← all publishers

SuperMate-Ai

@supermate-ai source repo

Report this profile Claim this profile

• 43 skills
• 0 followers
• 3 hours ago last updated

1. ▌
   H3 Prompt Writing · supermate-ai bundle

   Write MiniMax H3 video generation prompts for T2VA, I2VA, FL2VA, L2VA, and Ref2VA. Use when rewriting multimodal requests into H3 prompt structures, composing integrated_multimodal_description, overall_soundscape, and non_diegetic_music, aligning keyframes, or defining reference labels for images, videos, and audio.
   0
   installs
2. ▌
   Co Op Game Intro Generator · supermate-ai bundle

   For users creating a two-player co-op game menu or opening animation. Users provide two player names, a game title, a target visual style, and optional character reference images. The Skill locks identity cues, generates an approval image from a fixed menu framework with coordinated color, buttons, icons, and typography, then uses the approved result to rebuild the character, UI-copy, and event timing instructions for the final video. It outputs a co-op game intro featuring two characters, player cards, and menu interaction motion. Best for game concepts, character-led menus, and social content; not for playable game development, complex multi-page UI, exact brand-logo replication, or generic character-free title sequences.
   0
   installs
3. ▌
   3d Animation Short Generator · supermate-ai bundle

   Create complete stylized 3D animated shorts from a story idea through an ordered production workflow covering project brief, story outline, character and environment cards, standardized shot planning, text or optional pencil storyboards, video-model selection, single-shot generation, assembly, BGM matching, and final review. Use when the user wants an end-to-end narrative animation workflow with strong character consistency, scene continuity, timing, camera, performance, and audio control. Not for single images, simple edits, photorealistic live action, or one standalone clip.
   0
   installs
4. ▌
   Dsh Doc Site Sync · supermate-ai bundle

   Use when publishing, updating, moving, or removing DeepSeek Harness documentation website pages; editing website/docs.ts mappings or navigation; diagnosing a page missing from the VitePress site; fixing projected documentation links; or running the docs:dev, docs:check, and doc-sync workflow after website-content changes.
   0
   installs
5. ▌
   Handdrawn Live Video Generator · supermate-ai bundle

   For creators making surreal short videos that blend rough glowing hand-drawn animation with live-action spaces. Users provide a scene idea, contact object or hand, desired mood, and optional language or style constraints. The Skill clarifies the physical contact, designs continuous morphing, escape route, and delayed handheld chase movement, then writes a reusable 15-second 16:9 video prompt in the user's language. After user confirmation it recommends MiniMax H3 generation and checks contact realism, camera delay, rough glowing stroke texture, and non-horror tone. Best for single-scene creative clips, not polished CG, horror jump scares, plush characters, or multi-scene cuts.
   0
   installs
6. ▌
   Music Video Subtitle Generator · supermate-ai bundle

   For musicians, video creators, and social-media editors producing AI music videos or emotional short films with lyric typography. Users provide music, lyrics, references, characters, typography direction, mood, or target platform. The Skill analyzes beat and vocal timing, separates character, scene, and text references, designs beat-reactive spatial typography, decomposes long works into connected shots, audits prompts, and routes generation for H3 or other video tools. It outputs MV concepts, shot prompts, lyric text plans, and stitching guidance. Best for stylized MVs and subtitle-driven music visuals, not ordinary caption cleanup, licensed IP copying, or fully manual post-production editing.
   0
   installs
7. ▌
   Dsh Translate Docs · supermate-ai bundle

   Manually run the extended DeepSeek Harness bilingual-document workflow, including generated briefings, delegated prose translation, whole-document translation, and scoped pairing verification.
   0
   installs
8. ▌
   Minimalist Product Ad Generator · supermate-ai bundle

   Turn product images and ad requirements into minimalist product ad shorts for e-commerce promotion and product launches. The Skill confirms format and product variants, extracts selling points, writes concise English ad copy, builds product anchors, plans beat-synced typography/storyboards, and generates a clean product film with premium camera language. Not for KOC talking-head ads, general editing, or complex screen demos.
   0
   installs
9. ▌
   Dsh Pre Push Checks · supermate-ai bundle

   Use before pushing, force-pushing, marking ready for review, or claiming checks pass on a deepseek-harness branch, and immediately after gh stack sync publishes rewritten branches, to select the smallest tests and checks that cover the outgoing or just-published diff without reflexively running the full repository suite.
   0
   installs
10. ▌
    Papercraft Stop Motion Explainer · supermate-ai bundle

