6,089 件 ・ 関連度順
概要と使いどころ
Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.
日本語の概要は準備中です。原文の説明を表示しています。
Official NVIDIA-authored guidance for PhysicsNeMo ShardTensor domain parallelism — integrate domain parallelism into training/inference scripts (new or existing) with DDP or FSDP2, write and register shard patches to enable new layers/ops, and bootstrap multi-GPU correctness tests. Use when working with ShardTensor, scatter_tensor, domain parallelism, sequence/spatial sharding, ring attention, DeviceMesh + DDP/FSDP2 hybrid parallelism, or physicsnemo.domain_parallel. Do NOT use for generic PyTorch DDP/FSDP setup without domain parallelism, picking a PhysicsNeMo model or example (use physicsnemo-discover), or non-distributed training questions.
日本語の概要は準備中です。原文の説明を表示しています。
Build, deploy, run, monitor, tune, and stop Simulink Real-Time applications. Use when the user wants to build a Simulink Real-Time model, load/start/stop an `.mldatx` on a target, set stop time or start options, drive root inports/playback/parameter stimulation, inspect installed or running apps, tune parameters with `getparam`/`setparam`, read ad hoc signal values with `getsignal`, manage parameter sets, control recording, or choose between file logging, live streaming, and file-log import. Covers `slbuild`, `install`, `load`, `start`, `stop`, `setStopTime`, `tg.Stimulation`, `startRecording`/`stopRecording`, `tg.FileLog.import`, `slrealtime.exportRun`, `getparam`/`setparam`, `getsignal`, and installed app management for Speedgoat targets and Simulink Real-Time Linux devices.
日本語の概要は準備中です。原文の説明を表示しています。
Writes the character-performance layer of a video prompt as behavior under pressure, not displayed emotion — objective, obstacle, tactics, beats, subtext, listening, body/status/proxemics, and mandatory eye life. Produces a reusable 150–220-word acting master profile per character plus a per-scene rewrite of it, and a locked voice prompt. Use whenever a prompt needs acting, performance, or emotion direction; whenever characters read wooden, dead-eyed, or 'AI'; whenever a character must stay themselves across many shots; or when the user asks for character behavior, mannerisms, tics, a gait, or how someone reacts. Pairs with higgsfield-facs (muscle-level AU codes) and higgsfield-seedance (the prompt the paragraph goes into).
日本語の概要は準備中です。原文の説明を表示しています。
Use when the user wants to train a reusable face-faithful Soul Character identity on Higgsfield. One-time training returns a reference_id passed to Soul models in higgsfield-generate via `--soul-id`. Use this skill whenever: "train soul character", "create my Soul", "train my face", "make my digital twin", "build me an avatar", "create a character of me", "set up identity for video", "I want my face in generated images", "face-faithful character", "reusable character", "brand character training", "build a character from photos", "soul training", "soul-id create". After training, logs the soul_reference_id to `brand/assets.md`. NOT for: one-shot face swaps (use higgsfield-generate with --image), fictional/non-photo avatars (use higgsfield-generate with a prompt). Requires Higgsfield CLI + Basic plan or higher.
日本語の概要は準備中です。原文の説明を表示しています。
メモや文字起こしなどの素材を、読者が理解しやすい順序の記事に組み立てます。導入案や段落の役割を対話で確かめ、合意した文章を別ファイルに保存するスキル。
作業の依存関係を整理し、独立した調査・実装・検証を並行して進めるためのスキルです。変更先の競合を避け、各作業の完了状況と確認結果を表にまとめます。
複数のAIエージェントに開発や調査を分担させ、dmuxの画面で並行して進める手順をまとめます。担当範囲の切り分け、作業場所の分離、成果の統合を支援します。
Guides authoring of high-quality YARA-X detection rules for malware identification. Use when writing, reviewing, or optimizing YARA rules. Covers naming conventions, string selection, performance optimization, migration from legacy YARA, and false positive reduction. Triggers on: YARA, YARA-X, malware detection, threat hunting, IOC, signature, crx module, dex module.
日本語の概要は準備中です。原文の説明を表示しています。
Transform sequential Python code into parallel/concurrent implementations. Use when asked to parallelize Python code, improve code performance through concurrency, convert loops to parallel execution, or identify parallelization opportunities. Handles CPU-bound (multiprocessing), I/O-bound (asyncio, threading), and data-parallel (vectorization) scenarios.
