[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"extension-plugin-huggingface-huggingface-community-evals-de":3,"guides-for-huggingface-huggingface-community-evals":747,"similar-k17be75g9tc47hn3db5vaaxzwh86mvd4-de":748},{"_creationTime":4,"_id":5,"children":6,"community":7,"display":9,"evaluation":14,"identity":255,"isFallback":252,"parentExtension":259,"providers":294,"relations":298,"repo":299,"tags":745,"workflow":746},1778690773482.4836,"k17be75g9tc47hn3db5vaaxzwh86mvd4",[],{"reviewCount":8},0,{"description":10,"installMethods":11,"name":12,"sourceUrl":13},"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":12},"huggingface-community-evals","https://github.com/huggingface/skills",{"_creationTime":15,"_id":16,"extensionId":5,"locale":17,"result":18,"trustSignals":236,"workflow":253},1778690920957.7244,"kn77s8d00jd4z59cb3abqf4gzx86mghq","en",{"checks":19,"evaluatedAt":202,"extensionSummary":203,"features":204,"nonGoals":210,"promptVersionExtension":215,"promptVersionScoring":216,"purpose":217,"rationale":218,"score":219,"summary":220,"tags":221,"targetMarket":229,"tier":230,"useCases":231},[20,25,28,31,35,38,42,46,49,52,56,60,64,68,71,74,77,80,83,86,90,94,98,102,106,109,112,115,119,122,125,128,131,134,137,141,145,149,152,156,159,162,165,168,170,173,176,179,181,184,188,191,194,198],{"category":21,"check":22,"severity":23,"summary":24},"Practical Utility","Problem relevance","pass","The description clearly states the problem of running evaluations for Hugging Face models locally, mentioning specific tools and use cases.",{"category":21,"check":26,"severity":23,"summary":27},"Unique selling proposition","The skill offers a specific workflow for local model evaluation using inspect-ai and lighteval, with choices for inference backends, going beyond simple API wrapping.",{"category":21,"check":29,"severity":23,"summary":30},"Production readiness","The skill provides clear scripts and guidance for local evaluation, covering different backends and offering troubleshooting, indicating readiness for workflow integration.",{"category":32,"check":33,"severity":23,"summary":34},"Scope","Single responsibility principle","The skill focuses specifically on running local model evaluations and clearly delineates what it does NOT cover, such as remote execution or model card management.",{"category":32,"check":36,"severity":23,"summary":37},"Description quality","The displayed description accurately reflects the skill's capabilities and limitations as detailed in the SKILL.md file.",{"category":39,"check":40,"severity":23,"summary":41},"Invocation","Scoped tools","The skill exposes specific scripts like `inspect_eval_uv.py` and `lighteval_vllm_uv.py` which act as scoped tools for defined evaluation tasks.",{"category":43,"check":44,"severity":23,"summary":45},"Documentation","Configuration & parameter reference","The SKILL.md provides detailed information on scripts, arguments, backend selection, and hardware guidance, effectively documenting parameters and usage.",{"category":32,"check":47,"severity":23,"summary":48},"Tool naming","The primary tools are scripts with descriptive names like `inspect_eval_uv.py` and `lighteval_vllm_uv.py`, clearly indicating their purpose.",{"category":32,"check":50,"severity":23,"summary":51},"Minimal I/O surface","The scripts take clearly defined arguments like `--model`, `--task`, and `--limit`, and the expected output is the evaluation results, not diagnostic dumps.",{"category":53,"check":54,"severity":23,"summary":55},"License","License usability","The `LICENSE` file clearly indicates the Apache-2.0 license, which is permissive and usable.",{"category":57,"check":58,"severity":23,"summary":59},"Maintenance","Commit recency","The last commit was on 2026-05-12, which is recent (within the last 3 months).",{"category":57,"check":61,"severity":62,"summary":63},"Dependency Management","not_applicable","The skill utilizes `uv` for managing dependencies, which is a suitable mechanism.",{"category":65,"check":66,"severity":23,"summary":67},"Security","Secret Management","The skill mentions setting `HF_TOKEN` for gated models but does not expose or echo resolved secrets in its output or documentation.",{"category":65,"check":69,"severity":23,"summary":70},"Injection","The scripts are designed