    For creators explaining science, education, or general knowledge through tactile handmade papercraft visuals. Users provide a topic, core knowledge points, or source material and may specify audience, duration, aspect ratio, and deliverable type. The Skill extracts the learning goal and visual metaphor, proposes creative directions, designs paper characters, layered diorama sets, and props, creates preview concepts plus image and video prompts, and plans storyboards, camera movement, transitions, and sound with staged approvals and review checklists. It outputs a production-ready papercraft stop-motion explainer package, or selected assets such as still prompts, image-series prompts, short-video prompts, or storyboards. Best for cut-paper, pop-up-book, layered diorama, and miniature stop-motion explainers; not for standard 2D cartoons, line doodles, live action, or explainers without a paper-art look.
    0
    installs
11. ▌
    Paper Collage Explainer Generator · supermate-ai bundle

    For creators, educators, and social-video editors who need a tactile paper-collage language for narration, knowledge points, opinions, or abstract topics. Users provide source copy, story beats, or a core concept and may specify aspect ratio, duration, palette, and audio needs. The Skill extracts meaning, proposes visual metaphors, prepares a production plan and storyboard, generates approved halftone collage stills, then creates stop-motion clips with paper movement and tactile sound effects, with optional final assembly. By default it keeps collage SFX and does not add BGM, voiceover, or subtitles unless requested. Best for explainers, viewpoints, story visuals, and social B-roll; not for presenter ads, editable layers, complex typography, or prompt-only tasks.
    0
    installs
12. ▌
    Editing Cordis Compositions · supermate-ai

    Use when creating, changing, or validating a Cordis composition for this harness — writing or editing an agent preset, adding or removing a plugin row, deciding whether something belongs to the host composition or to one session, checking whether a preset you authored actually mounts, or diagnosing a row that mounted but contributed nothing.
    0
    installs
13. ▌
    Colleague Supermate 2 · supermate-ai

    超级搭档 SuperMate——基于 DeepSeek Harness 插件结构组装而成的真智能体。主脑=DeepSeek 文本模型；记忆=Obsidian vault + Hermes 复盘；眼睛=deepseek-eyes/Ollama；画笔=ComfyUI Z-Image；演播室=ComfyUI MiniMax H3 + h3-video-producer；风格=秋芝2046 科技口播；工具=ComfyUI API/ffmpeg/GitHub。当用户需要"超级搭档"端到端完成任务（看图/生成图与视频/记忆沉淀/项目复盘进化）时使用。
    0
    installs
14. ▌
    Dot Skill · supermate-ai bundle

    Unified meta-skill engine for distilling colleague, relationship, or celebrity characters into reusable Skills. | 统一的 meta-skill 引擎，把 colleague、relationship、celebrity 三类对象蒸馏成可复用 Skill。
    0
    installs
15. ▌
    Colleague Supermate · supermate-ai

    超级搭档 SuperMate——基于 DeepSeek Harness 插件结构组装而成的真智能体。主脑=DeepSeek 文本模型；记忆=Obsidian vault + Hermes 复盘；眼睛=deepseek-eyes/Ollama；画笔=ComfyUI Z-Image；演播室=ComfyUI MiniMax H3 + h3-video-producer；风格=秋芝2046 科技口播；工具=ComfyUI API/ffmpeg/GitHub。当用户需要"超级搭档"端到端完成任务（看图/生成图与视频/记忆沉淀/项目复盘进化）时使用。
    0
    installs
16. ▌
    Quark Drive · supermate-ai bundle

    夸克网盘下载/管理 skill——通过 CDP 操作夸克浏览器中的夸克网盘（clouddrive）实现文件下载与查看。⚠️ 必须**在夸克浏览器中运行 SuperMate Harness System**（DSH 跑在夸克标签页里），以复用夸克的登录状态（网盘免重复登录验证）。当需要从夸克网盘下载文件、查看网盘内容、检查传输状态时使用。零 API Key、零凭据——全靠夸克浏览器登录态。
    0
    installs
17. ▌
    Rh Workflow · supermate-ai bundle