日本語の概要は準備中です。原文の説明を表示しています。
Japanese (and other non-Hollywood) screenwriting methods (日本编剧方法/原创剧本实战) — distilled from ten Japanese directors and writers in ゼロからの脚本術; Kosawa Ryota (three acts as 起承转合, speed by jumping), Uchida Kenji (beautiful structure, time-axis tricks), Miki Satoshi (small-material notebook, theme emerges later), Sono Sion, Omiya Ellie (dialogue from inner monologue), Kakei Masaya ("if + moreover"), Fukuda Yuichi (character = actor + flaw, audience as tsukkomi), Yokohama Satoko, Takahashi Izumi ("pain"), Yukisada Isao (situation-born originals); plus Arai Haruhiko (IP adaptation, "write what you're ashamed of"). Neighbours; sw-korean-french-screenwriting, ozu-screenplay-style. Use when writing an original Japanese-style script or TV drama, working from small material and fragments, choosing between structure-first and character-first methods, building a Japanese 企画書, designing comedy tsukkomi systems, or when Hollywood models feel wrong for a quiet, daily-life, non-goal-driven story.
Create and control VFX in Unreal Engine 5 with Niagara: systems and emitters, modules and the spawn/update stages, exposed User parameters, and spawning or driving effects from Blueprints or C++. Use when building particle effects, NS_/NE_ assets, spawning a Niagara system at runtime, setting User parameters, or when the user mentions Niagara, VFX, or a particle system in Unreal.
日本語の概要は準備中です。原文の説明を表示しています。
Build respawn-safe Roblox character systems around Players, CharacterAdded/CharacterRemoving, Humanoid, HumanoidRootPart, Animator, R6/R15 rigs, movement, animations and markers, death, tools, accessories, ownership, and custom characters. Use when character scripts break after respawn, cache stale Humanoids, control movement or velocity, load AnimationTracks, handle death, equip Tools, modify avatars, or support custom player rigs.
日本語の概要は準備中です。原文の説明を表示しています。
Read and control Aqara smart home devices: plugs, switches, lights, AC, locks, curtains, scenes. Trigger phrases: aqara, aqara hub, aqara lock, aqara curtain.
日本語の概要は準備中です。原文の説明を表示しています。
Creates and manages reusable character profiles (Soul IDs) for consistent facial and stylistic identity across multiple image and video generations. Provides identity-vs-motion prompt separation, character sheet creation workflows, micro-expression direction, and Soul Cast AI actor configuration. Use when the user wants to maintain character consistency across multiple generations, asks about Soul ID, creating reusable characters, or generating consistent people across different scenes and shots.
日本語の概要は準備中です。原文の説明を表示しています。
The ultimate brand asset: a consistent, explorable universe generated by AI. This skill covers creating coherent brand worlds, consistent characters, architectural spaces, and visual universes that serve as the foundation for all marketing content. Traditional brand building creates guidelines. AI world building creates explorable realities. Characters with consistent personalities across thousands of images. Environments that feel like real places. Visual languages so coherent that any new asset feels like it belongs. This is where AI generation meets universe design. The practitioners of this skill don't just create images—they create the DNA that generates infinite consistent images. They build the source code for visual worlds. Use when "world building, brand world, universe, consistent character, character design, visual language, brand universe, style bible, world bible, coherent assets, character sheet, environment design, world-building, universe, character-design, consistency, visual-language, brand-world, style-guide" mentioned.
