yangwhale/CloseCrab

lecture-video

把「一节课的讲课录音 ＋ 课件网页」做成讲课视频：讲到哪句，页面平滑滚到对应的图并画框/圈/下划线，页面里的动画按时间同步播放，带句级字幕。一节一节做。触发：「生成课程视频」「把这一节做成视频」「给录音配画面」「讲课视频」「lecture video」「做第 N 节视频」。录音是主资产，视频只配画面，不改音频。

Idioma do texto original: chinês

lecture-voice

按讲义一节一节「正式开讲」并把现场语音存成课程录音（lecture-video 的输入）。用讲师口吻在语音通道里讲一节 → 等对方认可 → 把飞书「重播」那段原声存成 mp3、写文字稿、嵌进课件该节开头 → 讲下一节。触发：「开始第二轮讲解」「正式讲课录音」「一节一节讲给我听并录下来」「把音频放到小节前面」「录课」「lecture voice」。

Idioma do texto original: chinês

explainer-distill

全网找把同一个话题讲得最好的高手，蒸馏他们的讲法（开场钩子、比方、反例对照、顺序、图怎么画），借回来让自己的课件／讲义／报告／方案讲稿更有故事性、更好懂、更带得走。流程是：按节派研究 agent 去读点名的高手材料 → 挑能用的落地成图和讲法 → 讲师视角串讲查衔接与比方打架 → 麻瓜视角分段读查「想跳过／没看懂／看反了」→ 补薄弱处 →…

Idioma do texto original: chinês

manim-teaching-figures

用 Manim 给大模型 / AI 基础设施课程画「会动的」教学图 —— 矩阵与线性变换、梯度下降与优化器轨迹、动量与梯度场、loss 地形与鞍点、注意力与 KV cache 的时间演化、高维不可视时的降维表示。产出是挂 loop 的无声 mp4，配静态图与页面图注使用。当用户说「画个动图」「这段用 manim 画」「梯度下降/动量/矩阵变换要动起来」「三蓝一棕那种图」「给课件加动画」，或者一格内容的增量只能来自「时间」或「第三维」时使用。也用于检查已有动画的循环接缝对不对得上。

Idioma do texto original: chinês

tts-generator

Generate speech audio from text using Gemini 3.1 Flash TTS (with emotion control) or Edge TTS fallback. Use when the user says "/tts", "语音合成", "文字转语音", "念出来", "读出来", "生成语音", "帮我念", "说出来", or when asked to convert text to speech audio.

Idioma do texto original: Vários idiomas

wiki

个人知识 Wiki 管理（Quartz v2）。录入资料、查询知识、健康检查。当用户说"录入wiki"、"wiki ingest"、"加到wiki"、"wiki查一下"、"wiki lint"、"/wiki"等关键词时触发。

Idioma do texto original: chinês

browser-cli

Drive the already-logged-in Chrome on Chris's cloudtop from the command line (agent-browser over CDP), instead of chrome-devtools-mcp. Use for ANY interactive browser work on internal/SSO-gated sites — gHire, Buganizer UI, Google Chat, internal dashboards,…

Idioma do texto original: chinês

frontend-slides

Create stunning, animation-rich HTML presentations from scratch or by converting PowerPoint files. Use when the user wants to build a presentation, convert a PPT/PPTX to web, or create slides for a talk/pitch. Helps non-designers discover their aesthetic…

Idioma do texto original: inglês

veo-generator

Generate videos using Google Veo 3.1 on Vertex AI. Use when the user says "生成视频", "做个视频", "generate video", "帮我生成一段视频", "create a video", "视频生成", "text to video", "文生视频", or when you need to create video content.

