所有 Skills

找到 8742 个 Skills

Skills 列表

huggingface-best

huggingface-best

11Kagent-workflows

Use when the user asks about finding the best, top, or recommended model for a task, wants to know what AI model to use, or wants to compare models by benchmark scores. Triggers on: "best model for X", "what model should I use for", "top models for [task]", "which model runs on my laptop/machine/device", "recommend a model for", "what LLM should I use for", "compare models for", "what's state of the art for", or any question about choosing an AI model for a specific use case. Always use this skill when the user wants model recommendations or comparisons, even if they don't explicitly mention HuggingFace or benchmarks.

huggingface avatarhuggingface
获取
keyword-research

keyword-research

11Kmarketing-seo

Discover keyword opportunities, evaluate metrics and SERPs, and save/tag promising terms.

every-app avatarevery-app
获取
huggingface-vision-trainer

huggingface-vision-trainer

11Ktesting-qa

Trains and fine-tunes vision models for object detection (D-FINE, RT-DETR v2, DETR, YOLOS), image classification (timm models — MobileNetV3, MobileViT, ResNet, ViT/DINOv3 — plus any Transformers classifier), and SAM/SAM2 segmentation using Hugging Face Transformers on Hugging Face Jobs cloud GPUs. Covers COCO-format dataset preparation, Albumentations augmentation, mAP/mAR evaluation, accuracy metrics, SAM segmentation with bbox/point prompts, DiceCE loss, hardware selection, cost estimation, Trackio monitoring, and Hub persistence. Use when users mention training object detection, image classification, SAM, SAM2, segmentation, image matting, DETR, D-FINE, RT-DETR, ViT, timm, MobileNet, ResNet, bounding box models, or fine-tuning vision models on Hugging Face Jobs.

huggingface avatarhuggingface
获取
huggingface-trackio

huggingface-trackio

11Kbackend-api

Track and visualize ML training experiments with Trackio. Use when logging metrics during training (Python API), firing alerts for training diagnostics, or retrieving/analyzing logged metrics (CLI). Supports real-time dashboard visualization, alerts with webhooks, HF Space syncing, and JSON output for automation.

huggingface avatarhuggingface
获取
huggingface-community-evals

huggingface-community-evals

11Kproductivity

Run evaluations for Hugging Face Hub models using inspect-ai and lighteval on local hardware. Use for backend selection, local GPU evals, and choosing between vLLM / Transformers / accelerate. Not for HF Jobs orchestration, model-card PRs, .eval_results publication, or community-evals automation.

huggingface avatarhuggingface
获取
transformers-js

transformers-js

11Kresearch-knowledge

Use Transformers.js to run state-of-the-art machine learning models directly in JavaScript/TypeScript. Supports NLP (text classification, translation, summarization), computer vision (image classification, object detection), audio (speech recognition, audio classification), and multimodal tasks. Works in browsers and server-side runtimes (Node.js, Bun, Deno) with WebGPU/WASM using pre-trained models from Hugging Face Hub.

huggingface avatarhuggingface
获取
huggingface-llm-trainer

huggingface-llm-trainer

11Kdevops-cloud

Train or fine-tune language and vision models using TRL (Transformer Reinforcement Learning) or Unsloth with Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment. Includes guidance on the TRL Jobs package, UV scripts with PEP 723 format, dataset preparation and validation, hardware selection, cost estimation, Trackio monitoring, Hub authentication, model selection/leaderboards and model persistence. Use for tasks involving cloud GPU training, GGUF conversion, or when users mention training on Hugging Face Jobs without local GPU setup.

huggingface avatarhuggingface
获取
huggingface-papers

huggingface-papers

11Kresearch-knowledge

Look up and read Hugging Face paper pages in markdown, and use the papers API for structured metadata such as authors, linked models/datasets/spaces, Github repo and project page. Use when the user shares a Hugging Face paper page URL, an arXiv URL or ID, or asks to summarize, explain, or analyze an AI research paper.