to run local evaluations and do not appear to load untrusted third-party data as instructions.",{"category":65,"check":72,"severity":23,"summary":73},"Transitive Supply-Chain Grenades","The skill relies on locally installed dependencies managed by `uv` and does not fetch external code or data at runtime.",{"category":65,"check":75,"severity":23,"summary":76},"Sandbox Isolation","The scripts operate within the local project context and do not attempt to modify files outside of the designated project folder.",{"category":65,"check":78,"severity":23,"summary":79},"Sandbox escape primitives","No detached process spawns or deny-retry loops are evident in the provided skill documentation or scripts.",{"category":65,"check":81,"severity":23,"summary":82},"Data Exfiltration","The skill's purpose is local evaluation; there are no documented outbound calls for telemetry or data submission.",{"category":65,"check":84,"severity":23,"summary":85},"Hidden Text Tricks","The bundled content appears free of hidden-steering tricks, HTML comments, or invisible Unicode characters in its documentation and scripts.",{"category":87,"check":88,"severity":23,"summary":89},"Hooks","Opaque code execution","The skill's scripts (`.py`) are plain Python and not obfuscated, base64-decoded, or fetched at runtime.",{"category":91,"check":92,"severity":23,"summary":93},"Portability","Structural Assumption","The scripts reference paths relative to the skill's directory or use standard Python execution (`uv run`), avoiding assumptions about user project structure.",{"category":95,"check":96,"severity":23,"summary":97},"Trust","Issues Attention","With 4 issues opened and 6 closed in the last 90 days, the closure rate is sufficient and maintainer engagement appears adequate.",{"category":99,"check":100,"severity":23,"summary":101},"Versioning","Release Management","The repository has a recent commit date, and while no explicit versioning is found in SKILL.md, the use of `github:huggingface/skills` implies sourcing from a specific commit/tag, and the README guides installation via marketplace.",{"category":103,"check":104,"severity":23,"summary":105},"Code Execution","Validation","The scripts accept typed arguments via `uv run` and command-line parsing, and the documentation outlines expected inputs, suggesting adequate validation.",{"category":65,"check":107,"severity":23,"summary":108},"Unguarded Destructive Operations","The skill is focused on evaluation and does not perform destructive operations like file deletion or infrastructure changes.",{"category":103,"check":110,"severity":23,"summary":111},"Error Handling","The skill's documentation implies proper error handling with troubleshooting tips for common issues like CUDA OOM and backend failures.",{"category":103,"check":113,"severity":62,"summary":114},"Logging","The skill focuses on local execution and does not require a dedicated local audit file for its operations.",{"category":116,"check":117,"severity":23,"summary":118},"Compliance","GDPR","The skill operates on model evaluations and does not appear to handle personal data.",{"category":116,"check":120,"severity":23,"summary":121},"Target market","The extension is globally applicable as it focuses on local model evaluation and does not have regional restrictions.",{"category":91,"check":123,"severity":23,"summary":124},"Runtime stability","The skill uses standard Python execution (`uv run`) and common ML libraries, making it portable across POSIX-compatible systems.",{"category":43,"check":126,"severity":23,"summary":127},"README","The README provides a comprehensive overview of the Hugging Face Skills repository, installation instructions for various agents, and details on available skills.",{"category":32,"check":129,"severity":23,"summary":130},"Tool surface size","The skill primarily uses three main scripts for evaluation, fitting within the ideal range of 3-10 tools.",{"category":39,"check":132,"severity":23,"summary":133},"Overlapping near-synonym tools","The primary tools (`inspect_eval_uv.py`, `inspect_vllm_uv.py`, `lighteval_vllm_uv.py`) have distinct purposes and clearly defined use cases, avoiding functional overlap.",{"category":43,"check":135,"severity":23,"summary":136},"Phantom features","All advertised features related to local