[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"extension-plugin-huggingface-huggingface-datasets-zh-CN":3,"guides-for-huggingface-huggingface-datasets":747,"similar-k17a1t3nm72bs2xf80qmf4k9md86mk26-zh-CN":748},{"_creationTime":4,"_id":5,"children":6,"community":7,"display":9,"evaluation":14,"identity":256,"isFallback":253,"parentExtension":260,"providers":294,"relations":298,"repo":299,"tags":745,"workflow":746},1778690773482.4846,"k17a1t3nm72bs2xf80qmf4k9md86mk26",[],{"reviewCount":8},0,{"description":10,"installMethods":11,"name":12,"sourceUrl":13},"Explore, query, and extract data from any Hugging Face dataset using the Dataset Viewer REST API and npx tooling. Zero Python dependencies — covers split/config discovery, row pagination, text search, filtering, SQL via parquetlens, and dataset upload via CLI.",{"claudeCode":12},"huggingface-datasets","https://github.com/huggingface/skills",{"_creationTime":15,"_id":16,"extensionId":5,"locale":17,"result":18,"trustSignals":237,"workflow":254},1778691013020.0288,"kn75r9j5vhrqspjfck8cy291x986mx9j","en",{"checks":19,"evaluatedAt":205,"extensionSummary":206,"features":207,"nonGoals":213,"promptVersionExtension":217,"promptVersionScoring":218,"purpose":219,"rationale":220,"score":221,"summary":222,"tags":223,"targetMarket":230,"tier":231,"useCases":232},[20,25,28,31,35,38,42,47,50,53,57,61,65,69,72,75,78,81,84,87,91,95,99,103,107,110,114,117,121,124,127,130,133,136,139,143,147,150,153,157,160,163,166,169,172,175,178,181,184,187,191,194,197,201],{"category":21,"check":22,"severity":23,"summary":24},"Practical Utility","Problem relevance","pass","The description clearly states the problem of exploring and extracting data from Hugging Face datasets and highlights the benefits of using the Dataset Viewer API and CLI tooling.",{"category":21,"check":26,"severity":23,"summary":27},"Unique selling proposition","The extension provides direct access to Hugging Face datasets via an API and npx tooling, offering significant value beyond basic LLM capabilities by enabling structured data interaction and SQL queries.",{"category":21,"check":29,"severity":23,"summary":30},"Production readiness","The extension covers the complete lifecycle for interacting with Hugging Face datasets, including exploration, querying, filtering, and uploading, and is integrated with common coding agent tools.",{"category":32,"check":33,"severity":23,"summary":34},"Scope","Single responsibility principle","The plugin focuses specifically on Hugging Face dataset interaction, covering exploration, querying, and extraction, without extending into unrelated domains.",{"category":32,"check":36,"severity":23,"summary":37},"Description quality","The displayed description accurately reflects the plugin's capabilities, highlighting key features like API access, npx tooling, and specific functionalities such as row pagination and SQL querying.",{"category":39,"check":40,"severity":23,"summary":41},"Invocation","Scoped tools","The described API endpoints and CLI commands are specific verb-noun operations (e.g., '/splits', '/rows', 'upload datasets'), indicating well-scoped tools.",{"category":43,"check":44,"severity":45,"summary":46},"Documentation","Configuration & parameter reference","info","While the documentation details API endpoints and their parameters, specific default values for parameters like 'length' or 'offset' for all relevant endpoints are not explicitly listed, though max length is mentioned.",{"category":32,"check":48,"severity":23,"summary":49},"Tool naming","The tools and commands described (e.g., '/splits', '/rows', 'npx @huggingface/hub upload') are descriptive and clearly indicate their function within the Hugging Face dataset domain.",{"category":32,"check":51,"severity":23,"summary":52},"Minimal I/O surface","The documented API endpoints and CLI commands appear to request only necessary parameters for their stated tasks, and the output descriptions suggest focused responses.",{"category":54,"check":55,"severity":23,"summary":56},"License","License usability","The project includes a clear Apache-2.0 license file, indicating permissive open-source usage.",{"category":58,"check":59,"severity":23,"summary":60},"Maintenance","Commit recency","The repository shows recent commits, indicating active maintenance.",{"category":58,"check":62,"severity":63,"summary":64},"Dependency Management","not_applicable","The extension's README states 'Zero Python dependencies' for its core API usage and relies on npx for tooling, implying minimal external runtime dependencies managed by npm/npx rather than explicit dependency management within the skill itself.",{"category":66,"check":67,"severity":23,"summary":68},"Security","Secret Management","The documentation mentions using `HF_TOKEN` and `Authorization: Bearer \u003CHF_TOKEN>`, implying secrets are handled via environment variables, which is a standard practice and not inherently insecure if managed by the user.",{"category":66,"check":70,"severity":23,"summary":71},"Injection","The skills primarily interact with documented APIs and CLI commands, and there's no indication of executing untrusted external code or instructions.",{"category":66,"check":73,"severity":23,"summary":74},"Transitive Supply-Chain Grenades","The extension relies on npx for tooling and documented API interactions, with no evidence of runtime downloads or remote code execution beyond documented package installations.",{"category":66,"check":76,"severity":23,"summary":77},"Sandbox Isolation","The extension interacts with external APIs and local CLI commands via npx, with no indication of attempting to modify files outside its intended