Idioma do texto original: inglês

memory-lint

全面 memory 重构战役 — 不只 audit，是把"2026-05-22 战役"完整流程做一遍（健康审计 → 集群合并 → 孤儿处理 → 死链修复 → 小爱回归测试 → 备份）。当用户说「memory lint」「/memory lint」「记忆大扫除」「memory 大整理」「memory 重构」时触发。跟 memory-gc skill 互补 — memory-gc 只 audit，memory-lint 是大手术。

Idioma do texto original: chinês

chinese-dictation

Generate Chinese (traditional/simplified) dictation practice — extract words+pinyin from a screenshot, produce an HTML page with audio that reads each word aloud with configurable pause between words. Use when user says…

Idioma do texto original: inglês

lesson-read-aloud

Turn lesson PDFs into a read-aloud study page — extract the text from each PDF, embed the original page screenshots, and attach an English audio player that reads the whole lesson aloud (a real click-to-play progress bar). Built for Hong Kong primary-school…

Idioma do texto original: inglês

study-material

从 PDF/文档/图片中提取全部内容，整理成结构化 HTML 学习资料，并录入 Study Wiki。当用户说"提取文档"、"整理资料"、"学习资料"、"提取PDF"、"帮我整理这个文档"、"extract document"、"study material"，或上传 PDF/文档并要求提取内容时触发。

Idioma do texto original: chinês

music-generator

Generate music using Google Lyria 3 (DeepMind). Use for "生成音乐", "作曲", "写一首曲子", "generate music", "create a song", "compose", "piano piece", "钢琴曲", "make a beat", "生成钢琴曲", "来一首歌".

Idioma do texto original: inglês

imagen-generator

Generate or edit images using Gemini 3 Pro Image (Nano Banana) on Vertex AI. Use for text-to-image ("生成图片", "画一张", "generate image", "帮我画", "create an image", "图片生成") and image-to-image editing ("修图", "P图", "改图", "edit image", "在这张图上加 ...", "基于这张图 ...",…

Idioma do texto original: inglês

deck-builder

Build polished, on-brand slide decks (python-pptx → Google Slides) and richly formatted Google Docs, with a reusable design system, matplotlib charts, AI-generated imagery (Nano Banana), a render-verify loop, and link-stable Drive uploads (PATCH same fileId…

Idioma do texto original: inglês

wechat-reader

Read WeChat Official Account (微信公众号) articles. Trigger when user shares any mp.weixin.qq.com URL, or says "微信文章", "公众号文章", "读一下这个微信链接", "WeChat article". Bypasses WeChat's anti-scraping CAPTCHA by emulating the WeChat in-app browser. MUST be used instead of…

Idioma do texto original: inglês

hk-share-award-tax-dipn38

Compute Hong Kong salaries tax partial exemption for share awards (GSU / RSU / stock options) under DIPN No. 38 paragraphs 43-46 and section 8(1A)(a) of the Inland Revenue Ordinance. Full lifecycle: (1) auto-extract vest data from Morgan Stanley / IRD / IR56B…

Idioma do texto original: inglês

nemo-automodel-distributed-training

Guide for selecting and configuring distributed training strategies in NeMo AutoModel, including FSDP2, Megatron FSDP, DDP, and parallelism settings.

Idioma do texto original: inglês

nemo-automodel-model-onboarding

Guide for onboarding new model architectures into NeMo AutoModel, including architecture discovery, implementation patterns, registration, and validation.

Idioma do texto original: inglês

nemo-automodel-recipe-development

Create and modify NeMo AutoModel training and evaluation recipes, including YAML structure, builders, and execution flow.

Idioma do texto original: inglês

nemo-mbridge-perf-activation-recompute

Validate and use selective and full activation recompute in Megatron Bridge to reduce GPU memory usage at the cost of extra compute.

Idioma do texto original: inglês

nemo-mbridge-perf-cuda-graphs

Validate and use CUDA graph capture in Megatron Bridge, including local full-iteration graphs and Transformer Engine scoped graphs for attention, MLP, and MoE modules.

Idioma do texto original: inglês

nemo-mbridge-perf-expert-parallel-overlap

Validate and use MoE expert-parallel communication overlap in Megatron-Bridge, including overlap_moe_expert_parallel_comm, delay_wgrad_compute, and flex dispatcher backends such as DeepEP and HybridEP.

Idioma do texto original: inglês

nemo-mbridge-perf-memory-tuning

Techniques for reducing peak GPU memory in Megatron Bridge — expandable segments, parallelism resizing, activation recompute, CPU offloading constraints, and common OOM fixes.

