[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"extension-marketplace-huggingface-skills-zh-CN":3,"guides-for-huggingface-skills":954,"similar-k17es3r8wd37t5rrwqcpp5kwrh86mxx8-zh-CN":955},{"_creationTime":4,"_id":5,"children":6,"community":339,"display":340,"evaluation":345,"identity":532,"isFallback":529,"parentExtension":535,"providers":536,"relations":544,"repo":545,"tags":952,"workflow":953},1778690773482.4824,"k17es3r8wd37t5rrwqcpp5kwrh86mxx8",[7,45,66,90,113,137,162,182,202,223,247,270,294,317],{"_creationTime":8,"_id":9,"community":10,"display":12,"identity":17,"providers":23,"relations":39,"tags":41,"workflow":42},1778690773482.4827,"k175sdyyg60kjnnbkm4029s70186ngbn",{"reviewCount":11},0,{"description":13,"installMethods":14,"name":15,"sourceUrl":16},"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.",{"claudeCode":15},"huggingface-llm-trainer","https://github.com/huggingface/skills",{"basePath":18,"githubOwner":19,"githubRepo":20,"locale":21,"slug":15,"type":22},"skills/huggingface-llm-trainer","huggingface","skills","en","plugin",{"evaluate":24,"extract":37},{"promptVersionExtension":25,"promptVersionScoring":26,"score":27,"tags":28,"targetMarket":35,"tier":36},"3.0.0","4.4.0",99,[29,30,31,19,32,33,34],"llm","training","fine-tuning","trl","gguf","mlops","global","verified",{"commitSha":38},"HEAD",{"parentExtensionId":5,"repoId":40},"kd72xwt5xnc0ktc4p7smzfcp3986m959",[31,33,19,29,34,30,32],{"evaluatedAt":43,"extractAt":44,"updatedAt":43},1778690836703,1778690773482,{"_creationTime":46,"_id":47,"community":48,"display":49,"identity":53,"providers":55,"relations":62,"tags":63,"workflow":64},1778690773482.483,"k172w5kkdn7117xpqyyhcqsww186n17b",{"reviewCount":11},{"description":50,"installMethods":51,"name":52,"sourceUrl":16},"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.",{"claudeCode":52},"huggingface-local-models",{"basePath":54,"githubOwner":19,"githubRepo":20,"locale":21,"slug":52,"type":22},"skills/huggingface-local-models",{"evaluate":56,"extract":61},{"promptVersionExtension":25,"promptVersionScoring":26,"score":27,"tags":57,"targetMarket":35,"tier":36},[29,58,59,33,19,60],"local-models","llama-cpp","ml-ops",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[33,19,59,29,58,60],{"evaluatedAt":65,"extractAt":44,"updatedAt":65},1778690852474,{"_creationTime":67,"_id":68,"community":69,"display":70,"identity":74,"providers":76,"relations":86,"tags":87,"workflow":88},1778690773482.4832,"k178yjakvy2y11set9vw91xvnh86nfxr",{"reviewCount":11},{"description":71,"installMethods":72,"name":73,"sourceUrl":16},"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.",{"claudeCode":73},"huggingface-paper-publisher",{"basePath":75,"githubOwner":19,"githubRepo":20,"locale":21,"slug":73,"type":22},"skills/huggingface-paper-publisher",{"evaluate":77,"extract":85},{"promptVersionExtension":25,"promptVersionScoring":26,"score":78,"tags":79,"targetMarket":35,"tier":36},98,[19,80,81,82,83,84],"research","papers","publishing","documentation","arxiv",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[84,83,19,81,82,80],{"evaluatedAt":89,"extractAt":44,"updatedAt":89},1778690873816,{"_creationTime":91,"_id":92,"community":93,"display":94,"identity":99,"providers":101,"relations":109,"tags":110,"workflow":111},1778690773482.4834,"k179sm2kkyd7r7nz9jsx62jm9x86mw4a",{"reviewCount":11},{"description":95,"installMethods":96,"name":98,"sourceUrl":16},"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.",{"claudeCode":97},"huggingface-papers","Hugging Face Papers",{"basePath":100,"githubOwner":19,"githubRepo":20,"locale":21,"slug":97,"type":22},"skills/huggingface-papers",{"evaluate":102,"extract":107},{"promptVersionExtension":25,"promptVersionScoring":26,"score":103,"tags":104,"targetMarket":35,"tier":36},100,[19,81,84,105,80,106],"ai","metadata",{"commitSha":38,"license":108},"Apache-2.0",{"parentExtensionId":5,"repoId":40},[105,84,19,106,81,80],{"evaluatedAt":112,"extractAt":44,"updatedAt":112},1778690901306,{"_creationTime":114,"_id":115,"community":116,"display":117,"identity":121,"providers":123,"relations":133,"tags":134,"workflow":135},1778690773482.4836,"k17be75g9tc47hn3db5vaaxzwh86mvd4",{"reviewCount":11},{"description":118,"installMethods":119,"name":120,"sourceUrl":16},"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.",{"claudeCode":120},"huggingface-community-evals",{"basePath":122,"githubOwner":19,"githubRepo":20,"locale":21,"slug":120,"type":22},"skills/huggingface-community-evals",{"evaluate":124,"extract":132},{"promptVersionExtension":25,"promptVersionScoring":26,"score":78,"tags":125,"targetMarket":35,"tier":36},[126,19,127,128,129,130,131],"evaluation","ml","python","vllm","lighteval","inspect-ai",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[126,19,131,130,127,128,129],{"evaluatedAt":136,"extractAt":44,"updatedAt":136},1778690920957,{"_creationTime":138,"_id":139,"community":140,"display":141,"identity":145,"providers":147,"relations":158