evaluation with inspect-ai and lighteval have corresponding scripts and clear usage instructions.",{"category":138,"check":139,"severity":23,"summary":140},"Install","Installation instruction","The README provides clear installation instructions for Claude Code, Codex, Gemini CLI, and Cursor, including copy-paste examples for marketplace registration and skill installation.",{"category":142,"check":143,"severity":23,"summary":144},"Errors","Actionable error messages","The troubleshooting section in SKILL.md provides actionable advice for common errors like CUDA OOM and backend failures.",{"category":146,"check":147,"severity":23,"summary":148},"Execution","Pinned dependencies","The skill utilizes `uv` for dependency management, which implies pinned dependencies via lockfiles for reproducible builds.",{"category":32,"check":150,"severity":62,"summary":151},"Dry-run preview","The skill is focused on running evaluations, which are analytical in nature and do not involve state-changing operations or outbound data transmission.",{"category":153,"check":154,"severity":62,"summary":155},"Protocol","Idempotent retry & timeouts","The skill performs local evaluations and does not involve remote calls or state-changing operations that would require idempotency or timeouts.",{"category":116,"check":157,"severity":23,"summary":158},"Telemetry opt-in","The skill operates locally and does not emit telemetry data.",{"category":39,"check":160,"severity":23,"summary":161},"Name collisions","The skill `huggingface-community-evals` has a distinct name and purpose within the Hugging Face Skills repository.",{"category":39,"check":163,"severity":62,"summary":164},"Hooks-off mechanism","This extension is a plugin that does not appear to utilize hooks in a way that would require a hooks-off mechanism.",{"category":39,"check":166,"severity":62,"summary":167},"Hook matcher tightness","This extension does not appear to use hooks.",{"category":65,"check":169,"severity":62,"summary":167},"Hook security",{"category":87,"check":171,"severity":62,"summary":172},"Silent prompt rewriting","This extension does not have a `UserPromptSubmit` hook.",{"category":65,"check":174,"severity":62,"summary":175},"Permission Hook","This extension does not appear to use `PermissionRequest` hooks.",{"category":116,"check":177,"severity":62,"summary":178},"Hook privacy","This extension does not appear to use hooks that involve logging or telemetry over the network.",{"category":103,"check":180,"severity":62,"summary":167},"Hook dependency",{"category":43,"check":182,"severity":23,"summary":183},"Feature Transparency","The SKILL.md clearly describes the capabilities and limitations of the `huggingface-community-evals` skill.",{"category":185,"check":186,"severity":23,"summary":187},"Convention","Layout convention adherence","The repository follows standard conventions, with `SKILL.md` files in skill directories and installation instructions provided.",{"category":185,"check":189,"severity":62,"summary":190},"Plugin state","The skill operates locally and does not appear to manage persistent state that would require `${CLAUDE_PLUGIN_DATA}`.",{"category":65,"check":192,"severity":23,"summary":193},"Keychain-stored secrets","The skill mentions `HF_TOKEN` which would likely be managed via environment variables or user config, not directly in settings.json.",{"category":195,"check":196,"severity":23,"summary":197},"Dependencies","Tagged release sourcing","The skill is part of the `huggingface/skills` repository and is installed via marketplace or direct URL, implying sourcing from tagged releases or main branch.",{"category":199,"check":200,"severity":23,"summary":201},"Installation","Clean uninstall","The skill operates locally and does not install background daemons or services that would survive a clean uninstall.",1778690920501,"This plugin provides Python scripts to run local model evaluations using `inspect-ai` and `lighteval`, supporting various inference backends like vLLM, Transformers, and `accelerate`. It allows users to test models on Hugging Face Hub using specified tasks and hardware configurations, with clear guidance on when to use each script and how to troubleshoot common issues. It explicitly states it does not handle remote execution or direct model