scope.",{"category":66,"check":79,"severity":23,"summary":80},"Sandbox escape primitives","No evidence of detached process spawns or deny-retry loops that would indicate sandbox escape attempts.",{"category":66,"check":82,"severity":23,"summary":83},"Data Exfiltration","The extension's purpose is to query datasets; there are no documented instructions for reading and submitting confidential data to third parties.",{"category":66,"check":85,"severity":23,"summary":86},"Hidden Text Tricks","No hidden text tricks or obfuscated instructions were found in the provided README or skill descriptions.",{"category":88,"check":89,"severity":23,"summary":90},"Hooks","Opaque code execution","The extension relies on documented API calls and standard npx tooling, with no indication of obfuscated or minified code execution within hooks.",{"category":92,"check":93,"severity":23,"summary":94},"Portability","Structural Assumption","The skill interacts with Hugging Face Hub APIs and local files via npx, making no assumptions about specific project directory structures beyond standard dataset locations.",{"category":96,"check":97,"severity":23,"summary":98},"Trust","Issues Attention","The repository shows a healthy ratio of closed to open issues within the last 90 days, indicating active maintenance and responsiveness.",{"category":100,"check":101,"severity":23,"summary":102},"Versioning","Release Management","The repository has recent commit activity and a clearly defined Apache-2.0 license, suggesting a stable versioning approach, although explicit versioning in manifests or CHANGELOG is not detailed.",{"category":104,"check":105,"severity":45,"summary":106},"Execution","Validation","While API endpoints and CLI commands are described, there's no explicit mention of schema validation libraries (like Zod or pydantic) being used for input arguments or output sanitization.",{"category":66,"check":108,"severity":23,"summary":109},"Unguarded Destructive Operations","The dataset exploration and querying functions are read-only. Upload functionality is mentioned but implies standard CLI behavior which typically includes confirmation or clear action, and the focus is on data extraction, not destructive operations.",{"category":111,"check":112,"severity":45,"summary":113},"Code Execution","Error Handling","The skill documentation outlines API endpoints and their expected usage, but does not detail specific error handling mechanisms or structured error reporting for the agent.",{"category":111,"check":115,"severity":63,"summary":116},"Logging","The extension's primary function is interacting with external APIs and CLI tools for data retrieval, and it does not appear to perform destructive actions or outbound calls that would necessitate local audit logging.",{"category":118,"check":119,"severity":23,"summary":120},"Compliance","GDPR","The extension focuses on public Hugging Face dataset metadata and content, and any personal data would be within the datasets themselves, which the tool does not process or exfiltrate.",{"category":118,"check":122,"severity":23,"summary":123},"Target market","The extension interacts with Hugging Face Hub, which is a global platform, and the functionality is not geographically or legally restricted.",{"category":92,"check":125,"severity":23,"summary":126},"Runtime stability","The extension relies on standard npx tooling and Hugging Face APIs, which are designed to be cross-platform.",{"category":43,"check":128,"severity":23,"summary":129},"README","A comprehensive README exists, clearly explaining the extension's purpose, installation, usage, and available skills.",{"category":32,"check":131,"severity":23,"summary":132},"Tool surface size","The documentation lists a focused set of API endpoints and CLI commands relevant to dataset interaction, well within the recommended range.",{"category":39,"check":134,"severity":23,"summary":135},"Overlapping near-synonym tools","The tools and commands described are distinct and cover specific functionalities related to dataset exploration and querying, without significant overlap.",{"category":43,"check":137,"severity":23,"summary":138},"Phantom features","All features described in the README, such as row pagination, text search, and SQL via parquetlens, appear to be directly supported by the documented API endpoints and CLI commands.",{"category":140,"check":141,"severity":23,"summary":142},"Install","Installation instruction","Clear installation instructions are provided for Claude Code, Codex, Gemini CLI, and Cursor, including copy-pasteable commands and explanations for authentication.",{"category":144,"check":145,"severity":45,"summary":146},"Errors","Actionable error messages","The README describes API endpoints and their parameters but does not detail specific error messages or remediation steps for the agent.",{"category":104,"check":148,"severity":23,"summary":149},"Pinned dependencies","The extension relies on npx for tooling, which inherently handles dependency versioning and pinning through package.json and lockfiles.",{"category":32,"check":151,"severity":63,"summary":152},"Dry-run preview","The primary functions of this extension are data exploration and querying, which are read-only operations and do not require a dry-run preview.",{"category":154,"check":155,"severity":63,"summary":156},"Protocol","Idempotent retry & timeouts","The extension primarily interacts with external APIs for data retrieval; there are no state-changing operations that would require idempotency or specific retry logic beyond standard HTTP handling.",{"category":118,"check":158,"severity":23,"summary":159},"Telemetry opt-in","There is no mention of telemetry being collected or transmitted