,"tags":159,"workflow":160},1778690773482.484,"k175xxwhkryry1vj3f64maceh586npj6",{"reviewCount":11},{"description":142,"installMethods":143,"name":144,"sourceUrl":16},"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.",{"claudeCode":144},"huggingface-best",{"basePath":146,"githubOwner":19,"githubRepo":20,"locale":21,"slug":144,"type":22},"skills/huggingface-best",{"evaluate":148,"extract":157},{"promptVersionExtension":25,"promptVersionScoring":26,"score":149,"tags":150,"targetMarket":35,"tier":156},79,[19,151,152,153,154,155],"ai-model-recommendation","leaderboards","benchmarks","cli","developer-tools","community",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[151,153,154,155,19,152],{"evaluatedAt":161,"extractAt":44,"updatedAt":161},1778690943353,{"_creationTime":163,"_id":164,"community":165,"display":166,"identity":170,"providers":172,"relations":178,"tags":179,"workflow":180},1778690773482.4841,"k172m7ench1yy64c6nbcx5pmm586mphh",{"reviewCount":11},{"description":167,"installMethods":168,"name":169,"sourceUrl":16},"Execute Hugging Face Hub operations using the hf CLI. Download models/datasets, upload files, manage repos, and run cloud compute jobs.",{"claudeCode":169},"hf-cli",{"basePath":171,"githubOwner":19,"githubRepo":20,"locale":21,"slug":169,"type":22},"skills/hf-cli",{"evaluate":173,"extract":177},{"promptVersionExtension":25,"promptVersionScoring":26,"score":27,"tags":174,"targetMarket":35,"tier":36},[19,154,155,34,175,176],"datasets","models",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[154,175,155,19,34,176],{"evaluatedAt":181,"extractAt":44,"updatedAt":181},1778690967596,{"_creationTime":183,"_id":184,"community":185,"display":186,"identity":190,"providers":192,"relations":198,"tags":199,"workflow":200},1778690773482.4844,"k1785s3263hzxs6ne1nsbzas5186m86v",{"reviewCount":11},{"description":187,"installMethods":188,"name":189,"sourceUrl":16},"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.",{"claudeCode":189},"huggingface-trackio",{"basePath":191,"githubOwner":19,"githubRepo":20,"locale":21,"slug":189,"type":22},"skills/huggingface-trackio",{"evaluate":193,"extract":197},{"promptVersionExtension":25,"promptVersionScoring":26,"score":27,"tags":194,"targetMarket":35,"tier":36},[34,195,196,19,128,154],"experiment-tracking","monitoring",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[154,195,19,34,196,128],{"evaluatedAt":201,"extractAt":44,"updatedAt":201},1778690988981,{"_creationTime":203,"_id":204,"community":205,"display":206,"identity":210,"providers":212,"relations":219,"tags":220,"workflow":221},1778690773482.4846,"k17a1t3nm72bs2xf80qmf4k9md86mk26",{"reviewCount":11},{"description":207,"installMethods":208,"name":209,"sourceUrl":16},"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":209},"huggingface-datasets",{"basePath":211,"githubOwner":19,"githubRepo":20,"locale":21,"slug":209,"type":22},"skills/huggingface-datasets",{"evaluate":213,"extract":218},{"promptVersionExtension":25,"promptVersionScoring":26,"score":78,"tags":214,"targetMarket":35,"tier":36},[19,175,215,154,216,217],"api","data-exploration","npx",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[215,154,216,175,19,217],{"evaluatedAt":222,"extractAt":44,"updatedAt":222},1778691013020,{"_creationTime":224,"_id":225,"community":226,"display":227,"identity":232,"providers":234,"relations":243,"tags":244,"workflow":245},1778690773482.4849,"k1700wce4kdtr4t5rmewvk34q186meb7",{"reviewCount":11},{"description":228,"installMethods":229,"name":231,"sourceUrl":16},"Build reusable scripts for Hugging Face Hub and API workflows. Useful for chaining API calls, enriching Hub metadata, or automating repeated tasks.",{"claudeCode":230},"huggingface-tool-builder","Hugging Face Tool Builder",{"basePath":233,"githubOwner":19,"githubRepo":20,"locale":21,"slug":230,"type":22},"skills/huggingface-tool-builder",{"evaluate":235,"extract":242},{"promptVersionExtension":25,"promptVersionScoring":26,"score":236,"tags":237,"targetMarket":35,"tier":36},95,[19,215,154,238,239,128,240,241],"scripting","automation","bash","typescript",{"commitSha":38,"license":108},{"parentExtensionId":5,"repoId":40},[215,239,240,154,19,128,238,241],{"evaluatedAt":246,"extractAt":44,"updatedAt":246},1778691032588,{"_creationTime":248,"_id":249,"community":250,"display":251,"identity":255,"providers":257,"relations":266,"tags":267,"workflow":268},1778690773482.485,"k17d14kty8pzk8t0pf88jfj8ax86mr5j",{"reviewCount":11},{"description":252,"installMethods":253,"name":254,"sourceUrl":16},"Build Gradio web UIs and demos in Python. Use when creating or editing Gradio apps, components, event listeners, layouts, or