日本語の概要は準備中です。原文の説明を表示しています。
语音转文字 (ASR) (语音转文字 (ASR / 语音识别 / Speech-to-Text) 应用工程 (从业者视角) — 选型、集成、优化语音转文字能力，尤其面向移动端、低成本、快速识别的场景。覆盖: (a) 引擎/API 地图 — 云端 API(OpenAI gpt-4o-transcribe / Whisper API、Deepgram、AssemblyAI、Google / Azure / AWS Transcribe、讯飞、字节火山、阿里、腾讯、百度) vs 开源模型(Whisper / faster-whisper / whisper.cpp / distil-whisper、NVIDIA NeMo Parakeet / Canary、阿里 FunASR / Paraformer / SenseVoice、Moonshine、Vosk、Kaldi / k2 / icefall) vs 端侧·移动 SDK(whisper.cpp + CoreML / Metal、iOS Speech framework、Android SpeechRecognizer、Picovoice、SenseVoice 端侧); (b) 准确率(WER / CER) × 延迟(RTF / 流式) × 成本 三角权衡与选型决策树; (c) 成本优化 playbook — 端侧免费 / 批量折扣 / VAD 裁静音 / 量化(int8 / ggml) / 蒸馏 / 自托管 break-even; (d) 移动端集成 — 端侧 vs 云、流式 vs 批量、断点检测(endpointing)、隐私 / 离线; (e) 后处理 — 标点 / 数字规整(ITN) / 说话人分离(diarization) / 时间戳。学派分歧: 云 API vs 端侧自托管、通用大模型(Whisper) vs 专用流式(RNN-T / Conformer)、闭源 API vs 开源、准确率派 vs 成本派、英文优先 vs 中文 ASR(FunASR / SenseVoice / 讯飞)。不含: 文字转语音(TTS / 语音合成，方向相反)、声纹识别 / 说话人验证为主业、语音 agent / 对话式 AI、ASR 模型训练科研深水区。) Master OS — automated mastery of 语音转文字 (ASR / 语音识别 / Speech-to-Text) 应用工程 (从业者视角) — 选型、集成、优化语音转文字能力，尤其面向移动端、低成本、快速识别的场景。覆盖: (a) 引擎/API 地图 — 云端 API(OpenAI gpt-4o-transcribe / Whisper API、Deepgram、AssemblyAI、Google / Azure / AWS Transcribe、讯飞、字节火山、阿里、腾讯、百度) vs 开源模型(Whisper / faster-whisper / whisper.cpp / distil-whisper、NVIDIA NeMo Parakeet / Canary、阿里 FunASR / Paraformer / SenseVoice、Moonshine、Vosk、Kaldi / k2 / icefall) vs 端侧·移动 SDK(whisper.cpp + CoreML / Metal、iOS Speech framework、Android SpeechRecognizer、Picovoice、SenseVoice 端侧); (b) 准确率(WER / CER) × 延迟(RTF / 流式) × 成本 三角权衡与选型决策树; (c) 成本优化 playbook — 端侧免费 / 批量折扣 / VAD 裁静音 / 量化(int8 / ggml) / 蒸馏 / 自托管 break-even; (d) 移动端集成 — 端侧 vs 云、流式 vs 批量、断点检测(endpointing)、隐私 / 离线; (e) 后处理 — 标点 / 数字规整(ITN) / 说话人分离(diarization) / 时间戳。学派分歧: 云 API vs 端侧自托管、通用大模型(Whisper) vs 专用流式(RNN-T / Conformer)、闭源 API vs 开源、准确率派 vs 成本派、英文优先 vs 中文 ASR(FunASR / SenseVoice / 讯飞)。不含: 文字转语音(TTS / 语音合成，方向相反)、声纹识别 / 说话人验证为主业、语音 agent / 对话式 AI、ASR 模型训练科研深水区。: top builders' mental models, tool stack, current workflows, jargon, and where to keep up. Trigger this skill when the user works on 语音转文字 (ASR / 语音识别 / Speech-to-Text) 应用工程 (从业者视角) — 选型、集成、优化语音转文字能力，尤其面向移动端、低成本、快速识别的场景。覆盖: (a) 引擎/API 地图 — 云端 API(OpenAI gpt-4o-transcribe / Whisper API、Deepgram、AssemblyAI、Google / Azure / AWS Transcribe、讯飞、字节火山、阿里、腾讯、百度) vs 开源模型(Whisper / faster-whisper / whisper.cpp / distil-whisper、NVIDIA NeMo Parakeet / Canary、阿里 FunASR / Paraformer / SenseVoice、Moonshine、Vosk、Kaldi / k2 / icefall) vs 端侧·移动 SDK(whisper.cpp + CoreML / Metal、iOS Speech framework、Android SpeechRecognizer、Picovoice、SenseVoice 端侧); (b) 准确率(WER / CER) × 延迟(RTF / 流式) × 成本 三角权衡与选型决策树; (c) 成本优化 playbook — 端侧免费 / 批量折扣 / VAD 裁静音 / 量化(int8 / ggml) / 蒸馏 / 自托管 break-even; (d) 移动端集成 — 端侧 vs 云、流式 vs 批量、断点检测(endpointing)、隐私 / 离线; (e) 后处理 — 标点 / 数字规整(ITN) / 说话人分离(diarization) / 时间戳。学派分歧: 云 API vs 端侧自托管、通用大模型(Whisper) vs 专用流式(RNN-T / Conformer)、闭源 API vs 开源、准确率派 vs 成本派、英文优先 vs 中文 ASR(FunASR / SenseVoice / 讯飞)。不含: 文字转语音(TTS / 语音合成，方向相反)、声纹识别 / 说话人验证为主业、语音 agent / 对话式 AI、ASR 模型训练科研深水区。 problems and wants industry-grade thinking, tool selection, or workflow guidance. 触发词：「语音转文字」「语音识别」「asr」「speech to text」「stt」
