网站、作品展示、开发者作品集、设计师作品集。

审查某个分支、拉取请求或未提交的更改所引入或导致回归的界面问题，涵盖无障碍性、布局、文案、排版、色彩和UI。

通过一系列简短的高价值问题就计划、想法或设计对用户进行访谈，然后总结方向与下一步。适用于用户希望轻松访谈、轻量澄清或说“interview me”的场景，无需详尽盘问。

Answer a question about what is on the screen or in a recording; solves text, images and audio.

AI data-center design engine: size, validate & lay out Rubin-era data centers. Korea live.

MCP server for Qwen Image 3 AI image generation

Analyze videos: extract frames, transcribe audio, generate storyboard breakdowns.

Video Editing Ai MCP server. Tools: split scenes, generate subtitles, thumbnail data. Built by MEOK

Voice Audio MCP server. Tools: text to speech, list voices, transcribe. Built by MEOK AI Labs.

Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.

Let any LLM watch a video locally — and search everything it has ever watched.

Props, examples, design tokens and page blocks for the Halo React design system.

Google Gemini image generation, editing, and local processing via MCP

Gives LLMs ears. Spectral, harmonic, rhythm, stereo, and structural audio analysis.

Generate design systems: OKLCH colors, fluid type scales, spacing, shape, icon and motion tokens.

Analyze any video (local file or URL) into frames + transcript for Claude Code

An MCP server that converts Figma designs into React code.

InChat Image Viewer MCP - View images inline in AI chat by just providing the file path

MCP server for video transcripts, screenshots, and OCR on YouTube and web videos.

Bidirectional Figma MCP — AI draws UI on Figma canvas, reads designs back

Architecture diagrams for your agent: draw on the live OpenFlowKit canvas or edit diagram files.

Retrieve existing or generated YouTube transcripts with timestamps through Capslane.

Give AI agents full read & write access to Figma — 50+ tools for design automation

MCP server for context-aware royalty-free image search from Pexels and Unsplash

Animated b-roll overlays for short-form video. Paste a script, get an mp4. Free, MIT, local.

Figma MCP server in Rust: plugin bridge, no API token, no rate limits, 73 tools.

Read Lanhu design specs and export fine-grained UI assets with platform-adjusted units