chatbots.",{"claudeCode":254},"huggingface-gradio",{"basePath":256,"githubOwner":19,"githubRepo":20,"locale":21,"slug":254,"type":22},"skills/huggingface-gradio",{"evaluate":258,"extract":265},{"promptVersionExtension":25,"promptVersionScoring":26,"score":78,"tags":259,"targetMarket":35,"tier":36},[128,260,261,262,263,264],"web-development","ui","machine-learning","demo","gradio",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[263,264,262,128,261,260],{"evaluatedAt":269,"extractAt":44,"updatedAt":269},1778691060509,{"_creationTime":271,"_id":272,"community":273,"display":274,"identity":278,"providers":280,"relations":290,"tags":291,"workflow":292},1778690773482.4854,"k17745362t936z67p0p8w8mq0h86nmf0",{"reviewCount":11},{"description":275,"installMethods":276,"name":277,"sourceUrl":16},"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.",{"claudeCode":277},"transformers-js",{"basePath":279,"githubOwner":19,"githubRepo":20,"locale":21,"slug":277,"type":22},"skills/transformers-js",{"evaluate":281,"extract":289},{"promptVersionExtension":25,"promptVersionScoring":26,"score":282,"tags":283,"targetMarket":35,"tier":36},96,[262,284,241,285,286,287,288],"javascript","nlp","computer-vision","audio","multimodal",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[287,286,284,262,288,285,241],{"evaluatedAt":293,"extractAt":44,"updatedAt":293},1778691120894,{"_creationTime":295,"_id":296,"community":297,"display":298,"identity":302,"providers":304,"relations":313,"tags":314,"workflow":315},1778690773482.4856,"k179t4q3q7ywr9f9ka3wxrmbxs86m34d",{"reviewCount":11},{"description":299,"installMethods":300,"name":301,"sourceUrl":16},"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.",{"claudeCode":301},"huggingface-vision-trainer",{"basePath":303,"githubOwner":19,"githubRepo":20,"locale":21,"slug":301,"type":22},"skills/huggingface-vision-trainer",{"evaluate":305,"extract":312},{"promptVersionExtension":25,"promptVersionScoring":26,"score":282,"tags":306,"targetMarket":35,"tier":36},[262,286,307,308,309,310,311,128],"object-detection","image-classification","segmentation","hugging-face","transformers",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[286,310,308,262,307,128,309,311],{"evaluatedAt":316,"extractAt":44,"updatedAt":316},1778691153160,{"_creationTime":318,"_id":319,"community":320,"display":321,"identity":325,"providers":327,"relations":335,"tags":336,"workflow":337},1778690773482.4858,"k175rwqsqyx8atwtz5cs5b3fpx86m84e",{"reviewCount":11},{"description":322,"installMethods":323,"name":324,"sourceUrl":16},"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.",{"claudeCode":324},"train-sentence-transformers",{"basePath":326,"githubOwner":19,"githubRepo":20,"locale":21,"slug":324,"type":22},"skills/train-sentence-transformers",{"evaluate":328,"extract":334},{"promptVersionExtension":25,"promptVersionScoring":26,"score":27,"tags":329,"targetMarket":35,"tier":36},[262,285,330,331,332,333],"sentence-transformers","model-training","embeddings","reranking",{"commitSha":38},{"parentExtensionId":5,"repoId":40},[332,262,331,285,333,330],{"evaluatedAt":338,"extractAt":44,"updatedAt":338},1778691173389,{"reviewCount":11},{"description":341,"installMethods":342,"name":344,"sourceUrl":16},"Agent Skills for AI/ML tasks including dataset creation, model training, evaluation, and research paper publishing on Hugging Face Hub",{"claudeCode":343},"huggingface/skills","huggingface-skills",{"_creationTime":346,"_id":347,"extensionId":5,"locale":21,"result":348,"trustSignals":513,"workflow":530},1778690814090.0784,"kn774gkzvzd7aqrzcj797qzx9x86md21",{"checks":349,"evaluatedAt":492,"extensionSummary":493,"features":494,"nonGoals":499,"promptVersionExtension":503,"promptVersionScoring":26,"purpose":504,"rationale":505,"score":236,"summary":506,"tags":507,"targetMarket":35,"tier":36,"useCases":509},[350,355,358,361,365,368,372,377,380,383,387,391,394,398,401,404,407,410,413,416,420,424,428,432,436,439,443,446,450,453,457,461,464,467,470,473,476,479,483,486,489],{"category":351,"check":352,"severity":353,"summary":354},"Practical Utility","Problem relevance","pass","The description clearly names the user problem of accessing AI/ML tools and resources on Hugging Face Hub for tasks like dataset creation, model training, and research paper publishing.",{"category":351,"check":356,"severity":353,"summary":357},"Unique selling proposition","The skills offer distinct capabilities beyond basic LLM prompting, providing specialized tools for complex AI/ML workflows on Hugging Face Hub.",{"category":351,"check":359,"severity":353,"summary":360},"Production readiness","The collection of skills covers a comprehensive lifecycle for AI/ML tasks, from dataset creation and model training to evaluation and publishing, and integrates with multiple agent platforms.",{"category":362,"check":363,"severity":353,"summary":364},"Scope","Single responsibility principle","The marketplace curates skills related to Hugging Face AI/ML tasks, demonstrating a coherent thematic focus.",{"category":362,"check":366,"severity":353,"summary":367},"Description quality","The displayed description accurately reflects the purpose and scope of the skills available in the Hugging Face Hub for AI/ML