by this extension; it appears to be a direct interface to Hugging Face APIs and tooling.",{"category":39,"check":161,"severity":23,"summary":162},"Name collisions","The `huggingface-datasets` skill name is distinct and does not appear to conflict with other Hugging Face skills or built-in agent commands.",{"category":39,"check":164,"severity":63,"summary":165},"Hooks-off mechanism","This is a plugin marketplace offering multiple skills, not a single plugin with hooks that would require a hooks-off mechanism.",{"category":39,"check":167,"severity":63,"summary":168},"Hook matcher tightness","The extension is described as a plugin marketplace with individual skills, and there's no indication of specific hooks with broad matchers.",{"category":66,"check":170,"severity":63,"summary":171},"Hook security","The provided description and README do not indicate the use of hooks within the plugin's functionality.",{"category":88,"check":173,"severity":63,"summary":174},"Silent prompt rewriting","There is no indication that this plugin utilizes UserPromptSubmit hooks or modifies prompts silently.",{"category":66,"check":176,"severity":63,"summary":177},"Permission Hook","The extension does not appear to implement PermissionRequest hooks.",{"category":118,"check":179,"severity":63,"summary":180},"Hook privacy","No hooks are identified that would be used for logging or telemetry.",{"category":111,"check":182,"severity":63,"summary":183},"Hook dependency","The extension does not appear to use custom hooks that would require code execution analysis.",{"category":43,"check":185,"severity":23,"summary":186},"Feature Transparency","Critical functionality, such as data access and querying, is explained in the README, and the skills are clearly delineated.",{"category":188,"check":189,"severity":23,"summary":190},"Convention","Layout convention adherence","The repository follows standard structure for skills and includes necessary plugin manifests, indicating adherence to Claude Code conventions.",{"category":188,"check":192,"severity":63,"summary":193},"Plugin state","This plugin appears to be stateless, primarily interacting with external APIs and not requiring persistent state management within the Claude plugin data directory.",{"category":66,"check":195,"severity":23,"summary":196},"Keychain-stored secrets","The documentation suggests using environment variables for the HF_TOKEN, which is a common and generally acceptable method for secret management, implying it's not stored in plain JSON settings.",{"category":198,"check":199,"severity":23,"summary":200},"Dependencies","Tagged release sourcing","The extension is sourced directly from the Hugging Face GitHub repository, which is a reliable and tagged source.",{"category":202,"check":203,"severity":23,"summary":204},"Installation","Clean uninstall","The extension's primary mechanism is API interaction and npx tooling, which are session-scoped and do not involve background daemons or persistent installations that would survive an uninstall.",1778691012656,"This plugin allows users to explore, query, and extract data from Hugging Face datasets using the Dataset Viewer REST API and npx tooling. It supports discovering dataset splits and configurations, paginating through rows, performing text searches and filters, and querying data via SQL through parquetlens, all without requiring Python dependencies. It also includes functionality for dataset uploads via CLI.",[208,209,210,211,212],"Query Hugging Face datasets via REST API","Paginate rows and discover splits/configs","Search text and apply filters to dataset rows","Extract data using SQL via parquetlens","Upload datasets via npx tooling",[214,215,216],"Modifying Hugging Face dataset content directly via API (only upload supported).","Running complex data transformations or model training within this extension.","Providing a full-fledged Python-based data analysis environment.","3.0.0","4.4.0","To provide seamless access and manipulation of Hugging Face datasets for AI agents, enabling data-driven workflows without requiring complex setup or Python environments.","The extension is well-documented, has a clear purpose, and relies on well-defined APIs and tooling. No critical or warning findings were identified.",98,"High-quality plugin for interacting with Hugging Face datasets, offering robust exploration and extraction capabilities.",[224,225,226,227,228,229],"huggingface","datasets","api","cli","data-exploration","npx","global","verified",[233,234,235,236],"Exploring available Hugging Face datasets for a specific task.","Extracting subsets of data for model training or analysis.","Performing quick text searches within large dataset rows.","Querying dataset content using SQL syntax.",{"codeQuality":238,"collectedAt":240,"documentation":241,"maintenance":244,"security":250,"testCoverage":252},{"hasLockfile":239},false,1778690989255,{"descriptionLength":242,"readmeSize":243},260,9821,{"closedIssues90d":245,"forks":246,"hasChangelog":239,"openIssues90d":247,"pushedAt":248,"stars":249},6,663,4,1778593131000,10482,{"hasNpmPackage":239,"license":251,"smitheryVerified":239},"Apache-2.0",{"hasCi":253,"hasTests":239},true,{"updatedAt":255},1778691013020,{"basePath":257,"githubOwner":224,"githubRepo":258,"locale":17,"slug":12,"type":259},"skills/huggingface-datasets","skills","plugin",{"_creationTime":261,"_id":262,"community":263,"display":264,"identity":269,"parentExtension":272,"providers":273,"relations":288,"tags":290,"workflow":291},1778690773482.4824,"k17es3r8wd37t5rrwqcpp5kwrh86mxx8",{"reviewCount":8},{"description":265,"installMethods":266,"name":268,"sourceUrl":13},"Agent Skills for AI/ML tasks including dataset