Idioma do texto original: inglês

nemo-mbridge-perf-moe-comm-overlap

MoE expert-parallel communication overlap in Megatron Bridge. Covers dispatch/combine overlap, flex dispatcher backends, and expert wgrad scheduling.

Idioma do texto original: inglês

nemo-mbridge-perf-sequence-packing

Validate and use packed sequences and long-context training in Megatron-Bridge, distinguishing offline packed SFT for LLMs from in-batch packing for VLMs, and applying the right CP constraints.

Idioma do texto original: inglês

nemo-mbridge-recipe-recommender

Recommend and customize Megatron Bridge recipes for a user's model, GPU count, and training goal. Indexes library recipes (pretrain/SFT/PEFT) and performance recipes.

Idioma do texto original: inglês

tilegym-converting-cutile-to-triton

Converts cuTile GPU kernels (@ct.kernel) to Triton (@triton.jit). Handles standard in-repo conversion, debugging (cudaErrorIllegalAddress, shape mismatch, numerical mismatch), and mapping cuTile idioms (ct.load/ct.store, ct.Constant, ct.launch) to Triton…

Idioma do texto original: inglês

tilegym-monkey-patch-kernels-to-transformers

Integrate TileGym kernels into Hugging Face `transformers` models by replacing the library's submodule(s) and certain class(es)' implementations, and patching certain class(es)' init/forward/load weight methods prior to instantiating models. Used when the…

Idioma do texto original: inglês

hk-bus

香港智能通勤助手。Google Maps 路线推荐 + KMB/Citybus 实时到站融合。当用户说"几点有车"、"巴士到了吗"、"怎么去公司"、"坐什么车"、"查公交"、"bus ETA"、"下一班车"、"通勤"、"commute"等关键词时触发。

Idioma do texto original: chinês

feishu-user-msg

以 Chris 身份通过飞书 API 发消息到 bot DM 窗口。支持文字和语音。Siri 快捷指令的服务端。

Idioma do texto original: chinês

cc-pages-backup

CC Pages 增量备份到 GitHub private repo

Idioma do texto original: chinês

live-canvas

Real-time SVG whiteboard for teaching. Draw shapes, arrows, text step-by-step while explaining concepts. Viewers watch on any browser. Use when giving visual explanations, drawing architecture diagrams, or teaching algorithms.

Idioma do texto original: chinês

chat-style

聊天平台消息格式化规则。当 Claude Code 通过聊天平台（Discord、飞书等）运行时自动适配消息风格，避免渲染问题。当用户说"聊天风格"、"chat风格"、"消息格式"时触发。

Idioma do texto original: chinês

weather-forecast

查天气预报。香港用香港天文台官方 API（最准、秒回），全球用 Open-Meteo（免 key）。当用户说"天气"、"天气预报"、"今天天气"、"明天天气"、"几天天气"、"会不会下雨"、"weather"、"forecast"、"香港天气"、"某城市天气"等关键词时触发。

Idioma do texto original: chinês

multimodal-explainer

Generate a comprehensive multimodal HTML explainer document with embedded Gemini TTS voice narration segments, structured sections, callouts, diagrams, and tables. Optimized for explaining complex technical topics (PoC briefs, new domain concepts,…

Idioma do texto original: inglês

docx-form-filler

Fill a .docx template (forms, portfolios, application sheets, report cards) with data extracted from a source document (PDF, another docx, chat input) while preserving the template's formatting. Use when the user provides a .docx template plus a data source…

Idioma do texto original: inglês

feishu-mail

飞书企业邮箱管理与收发。当用户说"发邮件"、"收邮件"、"查邮件"、"回复邮件"、"send email"、"check email"、"加个邮箱"、"配置邮箱"、"create mailbox"、"setup email"等关键词时触发。

Idioma do texto original: chinês

memory-gc

记忆系统自动 GC — audit 当前 MEMORY.md / shared/ / disk memory 文件健康度。**Pressure-driven 不是 time-driven** — 没积累就不该清, cold ≠ garbage, 没新东西不要找事清。当用户说「memory audit」「记忆 GC」「记忆清理」「sleep gc」「/memory gc」「/memory audit」「check memory health」时触发。

Idioma do texto original: chinês