MLOps
模型部署、评估与运维
nemo-curator
magnetoid
GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16× faster), quality filtering (30+ heuristics), semantic deduplication, PII redaction, NSFW detection. Scales across GPUs with RAPIDS. Use for preparing high-quality training datasets, cleaning web data, or deduplicating large corpora.
go-create-gorm-model
cristiano-pacheco
Generate Go GORM models following Pingo modular architecture conventions. Use when creating or updating persistence models in internal/modules/<module>/model/, including table mapping, nullable SQL types, timestamps, and relation fields for identity and monitor modules.
distributed-llm-pretraining-torchtitan
magnetoid
Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.
crunch-compete
crunchdao
Use when working with Crunch competitions - setting up workspaces, exploring quickstarters, testing solutions locally, or submitting entries.
standards-alignment
Semprini
Use this skill when the user mentions a named industry standard (BIAN, ISO 20022, FHIR, ACORD, TM Forum, etc.), when modelling a recognised industry domain (banking, payments, insurance, healthcare, telecoms, retail), or when populating the Reference column of any MD-DDL summary table. Also use when the user asks whether a concept already exists in a standard, or when two modelled concepts might map to the same standard object.
sparse-autoencoder-training
magnetoid
Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.
ml-engineer
jgarrison929
Use when building ML pipelines, training models, deploying to production, implementing feature engineering, MLOps workflows, experiment tracking, model monitoring, or any machine learning engineering task.
ask-laravel-architect
NavanithanS
Laravel scaffolding for SQL or Mongo (Official/Jenssegers), SoftDeletes, API standards.
gcp-batch-inference
viktor-ferenczi
Running batch inference on Google Cloud (also known as Vertex AI)
pyside6-mvc
DS-codi
"Use this skill when building Python desktop applications using PySide6 with strict MVC architecture where all UI is defined by .ui files. Covers architecture patterns, controller/model/view separation, signal handling, and .ui file workflows."
fermi-estimation
tumf
Solve user questions with defensible Fermi estimates using explicit assumptions, source-backed inputs, uncertainty ranges, and sensitivity analysis. Use when exact data is unavailable or too slow to collect but the user still needs a quantitative answer for counts, market size, demand, costs, revenue, capacity, timing, or operational scale. Autonomously gather accessible evidence, prefer current primary sources, and drive to a conclusion without asking for approval.
model-usage
Heldinhow
Use CodexBar CLI local cost usage to summarize per-model usage for Codex or Claude, including the current (most recent) model or a full model breakdown. Trigger when asked for model-level usage/cost data from codexbar, or when you need a scriptable per-model summary from codexbar cost JSON.
azure-ai-formrecognizer-java
Tryboy869
"Build document analysis applications with Azure Document Intelligence (Form Recognizer) SDK for Java. Use when extracting text, tables, key-value pairs from documents, receipts, invoices, or building custom document models."
swiftui-view-refactor
dagba
Refactor and review SwiftUI view files for consistent structure, dependency injection, and Observation usage. Use when asked to clean up a SwiftUI view’s layout/ordering, handle view models safely (non-optional when possible), or standardize how dependencies and @Observable state are initialized and passed.
likec4-architecture
timseriakov
Builds and maintains software architecture as code with LikeC4 DSL. Use when requests mention architecture diagrams, C4 context/container/component views, system landscapes, dependency maps, integration maps, or architecture generated from code. Applies to creating new .c4/.likec4 models, updating existing models, validating with LikeC4 CLI, and preparing preview/build/export outputs.
project-development
ken-cavanagh-glean
Design and build LLM-powered projects from ideation through deployment. Use when starting new agent projects, choosing between LLM and traditional approaches, or structuring batch processing pipelines.
jackyshen-design-workshop-outline
mebusw
Use when user asks to "generate workshop outline", "create training agenda", "design course structure", "build workshop schedule", or requests help planning training sessions. Applies MECE structure, TfBR design (4Cs), and VAK-inclusive learning.
nuxt-layers
leeovery
Working with Nuxt layers (base, nuxt-ui, x-ui) that provide shared functionality. Use when understanding layer architecture, importing from layers, extending layer functionality, or creating new layers.
gitlab-ci
grandcamel
"GitLab CI/CD pipeline operations. ALWAYS use this skill when user wants to: (1) view pipeline status, (2) run/trigger pipelines, (3) view/retry jobs, (4) trace job logs, (5) download artifacts, (6) lint CI config."
pal-mcp-expert
mamba-mental
Expert guidance for using the Pal MCP Server (zen-pal-nas). This skill should be used when working with multi-model AI orchestration, tool workflows (chat, thinkdeep, planner, consensus, debug, codereview, precommit, clink), configuration troubleshooting, or optimizing model selection strategies. Activates automatically when user mentions Pal MCP, zen-pal-nas, or specific tool names.
pyside6-qml-views
DS-codi
Use this skill when creating QML view files, designing QML component hierarchies, building layouts, styling QML controls, creating reusable QML components, implementing QML navigation / page switching, or working with QML resources. Covers QML file structure, component patterns, Material/Controls styling, resource management, and common QML idioms for desktop applications.
context-fundamentals
ken-cavanagh-glean
Understand the components, mechanics, and constraints of context in agent systems. Use when designing agent architectures, debugging context-related failures, or optimizing context usage.
qwen3-tts-mlx
AGISeek
Local Qwen3-TTS speech synthesis on Apple Silicon via MLX. Use for offline narration, audiobooks, video voiceovers, and multilingual TTS.
run-claude
LostAbaddon
向助手发出协助请求，适用于需要在隔离的独立环境中执行复杂编程任务、任务规划、文稿撰写的场景。