creation, model training, evaluation, and research paper publishing on Hugging Face Hub",{"claudeCode":267},"huggingface/skills","huggingface-skills",{"basePath":270,"githubOwner":224,"githubRepo":258,"locale":17,"slug":258,"type":271},"","marketplace",null,{"evaluate":274,"extract":282},{"promptVersionExtension":275,"promptVersionScoring":218,"score":276,"tags":277,"targetMarket":230,"tier":231},"3.1.0",95,[278,224,225,279,280,281],"ai-ml","models","research","developer-tools",{"commitSha":283,"marketplace":284,"plugin":286},"HEAD",{"name":268,"pluginCount":285},14,{"mcpCount":8,"provider":287,"skillCount":8},"classify",{"repoId":289},"kd72xwt5xnc0ktc4p7smzfcp3986m959",[278,225,281,224,279,280],{"evaluatedAt":292,"extractAt":293,"updatedAt":292},1778690814090,1778690773482,{"evaluate":295,"extract":297},{"promptVersionExtension":217,"promptVersionScoring":218,"score":221,"tags":296,"targetMarket":230,"tier":231},[224,225,226,227,228,229],{"commitSha":283},{"parentExtensionId":262,"repoId":289},{"_creationTime":300,"_id":289,"identity":301,"providers":302,"workflow":741},1778689536128.5474,{"githubOwner":224,"githubRepo":258,"sourceUrl":13},{"classify":303,"discover":734,"github":737},{"commitSha":283,"extensions":304},[305,318,327,335,343,351,359,367,375,383,388,396,404,412,420,428,471,480,486,492,509,515,522,564,575,594,600,620,632,656,714],{"basePath":270,"description":265,"displayName":268,"installMethods":306,"rationale":307,"selectedPaths":308,"source":317,"sourceLanguage":17,"type":271},{"claudeCode":267},"marketplace.json at .claude-plugin/marketplace.json",[309,312,314],{"path":310,"priority":311},".claude-plugin/marketplace.json","mandatory",{"path":313,"priority":311},"README.md",{"path":315,"priority":316},"LICENSE","high","rule",{"basePath":319,"description":320,"displayName":321,"installMethods":322,"rationale":323,"selectedPaths":324,"source":317,"sourceLanguage":17,"type":259},"skills/huggingface-llm-trainer","Train or fine-tune language models using TRL on Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment. Includes hardware selection, cost estimation, Trackio monitoring, and Hub persistence.","huggingface-llm-trainer",{"claudeCode":321},"inline plugin source from marketplace.json at skills/huggingface-llm-trainer",[325],{"path":326,"priority":316},"SKILL.md",{"basePath":328,"description":329,"displayName":330,"installMethods":331,"rationale":332,"selectedPaths":333,"source":317,"sourceLanguage":17,"type":259},"skills/huggingface-local-models","Use to select models to run locally with llama.cpp and GGUF on CPU, Mac Metal, CUDA, or ROCm. Covers finding GGUFs, quant selection, running servers, exact GGUF file lookup, conversion, and OpenAI-compatible local serving.","huggingface-local-models",{"claudeCode":330},"inline plugin source from marketplace.json at skills/huggingface-local-models",[334],{"path":326,"priority":316},{"basePath":336,"description":337,"displayName":338,"installMethods":339,"rationale":340,"selectedPaths":341,"source":317,"sourceLanguage":17,"type":259},"skills/huggingface-paper-publisher","Publish and manage research papers on Hugging Face Hub. Supports creating paper pages, linking papers to models/datasets, claiming authorship, and generating professional markdown-based research articles.","huggingface-paper-publisher",{"claudeCode":338},"inline plugin source from marketplace.json at skills/huggingface-paper-publisher",[342],{"path":326,"priority":316},{"basePath":344,"description":345,"displayName":346,"installMethods":347,"rationale":348,"selectedPaths":349,"source":317,"sourceLanguage":17,"type":259},"skills/huggingface-papers","Look up and read Hugging Face paper pages in markdown, and use the papers API for structured metadata like authors, linked models, datasets, Spaces, and media URLs when needed.","huggingface-papers",{"claudeCode":346},"inline plugin source from marketplace.json at skills/huggingface-papers",[350],{"path":326,"priority":316},{"basePath":352,"description":353,"displayName":354,"installMethods":355,"rationale":356,"selectedPaths":357,"source":317,"sourceLanguage":17,"type":259},"skills/huggingface-community-evals","Add and manage evaluation results in Hugging Face model cards. Supports extracting eval tables from README content, importing scores from Artificial Analysis API, and running custom evaluations with vLLM/lighteval.","huggingface-community-evals",{"claudeCode":354},"inline plugin source from marketplace.json at skills/huggingface-community-evals",[358],{"path":326,"priority":316},{"basePath":360,"description":361,"displayName":362,"installMethods":363,"rationale":364,"selectedPaths":365,"source":317,"sourceLanguage":17,"type":259},"skills/huggingface-best","Find the best AI model for any task by querying Hugging Face leaderboards and benchmarks. Recommends top models based on task type, hardware constraints, and benchmark scores.","huggingface-best",{"claudeCode":362},"inline plugin source from marketplace.json at skills/huggingface-best",[366],{"path":326,"priority":316},{"basePath":368,"description":369,"displayName":370,"installMethods":371,"rationale":372,"selectedPaths":373,"source":317,"sourceLanguage":17,"type":259},"skills/hf-cli","Execute Hugging Face Hub operations using the hf CLI. Download models/datasets, upload files, manage repos, and run cloud compute jobs.","hf-cli",{"claudeCode":370},"inline plugin source from marketplace.json at skills/hf-cli",[374],{"path":326,"priority":316},{"basePath":376,"description":377,"displayName":378,"installMethods":379,"rationale":380,"selectedPaths":381,"source":317,"sourceLanguage":17,"type":259},"skills/huggingface-trackio","Track and visualize ML training