tasks.",{"category":369,"check":370,"severity":353,"summary":371},"Invocation","Scoped tools","The marketplace aggregates skills, and the individual skills themselves are expected to expose narrow, verb-noun specialist tools.",{"category":373,"check":374,"severity":375,"summary":376},"Documentation","Configuration & parameter reference","not_applicable","As this is a marketplace metadata-only extension, there are no tools or parameters to document.",{"category":362,"check":378,"severity":353,"summary":379},"Tool naming","The individual skills listed appear to follow descriptive naming conventions for their tools.",{"category":362,"check":381,"severity":353,"summary":382},"Minimal I/O surface","This check is not applicable to the marketplace itself; it would be evaluated at the individual skill level.",{"category":384,"check":385,"severity":353,"summary":386},"License","License usability","The repository is licensed under the Apache-2.0 license, which is a permissive open-source license.",{"category":388,"check":389,"severity":353,"summary":390},"Maintenance","Commit recency","The last commit was on 2026-05-12, which is recent.",{"category":388,"check":392,"severity":375,"summary":393},"Dependency Management","The marketplace extension itself does not have third-party dependencies; individual skills might.",{"category":395,"check":396,"severity":375,"summary":397},"Security","Secret Management","This is a metadata-only marketplace and does not handle secrets.",{"category":395,"check":399,"severity":375,"summary":400},"Injection","This is a metadata-only marketplace and does not load or execute third-party data.",{"category":395,"check":402,"severity":375,"summary":403},"Transitive Supply-Chain Grenades","This is a metadata-only marketplace and does not fetch remote content at runtime.",{"category":395,"check":405,"severity":375,"summary":406},"Sandbox Isolation","This is a metadata-only marketplace and does not modify files or interact with a file system.",{"category":395,"check":408,"severity":375,"summary":409},"Sandbox escape primitives","This is a metadata-only marketplace and has no scripts or hooks that could escape a sandbox.",{"category":395,"check":411,"severity":375,"summary":412},"Data Exfiltration","This is a metadata-only marketplace and does not handle or exfiltrate data.",{"category":395,"check":414,"severity":353,"summary":415},"Hidden Text Tricks","The bundled content is free of hidden-steering tricks, and descriptions are clean.",{"category":417,"check":418,"severity":375,"summary":419},"Hooks","Opaque code execution","This marketplace extension does not contain any executable code or scripts.",{"category":421,"check":422,"severity":375,"summary":423},"Portability","Structural Assumption","This is a metadata-only marketplace and makes no assumptions about user project structure.",{"category":425,"check":426,"severity":353,"summary":427},"Trust","Issues Attention","There are 4 open issues and 6 closed issues in the last 90 days, indicating active maintenance and response.",{"category":429,"check":430,"severity":353,"summary":431},"Versioning","Release Management","A meaningful version (1.0.2) is declared in the marketplace metadata, and the repository is actively maintained with recent commits.",{"category":433,"check":434,"severity":375,"summary":435},"Code Execution","Validation","This is a metadata-only marketplace and does not have executable code or structured output to validate.",{"category":395,"check":437,"severity":375,"summary":438},"Unguarded Destructive Operations","This is a metadata-only marketplace and has no destructive operations.",{"category":440,"check":441,"severity":375,"summary":442},"Errors","Error Handling","This is a metadata-only marketplace and has no user-facing error paths.",{"category":440,"check":444,"severity":375,"summary":445},"Actionable error messages","This is a metadata-only marketplace and has no error paths to provide messages for.",{"category":447,"check":448,"severity":375,"summary":449},"Execution","Pinned dependencies","This is a metadata-only marketplace and does not use third-party dependencies.",{"category":362,"check":451,"severity":375,"summary":452},"Dry-run preview","This marketplace extension is metadata-only and has no state-changing commands or outbound data sending.",{"category":454,"check":455,"severity":375,"summary":456},"Protocol","Idempotent retry & timeouts","This is a metadata-only marketplace and has no remote calls or state-changing operations.",{"category":458,"check":459,"severity":375,"summary":460},"Compliance","Telemetry opt-in","This is a metadata-only marketplace and does not emit telemetry.",{"category":362,"check":462,"severity":353,"summary":463},"Theme declaration","The marketplace clearly declares a curation theme focused on AI/ML tasks and Hugging Face Hub integration.",{"category":429,"check":465,"severity":353,"summary":466},"Per-entry version metadata","The marketplace JSON lists individual skills with explicit