card management.",[205,206,207,208,209],"Run local evaluations with inspect-ai","Run local evaluations with lighteval","Support for vLLM, Transformers, and accelerate backends","Guidance on task selection and hardware requirements","Troubleshooting for common evaluation issues",[211,212,213,214],"Orchestrating evaluations on Hugging Face Jobs","Directly editing Hugging Face model cards or publishing results","Automating community-evals workflows","Replacing remote Hugging Face compute infrastructure","3.0.0","4.4.0","To enable developers and researchers to run and manage AI model evaluations efficiently on their local hardware, facilitating model selection and comparison.","The extension is well-documented, production-ready, and has a clear single responsibility. The only minor point is the absence of a specific version tag within the SKILL.md, but installation via marketplace implies stable sourcing.",98,"High-quality plugin for local model evaluation with clear documentation and robust functionality.",[222,223,224,225,226,227,228],"evaluation","huggingface","ml","python","vllm","lighteval","inspect-ai","global","verified",[232,233,234,235],"Quickly test models from Hugging Face Hub locally","Compare model performance using standard benchmarks","Choose the best inference backend (vLLM, Transformers) for local GPU evaluations","Debug and troubleshoot evaluation setups before scaling to remote jobs",{"codeQuality":237,"collectedAt":239,"documentation":240,"maintenance":243,"security":249,"testCoverage":251},{"hasLockfile":238},false,1778690901617,{"descriptionLength":241,"readmeSize":242},214,9821,{"closedIssues90d":244,"forks":245,"hasChangelog":238,"openIssues90d":246,"pushedAt":247,"stars":248},6,663,4,1778593131000,10482,{"hasNpmPackage":238,"license":250,"smitheryVerified":238},"Apache-2.0",{"hasCi":252,"hasTests":238},true,{"updatedAt":254},1778690920957,{"basePath":256,"githubOwner":223,"githubRepo":257,"locale":17,"slug":12,"type":258},"skills/huggingface-community-evals","skills","plugin",{"_creationTime":260,"_id":261,"community":262,"display":263,"identity":268,"parentExtension":271,"providers":272,"relations":288,"tags":290,"workflow":291},1778690773482.4824,"k17es3r8wd37t5rrwqcpp5kwrh86mxx8",{"reviewCount":8},{"description":264,"installMethods":265,"name":267,"sourceUrl":13},"Agent Skills for AI/ML tasks including dataset creation, model training, evaluation, and research paper publishing on Hugging Face Hub",{"claudeCode":266},"huggingface/skills","huggingface-skills",{"basePath":269,"githubOwner":223,"githubRepo":257,"locale":17,"slug":257,"type":270},"","marketplace",null,{"evaluate":273,"extract":282},{"promptVersionExtension":274,"promptVersionScoring":216,"score":275,"tags":276,"targetMarket":229,"tier":230},"3.1.0",95,[277,223,278,279,280,281],"ai-ml","datasets","models","research","developer-tools",{"commitSha":283,"marketplace":284,"plugin":286},"HEAD",{"name":267,"pluginCount":285},14,{"mcpCount":8,"provider":287,"skillCount":8},"classify",{"repoId":289},"kd72xwt5xnc0ktc4p7smzfcp3986m959",[277,278,281,223,279,280],{"evaluatedAt":292,"extractAt":293,"updatedAt":292},1778690814090,1778690773482,{"evaluate":295,"extract":297},{"promptVersionExtension":215,"promptVersionScoring":216,"score":219,"tags":296,"targetMarket":229,"tier":230},[222,223,224,225,226,227,228],{"commitSha":283},{"parentExtensionId":261,"repoId":289},{"_creationTime":300,"_id":289,"identity":301,"providers":302,"workflow":741},1778689536128.5474,{"githubOwner":223,"githubRepo":257,"sourceUrl":13},{"classify":303,"discover":734,"github":737},{"commitSha":283,"extensions":304},[305,318,327,335,343,351,356,364,372,380,388,396,404,412,420,428,471,480,486,492,509,515,522,564,575,594,600,620,632,656,714],{"basePath":269,"description":264,"displayName":267,"installMethods":306,"rationale":307,"selectedPaths":308,"source":317,"sourceLanguage":17,"type":270},{"claudeCode":266},"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":258},"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":258},"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":258},"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":258},"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":256,"description":10,"displayName":12,"installMethods":352,"rationale":353,"selectedPaths":354,"source":317,"sourceLanguage":17,"type":258},{"claudeCode":12},"inline