experiments with Trackio. Log metrics via Python API and retrieve them via CLI. Supports real-time dashboards synced to HF Spaces.","huggingface-trackio",{"claudeCode":378},"inline plugin source from marketplace.json at skills/huggingface-trackio",[382],{"path":326,"priority":316},{"basePath":257,"description":10,"displayName":12,"installMethods":384,"rationale":385,"selectedPaths":386,"source":317,"sourceLanguage":17,"type":259},{"claudeCode":12},"inline plugin source from marketplace.json at skills/huggingface-datasets",[387],{"path":326,"priority":316},{"basePath":389,"description":390,"displayName":391,"installMethods":392,"rationale":393,"selectedPaths":394,"source":317,"sourceLanguage":17,"type":259},"skills/huggingface-tool-builder","Build reusable scripts for Hugging Face Hub and API workflows. Useful for chaining API calls, enriching Hub metadata, or automating repeated tasks.","huggingface-tool-builder",{"claudeCode":391},"inline plugin source from marketplace.json at skills/huggingface-tool-builder",[395],{"path":326,"priority":316},{"basePath":397,"description":398,"displayName":399,"installMethods":400,"rationale":401,"selectedPaths":402,"source":317,"sourceLanguage":17,"type":259},"skills/huggingface-gradio","Build Gradio web UIs and demos in Python. Use when creating or editing Gradio apps, components, event listeners, layouts, or chatbots.","huggingface-gradio",{"claudeCode":399},"inline plugin source from marketplace.json at skills/huggingface-gradio",[403],{"path":326,"priority":316},{"basePath":405,"description":406,"displayName":407,"installMethods":408,"rationale":409,"selectedPaths":410,"source":317,"sourceLanguage":17,"type":259},"skills/transformers-js","Run state-of-the-art machine learning models directly in JavaScript/TypeScript for NLP, computer vision, audio processing, and multimodal tasks. Works in Node.js and browsers with WebGPU/WASM using Hugging Face models.","transformers-js",{"claudeCode":407},"inline plugin source from marketplace.json at skills/transformers-js",[411],{"path":326,"priority":316},{"basePath":413,"description":414,"displayName":415,"installMethods":416,"rationale":417,"selectedPaths":418,"source":317,"sourceLanguage":17,"type":259},"skills/huggingface-vision-trainer","Train and fine-tune object detection models (RTDETRv2, YOLOS, DETR and others) and image classification models (timm and transformers models — MobileNetV3, MobileViT, ResNet, ViT/DINOv3) using Transformers Trainer API on Hugging Face Jobs infrastructure or locally. Includes COCO dataset format support, Albumentations augmentation, mAP/mAR metrics, trackio tracking, hardware selection, and Hub persistence.","huggingface-vision-trainer",{"claudeCode":415},"inline plugin source from marketplace.json at skills/huggingface-vision-trainer",[419],{"path":326,"priority":316},{"basePath":421,"description":422,"displayName":423,"installMethods":424,"rationale":425,"selectedPaths":426,"source":317,"sourceLanguage":17,"type":259},"skills/train-sentence-transformers","Train or fine-tune sentence-transformers models across all three architectures: SentenceTransformer (bi-encoder embeddings), CrossEncoder (rerankers), and SparseEncoder (SPLADE). Covers loss selection, hard-negative mining, evaluators, distillation, LoRA, Matryoshka, and Hugging Face Hub publishing.","train-sentence-transformers",{"claudeCode":423},"inline plugin source from marketplace.json at skills/train-sentence-transformers",[427],{"path":326,"priority":316},{"basePath":270,"description":265,"displayName":268,"installMethods":429,"license":251,"rationale":430,"selectedPaths":431,"source":317,"sourceLanguage":17,"type":259},{"claudeCode":268},"plugin manifest at .claude-plugin/plugin.json",[432,434,435,436,439,441,443,445,447,449,451,453,455,457,459,461,463,465,467,469],{"path":433,"priority":311},".claude-plugin/plugin.json",{"path":313,"priority":311},{"path":315,"priority":316},{"path":437,"priority":438},"skills/hf-cli/SKILL.md","medium",{"path":440,"priority":438},"skills/huggingface-best/SKILL.md",{"path":442,"priority":438},"skills/huggingface-community-evals/SKILL.md",{"path":444,"priority":438},"skills/huggingface-datasets/SKILL.md",{"path":446,"priority":438},"skills/huggingface-gradio/SKILL.md",{"path":448,"priority":438},"skills/huggingface-llm-trainer/SKILL.md",{"path":450,"priority":438},"skills/huggingface-local-models/SKILL.md",{"path":452,"priority":438},"skills/huggingface-paper-publisher/SKILL.md",{"path":454,"priority":438},"skills/huggingface-papers/SKILL.md",{"path":456,"priority":438},"skills/huggingface-tool-builder/SKILL.md",{"path":458,"priority":438},"skills/huggingface-trackio/SKILL.md",{"path":460,"priority":438},"skills/huggingface-vision-trainer/SKILL.md",{"path":462,"priority":438},"skills/train-sentence-transformers/SKILL.md",{"path":464,"priority":438},"skills/transformers-js/SKILL.md",{"path":466,"priority":311},".mcp.json",{"path":468,"priority":316},"agents/AGENTS.md",{"path":470,"priority":316},".cursor-plugin/plugin.json",{"basePath":472,"description":473,"displayName":474,"installMethods":475,"rationale":476,"selectedPaths":477,"source":317,"sourceLanguage":17,"type":479},"hf-mcp/skills/hf-mcp","Use Hugging Face Hub via MCP server tools. Search models, datasets, Spaces, papers. Get repo details, fetch documentation, run compute jobs, and use Gradio Spaces as AI tools. Available when connected to the HF MCP server.","hf-mcp",{"claudeCode":267},"SKILL.md frontmatter at hf-mcp/skills/hf-mcp/SKILL.md",[478],{"path":326,"priority":311},"skill",{"basePath":368,"description":481,"displayName":370,"installMethods":482,"rationale":483,"selectedPaths":484,"source":317,"sourceLanguage":17,"type":479},"Hugging