东方仙侠视觉导演：在统一原典下生成与诊断仙侠场景提示词
东方仙侠视觉导演是 SkillHub 社区 Skill（@user_11893326/xianxia-visual-director，v1.0.0，design-media），在继承式主原典下生成、派生、优化并诊断明亮电影感东方仙侠环境提示词，强调可度量神性尺度、五层空间、40–60% 呼吸留白、纪念碑式东方建筑、无栏杆边缘及仙人与服饰一致。覆盖云上仙城、仙界大境原典等路线，支持多种画幅且单条默认 16:9。requires_api_key=false。安装请按 skillhub.md 安装该坐标。
Read More生图助手：用九步流程把 AI 生图从试错变成可复现协作
生图助手是 SkillHub 社区 design-media 类 Skill（@user_7da98a98/image-gen-expert，v1.0.0），在用户需要生成图像或创作指导时激活。它用结构化九步流程串联提示工程、风格选择、多工具支持与版本管理，标签 requires_api_key=false。文章说明定位与核心能力，并给出按 skillhub.cn 安装文档检查 CLI、用 skillhub install 安装到 Cursor/Claude Code/Codex 等 skills 目录的步骤，以及典型对话式用法与适用注意。
Read MoreAI生图前端-移动端：用 Skill 约束 Agent 产出移动端界面概念图
AI生图前端-移动端（@user_3c6cb52e/imagegen-frontend-mobile-lr-rmhz，v1.0.0）是 SkillHub 社区发布的 MIT 技能，requires_api_key=false。SKILL.md 将 Agent 定位为移动产品设计艺术总监，仅生成 App 原生感界面与流程概念图，覆盖引导、认证、首页、聊天、电商、金融等场景，并列出常见 AI 移动端反模式。安装需 skillhub CLI，使用 skillhub install imagegen-frontend-mobile-lr-rmhz --namespace user_3c6cb52e --dir 指向各客户端 skills 目录。适合要多屏一致概念图、不写代码的团队。
Read More天美AI一键生成爆款图与视频：在 Agent 里做营销图与短视频
介绍 SkillHub 技能「天美AI一键生成爆款图与视频」（@user_101bb978/hirui-tianmei-video，v1.7.2）。天美AI由海睿数科出品，支持一键 AI 生图（海报、产品宣传、销售材料等）与生视频（免脚本拍摄剪辑取向）。说明 skillhub CLI 安装、各客户端 skills 目录、requires_api_key 配置注意，以及对话式使用方式与适用场景。
Read MoreGPT Image2：在 Agent 里通过 ClawdChat 调用 OpenAI gpt-image-2 生图
GPT Image2 是 SkillHub 社区 Skill（@user_ef6dcdec/gpt-image2，1.0.0，design-media），通过虾聊 ClawdChat 工具网关调用 OpenAI gpt-image-2。适用于对话里创建或生成图片、精确文字渲染、多元素指令跟随、图生图保人保物及吉卜力/Pixar 等命名风格。标签 requires_api_key=true，需在网关配置 API Key。公开数据：评分 100000，下载 7422。
Read More公众号爆款封面生成器：把文章信息收成可发布的微信封面
公众号爆款封面生成器是 SkillHub 社区 design-media 类 Skill（v0.2.7），输入标题、主题、摘要或参考图，生成公众号封面、首图、头图及主题配图；支持标题入图或无字背景留安全区，并可基于 Logo、人物、商品参考精修构图与裁切。本文介绍其定位、能力与 SkillHub/Cursor 安装方式，并说明 requires_api_key 与典型对话用法。
Read More微信视频号封面制作：用 AI 把选题与素材做成可发布的封面
介绍 SkillHub 上的微信视频号封面制作（@beatra-ai/wechat-channels-cover-maker，v0.1.9，beatra 发布）。该 Skill 使用 AI 将选题、标题、脚本、关键帧、人物照、商品图或参考图制成视频号封面、视频封面与缩略图，强调主体、封面钩子与标题安全区，并可精修为统一账号风格。说明 skillhub CLI 安装步骤、各 Agent 的 skills 目录，以及对话中的典型输入输出。requires_api_key=false，分类 design-media。
Read Moredsh-media-gen: A Unified Image and Video Generation Plugin for DeepSeek Harness
The DeepSeek Harness plugin `dsh-media-gen` integrates image and video generation capabilities into a unified toolset. It supports multiple providers, including OpenAI, DashScope, Volcengine, and Google Veo, and provides a CLI interface, a gallery view, and an automatic retry mechanism. Core tools include text-to-image, text-to-video, and image-to-video generation, with all generated artifacts uniformly stored in the `generate/` directory in the workspace. After installation, restart DSH and use the plugin in a new session. By unifying credential management and asynchronous task handling, the plugin reduces the configuration cost of multimodal development.
Read MoreDeepSeek Harness Plugin: Game Asset Master
The DSH plugin `dsh-game-material-master` integrates Volcano Ark, MiniMax, and local `ffmpeg` to automate the bulk generation of game assets from design sheets. Core features include eight-direction sprite generation, image cutout generation, animation frame sequence creation, and map/skeletal animation generation. The plugin supports chat-driven interaction and can generate transparent PNG assets; it requires configuring the relevant API key and installing `ffmpeg`. The installation command is `dsh plugin --profile web add dsh-game-material-master`.
Read Moretaxue-dsh-artisan: An Integrated Visual Creation Toolchain for DeepSeek Harness
`taxue-dsh-artisan` is an image generation plugin in the DeepSeek Harness ecosystem, designed to streamline visual creation workflows. It integrates prompt reverse-engineering, audit optimization, color palette extraction, and multi-provider image generation. The plugin supports extracting color palettes from reference images using the K-Means algorithm and directly outputs optimized prompts and aspect ratio suggestions. Its workflows support chained tool usage with token optimization and image-to-image control. It supports multiple providers, including Google Gemini, OpenAI, and ByteDance Seedream, and offers an asynchronous background execution mode.
Read Moredsh-image-plugins: Integrating Image Understanding and Generation for DeepSeek Harness
The DeepSeek Harness multimodal plugin, dsh-image-plugins, provides image understanding and generation capabilities for the DSH Web interface. The plugin consists of two independent tools, understand_image and generate_image, which respectively support reading workspace images and returning text descriptions, as well as generating and saving images based on prompts. The plugin supports both OpenAI-compatible endpoints and the native DashScope API, though DashScope image generation requires using the native Model Studio API to avoid 404 errors. After installation, the vision and image functions must be enabled separately via the Web settings panel or configuration file, and the corresponding baseUrl, apiKey, and model must be configured. Keys support both plaintext and environment variables.
Read More