source references, implying versioning can be managed per entry.",{"category":373,"check":468,"severity":353,"summary":469},"README","The README file clearly explains the purpose of the Hugging Face Skills repository and how to use them.",{"category":362,"check":471,"severity":375,"summary":472},"Tool surface size","This check is not applicable to a marketplace extension, which does not expose tools directly.",{"category":369,"check":474,"severity":375,"summary":475},"Overlapping near-synonym tools","This check is not applicable to a marketplace extension as it does not expose tools.",{"category":373,"check":477,"severity":353,"summary":478},"Phantom features","All features mentioned in the README and marketplace descriptions correspond to available skills.",{"category":480,"check":481,"severity":353,"summary":482},"Install","Installation instruction","The README provides clear, copy-pasteable installation instructions for multiple agent platforms (Claude Code, Codex, Gemini CLI, Cursor).",{"category":458,"check":484,"severity":375,"summary":485},"GDPR","This marketplace extension is metadata-only and does not process personal data.",{"category":458,"check":487,"severity":353,"summary":488},"Target market","The extension is global in scope, with no regional or jurisdictional restrictions detected.",{"category":421,"check":490,"severity":375,"summary":491},"Runtime stability","This is a metadata-only marketplace and does not have runtime stability concerns.",1778690813910,"This is a marketplace listing for Hugging Face Skills, providing definitions for AI/ML tasks such as dataset creation, model training, evaluation, and research paper publishing. It supports interoperability with major coding agent tools.",[495,496,497,498],"Curated catalog of AI/ML skills","Supports dataset creation, model training, evaluation, and research publishing","Interoperable with major coding agent tools","Clear installation instructions for multiple platforms",[500,501,502],"Hosting executable code","Providing runtime environments for skills","Bundling individual skills directly within the marketplace","3.1.0","To provide a curated catalog of specialized AI/ML skills, enabling coding agents to easily access and leverage Hugging Face Hub capabilities for a wide range of machine learning workflows.","The extension is well-documented, actively maintained, and clearly defines its scope, leading to a high trust score. No critical or warning findings were identified.","A comprehensive and well-maintained marketplace of AI/ML skills for Hugging Face Hub.",[508,19,175,176,80,155],"ai-ml",[510,511,512],"Finding and installing AI/ML skills for specific Hugging Face tasks","Enabling agents to perform advanced ML operations via curated skills","Browsing available AI/ML capabilities for development and research",{"codeQuality":514,"collectedAt":516,"documentation":517,"maintenance":520,"security":527,"testCoverage":528},{"hasLockfile":515},false,1778690776665,{"descriptionLength":518,"readmeSize":519},134,9821,{"closedIssues90d":521,"forks":522,"hasChangelog":515,"manifestVersion":523,"openIssues90d":524,"pushedAt":525,"stars":526},6,663,"1.0.2",4,1778593131000,10482,{"hasNpmPackage":515,"license":108,"smitheryVerified":515},{"hasCi":529,"hasTests":515},true,{"updatedAt":531},1778690814090,{"basePath":533,"githubOwner":19,"githubRepo":20,"locale":21,"slug":20,"type":534},"","marketplace",null,{"evaluate":537,"extract":539},{"promptVersionExtension":503,"promptVersionScoring":26,"score":236,"tags":538,"targetMarket":35,"tier":36},[508,19,175,176,80,155],{"commitSha":38,"marketplace":540,"plugin":542},{"name":344,"pluginCount":541},14,{"mcpCount":11,"provider":543,"skillCount":11},"classify",{"repoId":40},{"_creationTime":546,"_id":40,"identity":547,"providers":548,"workflow":948},1778689536128.5474,{"githubOwner":19,"githubRepo":20,"sourceUrl":16},{"classify":549,"discover":941,"github":944},{"commitSha":38,"extensions":550},[551,564,570,575,580,585,590,595,600,605,610,615,620,625,630,635,678,687,693,699,716,722,729,771,782,801,807,827,839,863,921],{"basePath":533,"description":341,"displayName":344,"installMethods":552,"rationale":553,"selectedPaths":554,"source":563,"sourceLanguage":21,"type":534},{"claudeCode":343},"marketplace.json at .claude-plugin/marketplace.json",[555,558,560],{"path":556,"priority":557},".claude-plugin/marketplace.json","mandatory",{"path":559,"priority":557},"README.md",{"path":561,"priority":562},"LICENSE","high","rule",{"basePath":18,"description":13,"displayName":15,"installMethods":565,"rationale":566,"selectedPaths":567,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":15},"inline plugin source from marketplace.json at skills/huggingface-llm-trainer",[568],{"path":569,"priority":562},"SKILL.md",{"basePath":54,"description":50,"displayName":52,"installMethods":571,"rationale":572,"selectedPaths":573,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":52},"inline plugin source from