Face Hub CLI (`hf`) for downloading, uploading, and managing models, datasets, spaces, buckets, repos, papers, jobs, and more on the Hugging Face Hub. Use when: handling authentication; managing local cache; managing Hugging Face Buckets; running or scheduling jobs on Hugging Face infrastructure; managing Hugging Face repos; discussions and pull requests; browsing models, datasets and spaces; reading, searching, or browsing academic papers; managing collections; querying datasets; configuring spaces; setting up webhooks; or deploying and managing HF Inference Endpoints. Make sure to use this skill whenever the user mentions 'hf', 'huggingface', 'Hugging Face', 'huggingface-cli', or 'hugging face cli', or wants to do anything related to the Hugging Face ecosystem and to AI and ML in general. Also use for cloud storage needs like training checkpoints, data pipelines, or agent traces. Use even if the user doesn't explicitly ask for a CLI command. Replaces the deprecated `huggingface-cli`.",{"claudeCode":267},"SKILL.md frontmatter at skills/hf-cli/SKILL.md",[485],{"path":326,"priority":311},{"basePath":360,"description":487,"displayName":362,"installMethods":488,"rationale":489,"selectedPaths":490,"source":317,"sourceLanguage":17,"type":479},"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.\n",{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-best/SKILL.md",[491],{"path":326,"priority":311},{"basePath":352,"description":493,"displayName":354,"installMethods":494,"rationale":495,"selectedPaths":496,"source":317,"sourceLanguage":17,"type":479},"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.",{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-community-evals/SKILL.md",[497,498,501,503,505,507],{"path":326,"priority":311},{"path":499,"priority":500},"examples/.env.example","low",{"path":502,"priority":500},"examples/USAGE_EXAMPLES.md",{"path":504,"priority":500},"scripts/inspect_eval_uv.py",{"path":506,"priority":500},"scripts/inspect_vllm_uv.py",{"path":508,"priority":500},"scripts/lighteval_vllm_uv.py",{"basePath":257,"description":510,"displayName":12,"installMethods":511,"rationale":512,"selectedPaths":513,"source":317,"sourceLanguage":17,"type":479},"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.\r",{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-datasets/SKILL.md",[514],{"path":326,"priority":311},{"basePath":397,"description":398,"displayName":399,"installMethods":516,"rationale":517,"selectedPaths":518,"source":317,"sourceLanguage":17,"type":479},{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-gradio/SKILL.md",[519,520],{"path":326,"priority":311},{"path":521,"priority":438},"examples.md",{"basePath":319,"description":523,"displayName":321,"installMethods":524,"rationale":525,"selectedPaths":526,"source":317,"sourceLanguage":17,"type":479},"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.",{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-llm-trainer/SKILL.md",[527,528,530,532,534,536,538,540,542,544,546,548,550,552,554,556,558,560,562],{"path":326,"priority":311},{"path":529,"priority":438},"references/gguf_conversion.md",{"path":531,"priority":438},"references/hardware_guide.md",{"path":533,"priority":438},"references/hub_saving.md",{"path":535,"priority":438},"references/local_training_macos.md",{"path":537,"priority":438},"references/reliability_principles.md",{"path":539,"priority":438},"references/trackio_guide.md",{"path":541,"priority":438},"references/training_methods.md",{"path":543,"priority":438},"references/training_patterns.md",{"path":545,"priority":438},"references/troubleshooting.md",{"path":547,"priority":438},"references/unsloth.md",{"path":549,"priority":500},"scripts/convert_to_gguf.py",{"path":551,"priority":500},"scripts/dataset_inspector.py",{"path":553,"priority":500},"scripts/estimate_cost.py",{"path":555,"priority":500},"scripts/hf_benchmarks.py",{"path":557,"priority":500},"scripts/train_dpo_example.py",{"path":559,"priority":500},"scripts/train_grpo_example.py",{"path":561,"priority":500},"scripts/train_sft_example.py",{"path":563,"priority":500},"scripts/unsloth_sft_example.py",{"basePath":328,"description":329,"displayName":330,"installMethods":565,"rationale":566,"selectedPaths":567,"source":317,"sourceLanguage":17,"type":479},{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-local-models/SKILL.md",[568,569,571,573],{"path":326,"priority":311},{"path":570,"priority":438},"references/hardware.md",{"path":572,"priority":438},"references/hub-discovery.md",{"path":574,"priority":438},"references/quantization.md",{"basePath":336,"description":337,"displayName":338,"installMethods":576,"rationale":577,"selectedPaths":578,"source":317,"sourceLanguage":17,"type":479},{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-paper-publisher/SKILL.md",[579,580,582,584,586,588,590,592],{"path":326,"priority":311},{"path":581,"priority":500},"examples/example_usage.md",{"path":583,"priority":438},"references/quick_reference.md",{"path":585,"priority":500},"scripts/paper_manager.py",{"path":587,"priority":500},"templates/arxiv.md",{"path":589,"priority":500},"templates/ml-report.md",{"path":591,"priority":500},"templates/modern.md",{"path":593,"priority":500},"templates/standard.md",{"basePath":344,"description":595,"displayName":346,"installMethods":596,"rationale":597,"selectedPaths":598,"source":317,"sourceLanguage":17,"type":479},"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.",{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-papers/SKILL.md",[599],{"path":326,"priority":311},{"basePath":389,"description":601,"displayName":391,"installMethods":602,"rationale":603,"selectedPaths":604,"source":317,"sourceLanguage":17,"type":479},"Use