    RunningHub（RH）工作流 API 调用——通过 ComfyUI 工作流跑 MiniMax H3 图生视频等任务。当用户提到 RunningHub/RH API、H3 工作流、图生视频、三视图首帧、跑 RH 脚本、或需要把参考图变成视频时使用。核心铁律：参考图绝不裁剪，只缩放到最长边 1280px；提示词充分信任 H3 理解力、写具体剧情与运镜。脚本位于 Local_LLM/RH_Workflow_Api/。
    0
    installs
18. ▌
    Dsh Im Wecom · supermate-ai bundle

    dsh-im-wecom — 企业微信 Bot × DeepSeek Harness 对话 Skill
    0
    installs
19. ▌
    Deepseek Eyes · supermate-ai bundle

    deepseek-eyes：协助文本 AI 模型做图片分析——给 DeepSeek Harness（DSH）里的文本模型长眼睛。Use when DSH needs to analyze an image file or a folder of images (asset entry, prompt management, screenshot understanding, image comparison, OCR-in-image)：first preprocess (convert any non-PNG/JPEG format such as WebP to PNG; downscale long edge to at most 1080px), then call a vision model — a local Ollama model (think off) OR an OpenAI-compatible vision API — and return a structured Chinese description plus retrieval tags that get injected into the text model's context. Prerequisite：the user must have either a local vision model or a vision API key.
    0
    installs
20. ▌
    Hermes Memory · supermate-ai

    Cross-session persistent memory and self-evolution for Harness, adapted from Hermes Agent's memory system. Use at the start of work (load MEMORY.md/USER.md context), at the end of significant tasks (consolidate lessons into memory), when the user mentions memory/persistence/self-improvement/remembering across sessions, when updating or creating skills from experience, or when deciding whether to record or skip information. Hermes 记忆系统方法论（跨会话持久记忆 + 自我进化）。
    0
    installs
21. ▌
    Colleague Musk · supermate-ai bundle

    马斯克, 特斯拉 / SpaceX 参谋长 第一性原理驱动的跨领域技术决策者, MBTI INTJ
    0
    installs
22. ▌
    Doubao Creator · supermate-ai bundle

    豆包（字节跳动 AI）网页操作 skill——通过 CDP 控制夸克浏览器中的豆包网页版（doubao.com）。能力：①看图分析/反推形象描述（普通对话看图，无限制）；②文生视频（Seedance 2.0，10s 竖屏等）；③图生视频（⚠️ 肖像保护：Seedance 不支持上传真实人脸参考图，会拒绝；破局法=先让豆包看图反推形象文字描述 → 再用文字锚定文生视频）。零 API Key，需豆包账号登录态。当需要豆包分析图片、生成视频、或做字节系 AI 任务时使用。
    0
    installs
23. ▌
    Image Analysis · supermate-ai bundle

    图片分析（本地视觉模型）
    0
    installs
24. ▌
    Modelscope API · supermate-ai bundle

    ModelScope（魔塔）免费 API 客户端——零依赖 Python 脚本，OpenAI 协议直连魔塔推理服务：chat 文本对话（Qwen3-235B / DeepSeek-V4-Flash / GLM-4.7-Flash 等）、vision 图片理解（Qwen3-VL-235B）、image 文生图（Z-Image-Turbo / Qwen-Image-2512 / FLUX.2-klein-9B）、models 列出可用模型。当用户要免费 LLM/视觉/生图、或提到魔塔/ModelScope/免费 API 通道时使用。铁律：API Key 只存本地（环境变量或本地 Api_key_modelscope.txt），绝不提交 GitHub。
    0
    installs
25. ▌
    Rh Workflow 9b · supermate-ai

    9B 执行 RunningHub（RH）工作流 API 的唯一行为准则（无漂移宪章）。当你（本地 9B 模型）作为 RH 通道执行器运行时必须先读本宪章：只允许用 run_i2v_workflow.py / run_batch.py 两个入口，参数白名单，禁止写新脚本/编造参数/换主题/手写轮询，先 --check 后提交，失败不重试写 issues 上报。违反任一条 = 违规 = 停止并上报 DeepSeek。
    0
    installs
26. ▌
    Context Handoff · supermate-ai bundle

    上下文迁移/任务交接技能。当对话上下文将满或已满（提示"达到输出token上限"、输出被截断、用户说"交接一下/上下文满了/换新对话/续命/保存进度"）时，把当前工作区状态与待办任务保存为交接文档，并生成"新对话引导消息"，让任务在上下文重新计算后无缝继续。适用于任何工作区，核心原则：产物全部落文件、交接文档用几百 token 压缩摘要替代几万 token 历史、绝不读取旧 session 文件。
    0
    installs
27. ▌
    Supermate Script · supermate-ai

    剧本生成器（连续剧 + 单集短片）：把小说/故事/文案转成完整剧本包（分集规划→角色bible→场景资产→零损耗分镜切割→H3运镜→MiniMax H3 Ref2VA 原生提示词）。当用户提供故事/小说并要"改编成连续剧/AI短剧/多集短片/生成H3分镜提示词/喂给MiniMax或即梦"时使用。
    0
    installs
28. ▌
    Colleague Cameron · supermate-ai bundle