huggingface avatarhuggingface
获取
huggingface-datasets

huggingface-datasets

11Kagent-workflows

Use this skill for Hugging Face Dataset Viewer API workflows that fetch subset/split metadata, paginate rows, search text, apply filters, download parquet URLs, and read size or statistics.

huggingface avatarhuggingface
获取
huggingface-gradio

huggingface-gradio

11Kbackend-api

Build Gradio web UIs and demos in Python. Use when creating or editing Gradio apps, components, event listeners, layouts, or chatbots.

huggingface avatarhuggingface
获取
ml-pipeline

ml-pipeline

11Kdevops-cloud

设计并实现生产级机器学习管道基础设施:使用 MLflow 或 Weights & Biases 配置实验跟踪,创建 Kubeflow 或 Airflow DAG 进行训练编排,使用 Feast 构建特征存储模式,部署模型注册表,并自动化重新训练和验证工作流。在构建 ML 管道、编排训练工作流、自动化模型生命周期、实现特征存储、管理实验跟踪系统、设置 DVC 进行数据版本控制、调整超参数或配置 MLOps 工具(如 Kubeflow、Airflow、MLflow 或 Prefect)时使用。

jeffallan avatarjeffallan
获取
spark-engineer

spark-engineer

11Kagent-workflows

在编写Spark作业、调试性能问题或为Apache Spark应用程序、分布式数据处理管道或大数据工作负载配置集群设置时使用。调用以编写DataFrame转换、优化Spark SQL查询、实现RDD管道、调优shuffle操作、配置执行器内存、处理.parquet文件、处理数据分区或构建结构化流分析。

jeffallan avatarjeffallan
获取
vue-expert-js

vue-expert-js

11Ktesting-qa

创建 Vue 3 组件、构建原生 JS 组合式函数、配置 Vite 项目,并使用纯 JavaScript(无 TypeScript)设置路由和状态管理。生成带有 @typedef、@param 和 @returns 注解的 JSDoc 类型代码,无需 TS 编译器即可实现完整类型覆盖。适用于仅使用 JavaScript(无 TypeScript)构建 Vue 3 应用、项目需要基于 JSDoc 的类型提示、从 Vue 2 Options API 迁移到 Composition API(JS 版)、团队偏好原生 JavaScript 和 .mjs 模块,或需要快速原型开发而无需 TypeScript 配置的场景。

jeffallan avatarjeffallan
获取
graphql-architect

graphql-architect

11Ksecurity

用于设计 GraphQL schema、实现 Apollo Federation 或构建实时订阅。调用场景包括 schema 设计、使用 DataLoader 的解析器、查询优化、联邦指令。

jeffallan avatarjeffallan
获取
chaos-engineer

chaos-engineer

11Kdevops-cloud

设计混沌实验、创建故障注入框架,并组织分布式系统的游戏日演练——生成运行手册、实验清单、回滚流程及事后复盘模板。在需要设计混沌实验、实施故障注入框架或开展游戏日演练时使用。适用于混沌实验、韧性测试、爆炸半径控制、游戏日、反脆弱系统、故障注入、Chaos Monkey、Litmus Chaos等场景。

jeffallan avatarjeffallan
获取
swift-expert

swift-expert

11Ktesting-qa

构建 iOS/macOS/watchOS/tvOS 应用程序,实现 SwiftUI 视图和状态管理,设计面向协议的架构,处理 async/await 并发,实现 actor 保证线程安全,并调试 Swift 特定问题。在构建使用 Swift 5.9+、SwiftUI 或 async/await 并发的 iOS/macOS 应用程序时使用。适用于面向协议编程、SwiftUI 状态管理、actor、服务器端 Swift、UIKit 集成、Combine 或 Vapor。