this skill when the user wants to build tool/scripts or achieve a task where using data from the Hugging Face API would help. This is especially useful when chaining or combining API calls or the task will be repeated/automated. This Skill creates a reusable script to fetch, enrich or process data.",{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-tool-builder/SKILL.md",[605,606,608,610,612,614,616,618],{"path":326,"priority":311},{"path":607,"priority":438},"references/baseline_hf_api.py",{"path":609,"priority":438},"references/baseline_hf_api.sh",{"path":611,"priority":438},"references/baseline_hf_api.tsx",{"path":613,"priority":438},"references/find_models_by_paper.sh",{"path":615,"priority":438},"references/hf_enrich_models.sh",{"path":617,"priority":438},"references/hf_model_card_frontmatter.sh",{"path":619,"priority":438},"references/hf_model_papers_auth.sh",{"basePath":376,"description":621,"displayName":378,"installMethods":622,"rationale":623,"selectedPaths":624,"source":317,"sourceLanguage":17,"type":479},"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.",{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-trackio/SKILL.md",[625,626,628,630],{"path":326,"priority":311},{"path":627,"priority":438},"references/alerts.md",{"path":629,"priority":438},"references/logging_metrics.md",{"path":631,"priority":438},"references/retrieving_metrics.md",{"basePath":413,"description":633,"displayName":415,"installMethods":634,"rationale":635,"selectedPaths":636,"source":317,"sourceLanguage":17,"type":479},"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.",{"claudeCode":267},"SKILL.md frontmatter at skills/huggingface-vision-trainer/SKILL.md",[637,638,640,641,643,645,646,648,649,650,652,654],{"path":326,"priority":311},{"path":639,"priority":438},"references/finetune_sam2_trainer.md",{"path":533,"priority":438},{"path":642,"priority":438},"references/image_classification_training_notebook.md",{"path":644,"priority":438},"references/object_detection_training_notebook.md",{"path":537,"priority":438},{"path":647,"priority":438},"references/timm_trainer.md",{"path":551,"priority":500},{"path":553,"priority":500},{"path":651,"priority":500},"scripts/image_classification_training.py",{"path":653,"priority":500},"scripts/object_detection_training.py",{"path":655,"priority":500},"scripts/sam_segmentation_training.py",{"basePath":421,"description":657,"displayName":423,"installMethods":658,"rationale":659,"selectedPaths":660,"source":317,"sourceLanguage":17,"type":479},"Train or fine-tune sentence-transformers models across `SentenceTransformer` (bi-encoder; dense or static embedding model; for retrieval, similarity, clustering, classification, paraphrase mining, dedup, multimodal), `CrossEncoder` (reranker; pair scoring for two-stage retrieval / pair classification), and `SparseEncoder` (SPLADE, sparse embedding model; for learned-sparse retrieval). Covers loss selection, hard-negative mining, evaluators, distillation, LoRA, Matryoshka, and Hugging Face Hub publishing. Use for any sentence-transformers training task.",{"claudeCode":267},"SKILL.md frontmatter at skills/train-sentence-transformers/SKILL.md",[661,662,664,666,668,670,672,673,675,677,679,681,683,685,687,688,690,692,694,696,698,700,702,704,706,708,710,712],{"path":326,"priority":311},{"path":663,"priority":438},"references/base_model_selection.md",{"path":665,"priority":438},"references/dataset_formats.md",{"path":667,"priority":438},"references/evaluators_cross_encoder.md",{"path":669,"priority":438},"references/evaluators_sentence_transformer.md",{"path":671,"priority":438},"references/evaluators_sparse_encoder.md",{"path":531,"priority":438},{"path":674,"priority":438},"references/hf_jobs_execution.md",{"path":676,"priority":438},"references/losses_cross_encoder.md",{"path":678,"priority":438},"references/losses_sentence_transformer.md",{"path":680,"priority":438},"references/losses_sparse_encoder.md",{"path":682,"priority":438},"references/model_architectures.md",{"path":684,"priority":438},"references/prompts_and_instructions.md",{"path":686,"priority":438},"references/training_args.md",{"path":545,"priority":438},{"path":689,"priority":500},"scripts/mine_hard_negatives.py",{"path":691,"priority":500},"scripts/train_cross_encoder_distillation_example.py",{"path":693,"priority":500},"scripts/train_cross_encoder_example.py",{"path":695,"priority":500},"scripts/train_cross_encoder_listwise_example.py",{"path":697,"priority":500},"scripts/train_sentence_transformer_distillation_example.py",{"path":699,"priority":500},"scripts/train_sentence_transformer_example.py",{"path":701,"priority":500},"scripts/train_sentence_transformer_make_multilingual_example.py",{"path":703,"priority":500},"scripts/train_sentence_transformer_matryoshka_example.py",{"path":705,"priority":500},"scripts/train_sentence_transformer_multi_dataset_example.py",{"path":707,"priority":500},"scripts/train_sentence_transformer_static_embedding_example.py",{"path":709,"priority":500},"scripts/train_sentence_transformer_with_lora_example.py",{"path":711,"priority":500},"scripts/train_sparse_encoder_distillation_example.py",{"path":713,"priority":500},"scripts/train_sparse_encoder_example.py",{"basePath":405,"description":715,"displayName":407,"installMethods":716,"rationale":717,"selectedPaths":718,"source":317,"sourceLanguage":17,"type":479},"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.",{"claudeCode":267},"SKILL.md frontmatter at