    卡梅隆, 创意导演 热点驱动的视频创意导演，把当下文化符号翻译成 H3 视频提示词 图片/信息 → 创意方向 → H3 视频提示词, MBTI INTJ
    0
    installs
29. ▌
    Comfyui Zimage 9b · supermate-ai

    9B 执行 ComfyUI Z-Image 批量文生图的唯一行为准则（无漂移宪章）。当你（本地 9B 模型，如 qwen3.5 / Ornith-1.5-9B）作为 Z-Image 通道执行器运行时必须先读本宪章：只准用 zimage_generate.py 单一入口，参数白名单，提示词必须纯文本（禁 JSON/markdown/--ar），先 --check 后提交，失败不重试写 issues 上报。违反任一条 = 违规 = 停止并上报 DeepSeek。
    0
    installs
30. ▌
    Quark Qwen Vision · supermate-ai bundle

    夸克浏览器专用视觉 skill——通过 CDP 网页操作调用夸克内置千问（qwen-vl）看图。当需要分析图片内容（识图/描述/OCR/反推提示词/图像对比）且本机已装夸克浏览器时使用。原理：夸克以调试模式（--remote-debugging-port=9222）运行 → CDP 控制千问对话页（p.quark.cn/pcquark-chat/sidebar）→ 模拟粘贴图片 + 提问 → 读取回复。零依赖：无需 Ollama、无需 API Key。前置条件：夸克浏览器正在调试模式运行且千问对话页已打开。
    0
    installs
31. ▌
    Dsh Upgrade Runbook · supermate-ai bundle

    DSH（DeepSeek Harness）官方升级避坑 Runbook + 数据预检脚本。当本机/笔记本/云电脑要升级 dsh（git clone 官方 tag、pnpm install+build），或升级后启动报 "file is not valid JSON" / "stored record does not match its schema"，或一键启动指向旧源码路径起不来时使用。含 preflight.ps1（剥 BOM + JSON 校验）。
    0
    installs
32. ▌
    Lyric Mv Storyboard · supermate-ai bundle

    歌词驱动 MV 分镜生成器：输入歌词（支持 .lrc 时间轴或纯文本+BPM）、曲风、角色设定，自动分段切句、解析情绪、用三层表演体系（行动逻辑/表演节拍/面部时序）设计表演，生成时序分镜表与可直接执行的 MiniMax H3 原生提示词（T2VA/I2VA/Ref2VA），可接线 Z-Image 角色板 + RH Ref2VA/豆包 Seedance 出片并 ffmpeg 拼接。用户说"根据歌词设计 MV 分镜/给歌词生成视频提示词/做歌曲 MV 脚本/这首歌词怎么拍"时使用；歌词排版型 MV 走 music-video-subtitle-generator。
    0
    installs
33. ▌
    Lyric To Storyboard · supermate-ai

    输入歌词、曲风、BPM、角色设定，自动生成时序化视频分镜脚本。内置三层表演导戏体系，输出分镜表格+AI视频提示词，适配MiniMax H3、LTX‑2.5多模态模型，可用于MV、剧情短片脚本创作。
    0
    installs
34. ▌
    Supermate Code Review · supermate-ai

    SuperMate 项目专属代码审查与交付规范。用于：改动 ComfyUI 节点包/工作流/发布物之前的自我审查、完成改动后的独立复核、以及交付前的验证。确保命名一一对应、无回归、资源安全、输出质量达标。
    0
    installs
35. ▌
    Visual File Organizer · supermate-ai bundle

    视觉文件整理——批量把文件夹里的图片交给夸克千问（qwen-vl）分析，按「内容总结(≤6字)+风格+文件类型」自动重命名。当用户希望整理图片文件夹、批量智能命名、按内容给图片起名时使用。原理：夸克以调试模式运行 → CDP 控制千问对话页 → 逐张模拟粘贴图片+提问「内容|风格」→ 以问题文本为锚点截取本次回复 → 用「内容+风格+扩展名」重命名。零 API Key。
    0
    installs
36. ▌
    Lyric To Storyboard Lrc · supermate-ai

    输入歌词（支持 .lrc 文件/文本，含 [mm:ss.xx] 精确时间标签）、曲风、BPM、角色设定，自动解析 LRC 时间轴或按 BPM 估算，切分段落、解析情绪，用三层表演导戏体系（行动逻辑/表演节拍/面部时序）生成时序分镜表 + MiniMax H3 原生视频提示词。可用于 MV、剧情短片脚本创作。
    0
    installs
37. ▌
    Video Production Studio · supermate-ai bundle