日本語の概要は準備中です。原文の説明を表示しています。
Routing libraries for React SPAs: React Router v6/v7 (most common), TanStack Router (typed, modern), wouter (minimal). Detect what's installed and apply matching patterns. Lazy loading, navigation guards, typed params. Use this skill to: - Configure routes (declarative or file-based). - Use navigation hooks (useNavigate, useParams, useSearchParams). - Lazy-load routes for code splitting. - Implement protected routes / auth guards. - Type-safe params via Zod or framework's built-ins. Do NOT use this skill for: - General React conventions (see react-conventions). - State management (see react-state-management). - Form handling (see react-forms). - Next.js routing (different model — see nextjs-plugin).
日本語の概要は準備中です。原文の説明を表示しています。
Read and control Aqara smart home devices: plugs, switches, lights, AC, locks, curtains, scenes. Trigger phrases: aqara, aqara hub, aqara lock, aqara curtain.
日本語の概要は準備中です。原文の説明を表示しています。
Drafts a defendant employer's Answer to a workplace harassment complaint. Responds paragraph-by-paragraph to allegations, asserts Faragher-Ellerth and other affirmative defenses, and follows federal/state civil procedure rules. Use when defending employers at the pleadings phase against Title VII, state FEHA, or other employment harassment claims.
日本語の概要は準備中です。原文の説明を表示しています。
Especialista em construir aplicações de voz em tempo real - de agentes de voz em tempo real a apps habilitados para voz. Abrange OpenAI Realtime API, Vapi para agentes de voz, Deepgram para transcrição, ElevenLabs para síntese, LiveKit para infraestrutura em tempo real e fundamentos de WebRTC. Sabe como construir experiências de voz com baixa latência e prontas para produção. Use quando: voice ai, voice agent, speech to text, text to speech, realtime voice.
日本語の概要は準備中です。原文の説明を表示しています。
Treina modelos de linguagem grandes (2B-462B de parâmetros) usando NVIDIA Megatron-Core com estratégias avançadas de paralelismo. Use ao treinar modelos >1B de parâmetros, precisar de máxima eficiência de GPU (47% MFU em H100), ou necessitar de paralelismo tensor/pipeline/sequência/contexto/expert. Framework pronto para produção usado em Nemotron, LLaMA, DeepSeek.
日本語の概要は準備中です。原文の説明を表示しています。
validateSearch, search param validation with Zod/Valibot/ArkType adapters, fallback(), search middlewares (retainSearchParams, stripSearchParams), custom serialization (parseSearch, stringifySearch), search param inheritance, loaderDeps for cache keys, reading and writing search params.
日本語の概要は準備中です。原文の説明を表示しています。
Edit EcoSIM PFT-parameter NetCDF files such as ecosim_pftpar_20260303.nc with the ParEditor class from applications/notebooks/scripts/ParamEditor.py. Use when asked to modify, scale, copy, compare, or delete PFT parameter records for a plant functional type (six-character pfts code such as gr3s43 or ndlf34), update microbe parameter NetCDF values, or maintain the JSONL change log that records past parameter changes. Never modify plant_trait.*.desc files; they are read-only references.
日本語の概要は準備中です。原文の説明を表示しています。