jeffallan avatarjeffallan
获取
feature-forge

feature-forge

11Kagent-workflows

进行结构化的需求研讨会,生成功能规格说明、用户故事、EARS格式的功能需求、验收标准和实施检查清单。在定义新功能、收集需求或编写规格说明时使用。适用于功能定义、需求收集、用户故事、EARS格式规格、产品需求文档、验收标准或需求矩阵。

jeffallan avatarjeffallan
获取
fine-tuning-expert

fine-tuning-expert

11Kprompting-reasoning

用于微调LLM、训练自定义模型或针对特定任务调整基础模型。调用以配置LoRA/QLoRA适配器、准备JSONL训练数据集、设置微调运行超参数、适配器训练、迁移学习、使用Hugging Face PEFT进行微调、OpenAI微调、指令微调、RLHF、DPO,或量化并部署微调模型。触发词包括:LoRA、QLoRA、PEFT、finetuning、fine-tuning、adapter tuning、LLM training、model training、custom model。

jeffallan avatarjeffallan
获取
mcp-developer

mcp-developer

11Kmcp-integrations

在构建、调试或扩展将AI系统与外部工具和数据源连接的MCP服务器或客户端时使用。调用以实现工具处理器、配置资源提供程序、设置stdio/HTTP/SSE传输层、使用Zod或Pydantic验证模式、调试协议合规性问题,或使用TypeScript或Python SDK搭建完整的MCP服务器/客户端项目。

jeffallan avatarjeffallan
获取
legacy-modernizer

legacy-modernizer

11Ktesting-qa

设计增量迁移策略,识别服务边界,生成依赖关系图和迁移路线图,并为老旧代码库生成API外观设计。适用于现代化遗留系统、实施绞杀者模式或分支抽象、分解单体应用、升级框架或语言、在不中断业务运营的情况下减少技术债务。

jeffallan avatarjeffallan
获取
the-fool

the-fool

11Kagent-workflows

当需要运用结构化批判性思维挑战想法、计划、决策或提案时使用。调用此技能可扮演魔鬼代言人、进行事前验尸、红队测试或审计证据与假设。

jeffallan avatarjeffallan
获取
django-expert

django-expert

11Ktesting-qa

适用于构建 Django Web 应用或基于 Django REST Framework (DRF) 的 REST API。在处理 settings.py、models.py、manage.py 或任何 Django 项目文件时调用。能够创建带合理索引的 Django 模型,使用 select_related/prefetch_related 优化 ORM 查询,构建 DRF 序列化器 (Serializer) 与视图集 (ViewSet),并配置 JWT 身份认证。触发词:Django, DRF, Django REST Framework, Django ORM, Django model, serializer, viewset, Python web。

jeffallan avatarjeffallan
获取
spec-miner

spec-miner

11Ktesting-qa

代码逆向工程专家,专注于从现有代码库中提取需求规格与系统规范。适用于接手遗留代码、无文档系统、接管项目或缺乏文档的老旧代码库。调用此 Skill 可用于梳理代码依赖关系、从源码生成 API 文档、挖掘未记录的业务逻辑、厘清代码实际运行机制,或根据现有实现反推架构文档。触发词:reverse engineer(逆向工程)、old codebase(老旧代码库)、no docs / no documentation(无文档)、figure out how this works(搞清工作原理)、inherited project(接管项目)、legacy analysis(遗留系统分析)、code archaeology(代码考古)、undocumented features(未记录的功能)。

jeffallan avatarjeffallan
获取
vue-expert

vue-expert

11Kfrontend

使用组合式 API 模式构建 Vue 3 组件,配置 Nuxt 3 SSR/SSG 项目,设置 Pinia 存储,搭建 Quasar/Capacitor 移动应用,实现 PWA 功能,并优化 Vite 构建。适用于创建 Vue 3 组合式 API 应用、编写可复用的组合函数、使用 Pinia 管理状态、使用 Quasar 或 Capacitor 构建混合移动应用、配置 Service Worker,或调整 Vite 配置和 TypeScript 集成。

jeffallan avatarjeffallan
获取
想按分类查看?试试 /category/writing-content.