skills/transformers-js/SKILL.md",[719,720,722,724,726,728,730,732],{"path":326,"priority":311},{"path":721,"priority":438},"references/CACHE.md",{"path":723,"priority":438},"references/CONFIGURATION.md",{"path":725,"priority":438},"references/EXAMPLES.md",{"path":727,"priority":438},"references/MODEL_ARCHITECTURES.md",{"path":729,"priority":438},"references/MODEL_REGISTRY.md",{"path":731,"priority":438},"references/PIPELINE_OPTIONS.md",{"path":733,"priority":438},"references/TEXT_GENERATION.md",{"sources":735},[736],"manual",{"closedIssues90d":245,"description":738,"forks":246,"homepage":739,"license":251,"openIssues90d":247,"pushedAt":248,"readmeSize":243,"stars":249,"topics":740},"Give your agents the power of the Hugging Face ecosystem","https://huggingface.co",[],{"classifiedAt":742,"discoverAt":743,"extractAt":744,"githubAt":744,"updatedAt":742},1778690772996,1778689536128,1778690770714,[226,227,228,225,224,229],{"evaluatedAt":255,"extractAt":293,"updatedAt":255},[],[749,766,802,822,840,860],{"_creationTime":750,"_id":751,"community":752,"display":753,"identity":755,"providers":756,"relations":762,"tags":763,"workflow":764},1778690773482.4841,"k172m7ench1yy64c6nbcx5pmm586mphh",{"reviewCount":8},{"description":369,"installMethods":754,"name":370,"sourceUrl":13},{"claudeCode":370},{"basePath":368,"githubOwner":224,"githubRepo":258,"locale":17,"slug":370,"type":259},{"evaluate":757,"extract":761},{"promptVersionExtension":217,"promptVersionScoring":218,"score":758,"tags":759,"targetMarket":230,"tier":231},99,[224,227,281,760,225,279],"mlops",{"commitSha":283},{"parentExtensionId":262,"repoId":289},[227,225,281,224,760,279],{"evaluatedAt":765,"extractAt":293,"updatedAt":765},1778690967596,{"_creationTime":767,"_id":768,"community":769,"display":770,"identity":776,"providers":780,"relations":793,"tags":797,"workflow":798},1778693655824.435,"k17046jensvb82jshmksyrt4ps86nxe5",{"reviewCount":8},{"description":771,"installMethods":772,"name":774,"sourceUrl":775},"访问 Microsoft 官方文档、API 参考和代码示例，涵盖 Azure、.NET、Windows 等。",{"claudeCode":773},"microsoft-docs","Microsoft Learn MCP 服务器","https://github.com/MicrosoftDocs/mcp",{"basePath":270,"githubOwner":777,"githubRepo":778,"locale":779,"slug":778,"type":259},"MicrosoftDocs","mcp","zh-CN",{"evaluate":781,"extract":789},{"promptVersionExtension":217,"promptVersionScoring":218,"score":782,"tags":783,"targetMarket":230,"tier":231},100,[784,785,786,787,226,788,227],"microsoft","documentation","azure","net","rag",{"commitSha":283,"license":790,"plugin":791},"CC-BY-4.0",{"mcpCount":8,"provider":287,"skillCount":792},3,{"parentExtensionId":794,"repoId":795,"translatedFrom":796},"k17cyy5a1yyy3kgamhnat6m15x86n6r3","kd7a5v3pbwtsn0qajecay1jdcs86nn0z","k1735x1w1m3nbt4dfnr954mjsd86mkhc",[226,786,227,785,784,787,788],{"evaluatedAt":799,"extractAt":800,"updatedAt":801},1778693508577,1778693447172,1778693655824,{"_creationTime":803,"_id":804,"community":805,"display":806,"identity":809,"providers":810,"relations":818,"tags":819,"workflow":820},1778690773482.4834,"k179sm2kkyd7r7nz9jsx62jm9x86mw4a",{"reviewCount":8},{"description":345,"installMethods":807,"name":808,"sourceUrl":13},{"claudeCode":346},"Hugging Face Papers",{"basePath":344,"githubOwner":224,"githubRepo":258,"locale":17,"slug":346,"type":259},{"evaluate":811,"extract":817},{"promptVersionExtension":217,"promptVersionScoring":218,"score":782,"tags":812,"targetMarket":230,"tier":231},[224,813,814,815,280,816],"papers","arxiv","ai","metadata",{"commitSha":283,"license":251},{"parentExtensionId":262,"repoId":289},[815,814,224,816,813,280],{"evaluatedAt":821,"extractAt":293,"updatedAt":821},1778690901306,{"_creationTime":823,"_id":824,"community":825,"display":826,"identity":828,"providers":829,"relations":836,"tags":837,"workflow":838},1778690773482.4844,"k1785s3263hzxs6ne1nsbzas5186m86v",{"reviewCount":8},{"description":377,"installMethods":827,"name":378,"sourceUrl":13},{"claudeCode":378},{"basePath":376,"githubOwner":224,"githubRepo":258,"locale":17,"slug":378,"type":259},{"evaluate":830,"extract":835},{"promptVersionExtension":217,"promptVersionScoring":218,"score":758,"tags":831,"targetMarket":230,"tier":231},[760,832,833,224,834,227],"experiment-tracking","monitoring","python",{"commitSha":283},{"parentExtensionId":262,"repoId":289},[227,832,224,760,833,834],{"evaluatedAt":839,"extractAt":293,"updatedAt":839},1778690988981,{"_creationTime":841,"_id":842,"community":843,"display":844,"identity":846,"providers":847,"relations":856,"tags":857,"workflow":858},1778690773482.483,"k172w5kkdn7117xpqyyhcqsww186n17b",{"reviewCount":8},{"description":329,"installMethods":845,"name":330,"sourceUrl":13},{"claudeCode":330},{"basePath":328,"githubOwner":224,"githubRepo":258,"locale":17,"slug":330,"type":259},{"evaluate":848,"extract":855},{"promptVersionExtension":217,"promptVersionScoring":218,"score":758,"tags":849,"targetMarket":230,"tier":231},[850,851,852,853,224,854],"llm","local-models","llama-cpp","gguf","ml-ops",{"commitSha":283},{"parentExtensionId":262,"repoId":289},[853,224,852,850,851,854],{"evaluatedAt":859,"extractAt":293,"updatedAt":859},1778690852474,{"_creationTime":861,"_id":862,"community":863,"display":864,"identity":866,"providers":867,"relations":874,"tags":875,"workflow":876},1778690773482.4827,"k175sdyyg60kjnnbkm4029s70186ngbn",{"reviewCount":8},{"description":320,"installMethods":865,"name":321,"sourceUrl":13},{"claudeCode":321},{"basePath":319,"githubOwner":224,"githubRepo":258,"locale":17,"slug":321,"type":259},{"evaluate":868,"extract":873},{"promptVersionExtension":217,"promptVersionScoring":218,"score":758,"tags":869,"targetMarket":230,"tier":231},[850,870,871,224,872,853,760],"training","fine-tuning","trl",{"commitSha":283},{"parentExtensionId":262,"repoId":289},[871,853,224,850,760,870,872],{"evaluatedAt":877,"extractAt":293,"updatedAt":877},1778690836703]