marketplace.json at skills/huggingface-local-models",[574],{"path":569,"priority":562},{"basePath":75,"description":71,"displayName":73,"installMethods":576,"rationale":577,"selectedPaths":578,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":73},"inline plugin source from marketplace.json at skills/huggingface-paper-publisher",[579],{"path":569,"priority":562},{"basePath":100,"description":95,"displayName":97,"installMethods":581,"rationale":582,"selectedPaths":583,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":97},"inline plugin source from marketplace.json at skills/huggingface-papers",[584],{"path":569,"priority":562},{"basePath":122,"description":118,"displayName":120,"installMethods":586,"rationale":587,"selectedPaths":588,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":120},"inline plugin source from marketplace.json at skills/huggingface-community-evals",[589],{"path":569,"priority":562},{"basePath":146,"description":142,"displayName":144,"installMethods":591,"rationale":592,"selectedPaths":593,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":144},"inline plugin source from marketplace.json at skills/huggingface-best",[594],{"path":569,"priority":562},{"basePath":171,"description":167,"displayName":169,"installMethods":596,"rationale":597,"selectedPaths":598,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":169},"inline plugin source from marketplace.json at skills/hf-cli",[599],{"path":569,"priority":562},{"basePath":191,"description":187,"displayName":189,"installMethods":601,"rationale":602,"selectedPaths":603,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":189},"inline plugin source from marketplace.json at skills/huggingface-trackio",[604],{"path":569,"priority":562},{"basePath":211,"description":207,"displayName":209,"installMethods":606,"rationale":607,"selectedPaths":608,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":209},"inline plugin source from marketplace.json at skills/huggingface-datasets",[609],{"path":569,"priority":562},{"basePath":233,"description":228,"displayName":230,"installMethods":611,"rationale":612,"selectedPaths":613,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":230},"inline plugin source from marketplace.json at skills/huggingface-tool-builder",[614],{"path":569,"priority":562},{"basePath":256,"description":252,"displayName":254,"installMethods":616,"rationale":617,"selectedPaths":618,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":254},"inline plugin source from marketplace.json at skills/huggingface-gradio",[619],{"path":569,"priority":562},{"basePath":279,"description":275,"displayName":277,"installMethods":621,"rationale":622,"selectedPaths":623,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":277},"inline plugin source from marketplace.json at skills/transformers-js",[624],{"path":569,"priority":562},{"basePath":303,"description":299,"displayName":301,"installMethods":626,"rationale":627,"selectedPaths":628,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":301},"inline plugin source from marketplace.json at skills/huggingface-vision-trainer",[629],{"path":569,"priority":562},{"basePath":326,"description":322,"displayName":324,"installMethods":631,"rationale":632,"selectedPaths":633,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":324},"inline plugin source from marketplace.json at skills/train-sentence-transformers",[634],{"path":569,"priority":562},{"basePath":533,"description":341,"displayName":344,"installMethods":636,"license":108,"rationale":637,"selectedPaths":638,"source":563,"sourceLanguage":21,"type":22},{"claudeCode":344},"plugin manifest at .claude-plugin/plugin.json",[639,641,642,643,646,648,650,652,654,656,658,660,662,664,666,668,670,672,674,676],{"path":640,"priority":557},".claude-plugin/plugin.json",{"path":559,"priority":557},{"path":561,"priority":562},{"path":644,"priority":645},"skills/hf-cli/SKILL.md","medium",{"path":647,"priority":645},"skills/huggingface-best/SKILL.md",{"path":649,"priority":645},"skills/huggingface-community-evals/SKILL.md",{"path":651,"priority":645},"skills/huggingface-datasets/SKILL.md",{"path":653,"priority":645},"skills/huggingface-gradio/SKILL.md",{"path":655,"priority":645},"skills/huggingface-llm-trainer/SKILL.md",{"path":657,"priority":645},"skills/huggingface-local-models/SKILL.md",{"path":659,"priority":645},"skills/huggingface-paper-publisher/SKILL.md",{"path":661,"priority":645},"skills/huggingface-papers/SKILL.md",{"path":663,"priority":645},"skills/huggingface-tool-builder/SKILL.md",{"path":665,"priority":645},"skills/huggingface-trackio/SKILL.md",{"path":667,"priority":645},"skills/huggingface-vision-trainer/SKILL.md",{"path":669,"priority":645},"skills/train-sentence-transformers/SKILL.md",{"path":671,"priority":645},"skills/transformers-js/SKILL.md",{"path":673,"priority":557},".mcp.json",{"path":675,"priority":562},"agents/AGENTS.md",{"path":677,"priority":562},".cursor-plugin/plugin.json",{"basePath":679,"description":680,"displayName":681,"installMethods":682,"rationale":683,"selectedPaths":684,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at