plugin source from marketplace.json at skills/huggingface-community-evals",[355],{"path":326,"priority":316},{"basePath":357,"description":358,"displayName":359,"installMethods":360,"rationale":361,"selectedPaths":362,"source":317,"sourceLanguage":17,"type":258},"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":359},"inline plugin source from marketplace.json at skills/huggingface-best",[363],{"path":326,"priority":316},{"basePath":365,"description":366,"displayName":367,"installMethods":368,"rationale":369,"selectedPaths":370,"source":317,"sourceLanguage":17,"type":258},"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":367},"inline plugin source from marketplace.json at skills/hf-cli",[371],{"path":326,"priority":316},{"basePath":373,"description":374,"displayName":375,"installMethods":376,"rationale":377,"selectedPaths":378,"source":317,"sourceLanguage":17,"type":258},"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":375},"inline plugin source from marketplace.json at skills/huggingface-trackio",[379],{"path":326,"priority":316},{"basePath":381,"description":382,"displayName":383,"installMethods":384,"rationale":385,"selectedPaths":386,"source":317,"sourceLanguage":17,"type":258},"skills/huggingface-datasets","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.","huggingface-datasets",{"claudeCode":383},"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":258},"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":258},"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":258},"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":258},"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":258},"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":269,"description":264,"displayName":267,"installMethods":429,"license":250,"rationale":430,"selectedPaths":431,"source":317,"sourceLanguage":17,"type":258},{"claudeCode":267},"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":266},"SKILL.md frontmatter at hf-mcp/skills/hf-mcp/SKILL.md",[478],{"path":326,"priority":311},"skill",{"basePath":365,"description":481,"displayName":367,"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":266},"SKILL.md frontmatter at skills/hf-cli/SKILL.md",[485],{"path":326,"priority":311},{"basePath":357,"description":487,"displayName":359,"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":266},"SKILL.md frontmatter at skills/huggingface-best/SKILL.md",[491],{"path":326,"priority":311},{"basePath":256,"description":493,"displayName":12,"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":266},"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":381,"description":510,"displayName":383,"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":266},"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":266},"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":266},"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":266},"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":266},"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":266},"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":266},"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":373,"description":621,"displayName":375,"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":266},"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":266},"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":266},"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":266},"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":244,"description":738,"forks":245,"homepage":739,"license":250,"openIssues90d":246,"pushedAt":247,"readmeSize":242,"stars":248,"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,[222,223,228,227,224,225,226],{"evaluatedAt":254,"extractAt":293,"updatedAt":254},[],[749,770,790,805,825,843],{"_creationTime":750,"_id":751,"community":752,"display":753,"identity":756,"providers":757,"relations":766,"tags":767,"workflow":768},1778690773482.4834,"k179sm2kkyd7r7nz9jsx62jm9x86mw4a",{"reviewCount":8},{"description":345,"installMethods":754,"name":755,"sourceUrl":13},{"claudeCode":346},"Hugging Face