    通用视频制作流水线（短剧/漫剧/口播/广告/科普/MV/产品演示等）：从点子、小说、文案或现成剧本出发，经五份创作文档（剧本/视觉设定/分镜/图片提示词/视频提示词）产出可直接生产的 MiniMax H3 原生提示词，按"预览→明确确认→生产"闸门调用本地通道（ComfyUI Z-Image、RunningHub H3 Ref2VA/I2V、本地 H3、豆包 Seedance）生成图片与视频，再合成交付。用户说"做一个视频/拍短剧/把小说改编成短剧/写剧本分镜/H3 提示词/生成并合成视频"或提交剧本要求进入制作时使用；口播带货专案走 h3-video-producer，纯图片风格化走 zimage。
    0
    installs
38. ▌
    Multi Agent Orchestrator · supermate-ai

    在线多智能体对话总指挥。DeepSeek 主脑通过 CDP 调度夸克浏览器里的千问（p.quark.cn/pcquark-chat/sidebar）与豆包（doubao.com/chat）等在线智能体，并行回答同一问题、收集数据，再提取结构化结果并用本地数据验证。支持文字/图片附件；视频走抽帧、音频暂不支持（转文字替代）。当用户要"让千问/豆包一起回答"、需要在线智能体收集资料/多视角答案、或要对比多家AI观点时使用。核心纪律：在线智能体只当"候选假设生成器"，最终结论必须经本地数据/回测裁决。
    0
    installs
39. ▌
    H3 Video Producer · supermate-ai bundle

    用本地 ComfyUI 的 MiniMax H3 工作流从零生产完整视频（口播/科普/产品演示/教学），无需任何云端视频 API。流程：接收视频需求 → 写文案与分镜（三模式：全屏主持人/小头像/纯画外音）→ 按 h3-prompt-writing 规范生成 H3 分段提示词 → 本地 ComfyUI 逐段生成（链式衔接+智子等角色身份锚定+尾帧质检+卸载视觉模型）→ ffmpeg 合成（硬切、正圆羽化小窗、图表缓慢推镜、白字字幕+核心词黄高亮、BGM 避让、结尾淡出）→ 校验交付。Use when the user wants to make a video with the local H3 pipeline (e.g. "用本地 H3 做个口播视频", "做科普视频", "视频分镜合成").
    0
    installs
40. ▌
    Dsh Deepseek Vision Bridge · supermate-ai bundle

    dsh-deepseek-vision-bridge — DeepSeek 原生看图桥（Harness GUI 直接收图）
    0
    installs
41. ▌
    Supermate Harness Launcher · supermate-ai bundle

    一键启动 DeepSeek Harness + 夸克浏览器（千问悬浮侧边栏）。双击入口后自动：① 启动 Harness 网页服务（可配置命令，未运行才启动）→ ② 确保夸克浏览器以调试端口 9222 运行 → ③ 打开 Harness 界面与千问侧边栏页 → ④ 截图定位夸克原生"问AI"按钮并点击，弹出千问悬浮侧边栏面板。跨机型/分辨率/DPI 自适应（位置比例不变、像素变化，用截图+颜色扫描定位）。当用户想"用夸克打开 Harness/千问侧边栏/一键进入工作环境"时使用。
    0
    installs
42. ▌
    Brand Promo Video Generator · supermate-ai bundle

    For marketers and creators producing promotional content for brands, products, websites, apps, shops, or personal projects. Users provide logos, product images, interface screenshots, official links, or other verifiable assets and confirm duration, aspect ratio, audience, and campaign focus. The Skill organizes brand facts and asset provenance, selects a narrative direction, plans precise beats and shots, generates needed imagery, video, voiceover, or music, and completes assembly and pre-delivery review. It outputs a promotional short that highlights product capabilities, use cases, and a call to action. Best for launches, website showcases, and social promotion; not for imitating real brand marks without authorized assets, inventing product claims, or producing long-form narrative films.
    0
    installs
43. ▌
    Lyric To Storyboard Face Lock · supermate-ai

    输入歌词（支持 .lrc 时间轴）、曲风、BPM、角色设定，生成时序分镜表 + MiniMax H3 原生提示词。内置可执行防变脸流程：characterRef → 最小识别锚点 → 可选 Z-Image 角色板 → Ref2VA 锁身份 + 参考控制边界（may_control/must_not_control）+ 剪影/溯源/三距测试，跨镜跨段人物外观零漂移。
    0
    installs