hf-mcp/skills/hf-mcp/SKILL.md",[685],{"path":569,"priority":557},"skill",{"basePath":171,"description":688,"displayName":169,"installMethods":689,"rationale":690,"selectedPaths":691,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/hf-cli/SKILL.md",[692],{"path":569,"priority":557},{"basePath":146,"description":694,"displayName":144,"installMethods":695,"rationale":696,"selectedPaths":697,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/huggingface-best/SKILL.md",[698],{"path":569,"priority":557},{"basePath":122,"description":700,"displayName":120,"installMethods":701,"rationale":702,"selectedPaths":703,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/huggingface-community-evals/SKILL.md",[704,705,708,710,712,714],{"path":569,"priority":557},{"path":706,"priority":707},"examples/.env.example","low",{"path":709,"priority":707},"examples/USAGE_EXAMPLES.md",{"path":711,"priority":707},"scripts/inspect_eval_uv.py",{"path":713,"priority":707},"scripts/inspect_vllm_uv.py",{"path":715,"priority":707},"scripts/lighteval_vllm_uv.py",{"basePath":211,"description":717,"displayName":209,"installMethods":718,"rationale":719,"selectedPaths":720,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/huggingface-datasets/SKILL.md",[721],{"path":569,"priority":557},{"basePath":256,"description":252,"displayName":254,"installMethods":723,"rationale":724,"selectedPaths":725,"source":563,"sourceLanguage":21,"type":686},{"claudeCode":343},"SKILL.md frontmatter at skills/huggingface-gradio/SKILL.md",[726,727],{"path":569,"priority":557},{"path":728,"priority":645},"examples.md",{"basePath":18,"description":730,"displayName":15,"installMethods":731,"rationale":732,"selectedPaths":733,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/huggingface-llm-trainer/SKILL.md",[734,735,737,739,741,743,745,747,749,751,753,755,757,759,761,763,765,767,769],{"path":569,"priority":557},{"path":736,"priority":645},"references/gguf_conversion.md",{"path":738,"priority":645},"references/hardware_guide.md",{"path":740,"priority":645},"references/hub_saving.md",{"path":742,"priority":645},"references/local_training_macos.md",{"path":744,"priority":645},"references/reliability_principles.md",{"path":746,"priority":645},"references/trackio_guide.md",{"path":748,"priority":645},"references/training_methods.md",{"path":750,"priority":645},"references/training_patterns.md",{"path":752,"priority":645},"references/troubleshooting.md",{"path":754,"priority":645},"references/unsloth.md",{"path":756,"priority":707},"scripts/convert_to_gguf.py",{"path":758,"priority":707},"scripts/dataset_inspector.py",{"path":760,"priority":707},"scripts/estimate_cost.py",{"path":762,"priority":707},"scripts/hf_benchmarks.py",{"path":764,"priority":707},"scripts/train_dpo_example.py",{"path":766,"priority":707},"scripts/train_grpo_example.py",{"path":768,"priority":707},"scripts/train_sft_example.py",{"path":770,"priority":707},"scripts/unsloth_sft_example.py",{"basePath":54,"description":50,"displayName":52,"installMethods":772,"rationale":773,"selectedPaths":774,"source":563,"sourceLanguage":21,"type":686},{"claudeCode":343},"SKILL.md frontmatter at skills/huggingface-local-models/SKILL.md",[775,776,778,780],{"path":569,"priority":557},{"path":777,"priority":645},"references/hardware.md",{"path":779,"priority":645},"references/hub-discovery.md",{"path":781,"priority":645},"references/quantization.md",{"basePath":75,"description":71,"displayName":73,"installMethods":783,"rationale":784,"selectedPaths":785,"source":563,"sourceLanguage":21,"type":686},{"claudeCode":343},"SKILL.md frontmatter at skills/huggingface-paper-publisher/SKILL.md",[786,787,789,791,793,795,797,799],{"path":569,"priority":557},{"path":788,"priority":707},"examples/example_usage.md",{"path":790,"priority":645},"references/quick_reference.md",{"path":792,"priority":707},"scripts/paper_manager.py",{"path":794,"priority":707},"templates/arxiv.md",{"path":796,"priority":707},"templates/ml-report.md",{"path":798,"priority":707},"templates/modern.md",{"path":800,"priority":707},"templates/standard.md",{"basePath":100,"description":802,"displayName":97,"installMethods":803,"rationale":804,"selectedPaths":805,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/huggingface-papers/SKILL.md",[806],{"path":569,"priority":557},{"basePath":233,"description":808,"displayName":230,"installMethods":809,"rationale":810,"selectedPaths":811,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/huggingface-tool-builder/SKILL.md",[812,813,815,817,819,821,823,825],{"path":569,"priority":557},{"path":814,"priority":645},"references/baseline_hf_api.py",{"path":816,"priority":645},"references/baseline_hf_api.sh",{"path":818,"priority":645},"references/baseline_hf_api.tsx",{"path":820,"priority":645},"references/find_models_by_paper.