Papers",{"basePath":344,"githubOwner":223,"githubRepo":257,"locale":17,"slug":346,"type":258},{"evaluate":758,"extract":765},{"promptVersionExtension":215,"promptVersionScoring":216,"score":759,"tags":760,"targetMarket":229,"tier":230},100,[223,761,762,763,280,764],"papers","arxiv","ai","metadata",{"commitSha":283,"license":250},{"parentExtensionId":261,"repoId":289},[763,762,223,764,761,280],{"evaluatedAt":769,"extractAt":293,"updatedAt":769},1778690901306,{"_creationTime":771,"_id":772,"community":773,"display":774,"identity":776,"providers":777,"relations":786,"tags":787,"workflow":788},1778690773482.4844,"k1785s3263hzxs6ne1nsbzas5186m86v",{"reviewCount":8},{"description":374,"installMethods":775,"name":375,"sourceUrl":13},{"claudeCode":375},{"basePath":373,"githubOwner":223,"githubRepo":257,"locale":17,"slug":375,"type":258},{"evaluate":778,"extract":785},{"promptVersionExtension":215,"promptVersionScoring":216,"score":779,"tags":780,"targetMarket":229,"tier":230},99,[781,782,783,223,225,784],"mlops","experiment-tracking","monitoring","cli",{"commitSha":283},{"parentExtensionId":261,"repoId":289},[784,782,223,781,783,225],{"evaluatedAt":789,"extractAt":293,"updatedAt":789},1778690988981,{"_creationTime":791,"_id":792,"community":793,"display":794,"identity":796,"providers":797,"relations":801,"tags":802,"workflow":803},1778690773482.4841,"k172m7ench1yy64c6nbcx5pmm586mphh",{"reviewCount":8},{"description":366,"installMethods":795,"name":367,"sourceUrl":13},{"claudeCode":367},{"basePath":365,"githubOwner":223,"githubRepo":257,"locale":17,"slug":367,"type":258},{"evaluate":798,"extract":800},{"promptVersionExtension":215,"promptVersionScoring":216,"score":779,"tags":799,"targetMarket":229,"tier":230},[223,784,281,781,278,279],{"commitSha":283},{"parentExtensionId":261,"repoId":289},[784,278,281,223,781,279],{"evaluatedAt":804,"extractAt":293,"updatedAt":804},1778690967596,{"_creationTime":806,"_id":807,"community":808,"display":809,"identity":811,"providers":812,"relations":821,"tags":822,"workflow":823},1778690773482.483,"k172w5kkdn7117xpqyyhcqsww186n17b",{"reviewCount":8},{"description":329,"installMethods":810,"name":330,"sourceUrl":13},{"claudeCode":330},{"basePath":328,"githubOwner":223,"githubRepo":257,"locale":17,"slug":330,"type":258},{"evaluate":813,"extract":820},{"promptVersionExtension":215,"promptVersionScoring":216,"score":779,"tags":814,"targetMarket":229,"tier":230},[815,816,817,818,223,819],"llm","local-models","llama-cpp","gguf","ml-ops",{"commitSha":283},{"parentExtensionId":261,"repoId":289},[818,223,817,815,816,819],{"evaluatedAt":824,"extractAt":293,"updatedAt":824},1778690852474,{"_creationTime":826,"_id":827,"community":828,"display":829,"identity":831,"providers":832,"relations":839,"tags":840,"workflow":841},1778690773482.4827,"k175sdyyg60kjnnbkm4029s70186ngbn",{"reviewCount":8},{"description":320,"installMethods":830,"name":321,"sourceUrl":13},{"claudeCode":321},{"basePath":319,"githubOwner":223,"githubRepo":257,"locale":17,"slug":321,"type":258},{"evaluate":833,"extract":838},{"promptVersionExtension":215,"promptVersionScoring":216,"score":779,"tags":834,"targetMarket":229,"tier":230},[815,835,836,223,837,818,781],"training","fine-tuning","trl",{"commitSha":283},{"parentExtensionId":261,"repoId":289},[836,818,223,815,781,835,837],{"evaluatedAt":842,"extractAt":293,"updatedAt":842},1778690836703,{"_creationTime":844,"_id":845,"community":846,"display":847,"identity":852,"providers":856,"relations":864,"tags":867,"workflow":868},1778699018122.7842,"k17esy2nrsh2regbf65s788sfn86mg0e",{"reviewCount":8},{"description":848,"installMethods":849,"name":850,"sourceUrl":851},"Three-layer quality evaluation framework for Claude Code plugins with Elo ranking",{"claudeCode":850},"plugin-eval","https://github.com/wshobson/agents",{"basePath":853,"githubOwner":854,"githubRepo":855,"locale":17,"slug":850,"type":258},"plugins/plugin-eval","wshobson","agents",{"evaluate":857,"extract":861},{"promptVersionExtension":215,"promptVersionScoring":216,"score":219,"tags":858,"targetMarket":229,"tier":230},[222,859,860,815,784],"quality","testing",{"commitSha":283,"plugin":862},{"mcpCount":8,"provider":287,"skillCount":863},1,{"parentExtensionId":865,"repoId":866},"k17cywe30jfsfw3cdpncjfn8y186nvyw","kd74de64zj0axtg5b8t7eqqe2x86nske",[784,222,815,859,860],{"evaluatedAt":869,"extractAt":870,"updatedAt":869},1778700256992,1778699018122]