sh",{"path":822,"priority":645},"references/hf_enrich_models.sh",{"path":824,"priority":645},"references/hf_model_card_frontmatter.sh",{"path":826,"priority":645},"references/hf_model_papers_auth.sh",{"basePath":191,"description":828,"displayName":189,"installMethods":829,"rationale":830,"selectedPaths":831,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/huggingface-trackio/SKILL.md",[832,833,835,837],{"path":569,"priority":557},{"path":834,"priority":645},"references/alerts.md",{"path":836,"priority":645},"references/logging_metrics.md",{"path":838,"priority":645},"references/retrieving_metrics.md",{"basePath":303,"description":840,"displayName":301,"installMethods":841,"rationale":842,"selectedPaths":843,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/huggingface-vision-trainer/SKILL.md",[844,845,847,848,850,852,853,855,856,857,859,861],{"path":569,"priority":557},{"path":846,"priority":645},"references/finetune_sam2_trainer.md",{"path":740,"priority":645},{"path":849,"priority":645},"references/image_classification_training_notebook.md",{"path":851,"priority":645},"references/object_detection_training_notebook.md",{"path":744,"priority":645},{"path":854,"priority":645},"references/timm_trainer.md",{"path":758,"priority":707},{"path":760,"priority":707},{"path":858,"priority":707},"scripts/image_classification_training.py",{"path":860,"priority":707},"scripts/object_detection_training.py",{"path":862,"priority":707},"scripts/sam_segmentation_training.py",{"basePath":326,"description":864,"displayName":324,"installMethods":865,"rationale":866,"selectedPaths":867,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/train-sentence-transformers/SKILL.md",[868,869,871,873,875,877,879,880,882,884,886,888,890,892,894,895,897,899,901,903,905,907,909,911,913,915,917,919],{"path":569,"priority":557},{"path":870,"priority":645},"references/base_model_selection.md",{"path":872,"priority":645},"references/dataset_formats.md",{"path":874,"priority":645},"references/evaluators_cross_encoder.md",{"path":876,"priority":645},"references/evaluators_sentence_transformer.md",{"path":878,"priority":645},"references/evaluators_sparse_encoder.md",{"path":738,"priority":645},{"path":881,"priority":645},"references/hf_jobs_execution.md",{"path":883,"priority":645},"references/losses_cross_encoder.md",{"path":885,"priority":645},"references/losses_sentence_transformer.md",{"path":887,"priority":645},"references/losses_sparse_encoder.md",{"path":889,"priority":645},"references/model_architectures.md",{"path":891,"priority":645},"references/prompts_and_instructions.md",{"path":893,"priority":645},"references/training_args.md",{"path":752,"priority":645},{"path":896,"priority":707},"scripts/mine_hard_negatives.py",{"path":898,"priority":707},"scripts/train_cross_encoder_distillation_example.py",{"path":900,"priority":707},"scripts/train_cross_encoder_example.py",{"path":902,"priority":707},"scripts/train_cross_encoder_listwise_example.py",{"path":904,"priority":707},"scripts/train_sentence_transformer_distillation_example.py",{"path":906,"priority":707},"scripts/train_sentence_transformer_example.py",{"path":908,"priority":707},"scripts/train_sentence_transformer_make_multilingual_example.py",{"path":910,"priority":707},"scripts/train_sentence_transformer_matryoshka_example.py",{"path":912,"priority":707},"scripts/train_sentence_transformer_multi_dataset_example.py",{"path":914,"priority":707},"scripts/train_sentence_transformer_static_embedding_example.py",{"path":916,"priority":707},"scripts/train_sentence_transformer_with_lora_example.py",{"path":918,"priority":707},"scripts/train_sparse_encoder_distillation_example.py",{"path":920,"priority":707},"scripts/train_sparse_encoder_example.py",{"basePath":279,"description":922,"displayName":277,"installMethods":923,"rationale":924,"selectedPaths":925,"source":563,"sourceLanguage":21,"type":686},"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":343},"SKILL.md frontmatter at skills/transformers-js/SKILL.md",[926,927,929,931,933,935,937,939],{"path":569,"priority":557},{"path":928,"priority":645},"references/CACHE.md",{"path":930,"priority":645},"references/CONFIGURATION.md",{"path":932,"priority":645},"references/EXAMPLES.md",{"path":934,"priority":645},"references/MODEL_ARCHITECTURES.md",{"path":936,"priority":645},"references/MODEL_REGISTRY.md",{"path":938,"priority":645},"references/PIPELINE_OPTIONS.md",{"path":940,"priority":645},"references/TEXT_GENERATION.md",{"sources":942},[943],"manual",{"closedIssues90d":521,"description":945,"forks":522,"homepage":946,"license":108,"openIssues90d":524,"pushedAt":525,"readmeSize":519,"stars":526,"topics":947},"Give your agents the power of the Hugging Face ecosystem","https://huggingface.co",[],{"classifiedAt":949,"discoverAt":950,"extractAt":951,"githubAt":951,"updatedAt":949},1778690772996,1778689536128,1778690770714,[508,175,155,19,176,80],{"evaluatedAt":531,"extractAt":44,"updatedAt":531},[],[]]