huggingface Skills

Fixes broken typing checks detected by ty, make typing, or make check-repo. Use when typing errors appear in local runs, CI, or PR logs.

> Use when the user has written (or wants to write) a `ModularPipelineBlocks` subclass in a local Python file and needs to package it into a Hub-uploadable directory. Covers the workflow from a single `block.py` file to a published custom-block repo that consumers can load via `ModularPipeline.from_pretrained(<repo>, trust_remote_code=True)`.

> Use when the user wants to run a diffusers pipeline from a terminal (one-off generation, batch jobs, smoke-testing a new model), run on HF Sandbox hardware via `--remote`, introspect a pipeline's input schema before calling it, or attach a LoRA at inference time. Prefer this over writing ad-hoc Python scripts for generation tasks.

> Use when adding a new model or pipeline to diffusers, setting up file structure for a new model, converting a pipeline to modular format, or converting weights for a new version of an already-supported model.

> Use before opening a PR, or whenever asked to self-review a diffusers contribution. Applies the same rubric as the `@claude` CI (checks the diff against .ai/review-rules.md, traces call paths for dead code). Reports findings grouped by severity, flagging what to fix before submitting (blocking issues + dead code) vs what to leave for the actual review. Report-only — does not edit files.

Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands.

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.

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`.

Discover the user's local AWS context (active profile, region, account ID, caller identity) at the start of any AWS task. Use this skill before any other AWS work — deploying to SageMaker, creating resources, calling AWS APIs, or anything that touches an AWS account. Use it especially when the user has not specified a region or profile explicitly, when they say things like "use my AWS account", "deploy to AWS", "use my profile", or when about to make any AWS CLI or SDK call. Never guess the region or account ID — always use this skill to read it from the local configuration first.

Set up an isolated Python environment for SageMaker / AWS work, with the right Python version and current boto3. Use this skill whenever Python code will be executed for a SageMaker deployment, training job, or any AWS automation — including when about to run `pip install`, when about to invoke `boto3`, when creating or activating a virtualenv, or when the user asks to "set up the environment". Never use system Python and never `pip install` into it. Always isolate. This skill prevents the most common failure modes: wrong Python version, dependency conflicts, and stale SDKs.

Plan and coordinate the deployment of a model to Amazon SageMaker AI. Use this skill whenever the user wants to deploy, host, serve, or expose a model on SageMaker or AWS — including phrases like "deploy a model", "host this LLM on AWS", "serve this embedding model", "deploy a reranker", "deploy a text-to-image / diffusion model", "host this for async inference", "create an endpoint", "serve my fine-tuned model", or any request that involves making a model available for inference on AWS. Use this even when the user is vague (e.g. "I just want to get this running on AWS, you figure it out"). Works for text-generation LLMs, embedding models, rerankers, classifiers, text-to-image / diffusion models — picks the right serving stack and chooses between real-time and async inference. This is the entry-point skill for SageMaker deployment work — it asks clarifying questions, picks a deployment pathway, and coordinates the other deployment skills.

Ensure a usable SageMaker execution role exists before deploying or training. Use this skill whenever about to create a SageMaker endpoint, model, training job, or any resource that requires an execution role. Use it especially when the user has not provided a role ARN explicitly, when scripts are about to call `iam:CreateRole`, or when an AccessDenied error mentions an IAM action. Never blindly call `iam:CreateRole` — always check for existing roles first. This skill prevents the most common SageMaker deployment failure: trying to create IAM resources from an SSO principal that has no IAM write permissions.

Create a SageMaker endpoint (real-time, real-time scale-to-zero, or async) with autoscaling, CloudWatch alarms, and tagging enabled by default. Use this skill whenever about to create a SageMaker endpoint, write deployment code that calls `create_endpoint`, or finalize a deployment after the image URI and IAM role are known. Provides deploy.py for real-time endpoints, deploy_ic.py for real-time endpoints that scale to zero instances via inference components, and deploy_async.py for async endpoints (also scale-to-zero). This is the last step in the SageMaker deployment workflow. Never generate a bare `create_endpoint` call without these defaults — endpoints without autoscaling or alarms are demos, not deployments.

Pick the right serving container for a SageMaker model deployment and find its current image URI. Use this skill whenever about to deploy a model to a SageMaker endpoint and an image URI needs to be chosen — including when the user says "deploy this LLM", "host this HuggingFace model", "serve this fine-tuned model", "deploy this embedding model", "host a reranker", "serve a sentence-transformers model", or when about to hardcode any container URI in deployment code. HuggingFace-curated Deep Learning Containers are ALWAYS preferred: HuggingFace vLLM (LLMs and generative rerankers), HuggingFace vLLM-Omni (multimodal), TEI (embeddings/cross-encoder rerankers), HF Inference Toolkit (other transformers). Generic images (AWS vLLM, DJL-LMI, SGLang) are used only when no HuggingFace image is compatible — never merely because they carry a newer version. Never hardcode a container URI from memory and never default to TGI. Prevents stale-image failures and wrong-region URIs.

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.

Hugging Face CLI to estimate the required memory to load Safetensors or GGUF model weights for inference from the Hugging Face Hub

>- 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. "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.

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.

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

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.

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.

Build and publish a Gradio demo on Hugging Face Spaces for a user-provided LoRA. Use when someone asks to create, generate, ship, or publish a Space, demo, Gradio app, or playground for a LoRA — including LoRAs for Qwen-Image, Qwen-Image-Edit, LTX-Video, Wan, FLUX, SDXL, or other diffusion base models. Also triggers when someone describes a LoRA they trained or hosts on the Hub and wants to share it. Covers picking the right base pipeline and `diffusers` inference recipe, designing a UI tailored to the LoRA's task and inputs (Union/multi-task control, edit, video, image, etc.), respecting model-card recommendations (trigger words, steps, guidance, LoRA scale, example inputs), and shipping to ZeroGPU hardware as a private Space by default.

Build, deploy, and maintain applications on Hugging Face Spaces — Gradio / Docker / Static SDKs, ZeroGPU and dedicated hardware, model loading, debugging, buckets, inference providers, community grants. Use whenever the user asks to create or host an app on Hugging Face, port code onto ZeroGPU, fix a Space that won't build or run, or otherwise work with `hf spaces …`, `@spaces.GPU`, Space README frontmatter, or the `spaces` Python package.

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.

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.

AI demos and GPU compute with Gradio Spaces and Hugging Face Spaces ZeroGPU. Use when writing or reviewing code that uses `@spaces.GPU`, configuring `python_version` or `requirements.txt` for a ZeroGPU Space, or handling ZeroGPU-specific code constraints — pickle-based process isolation, `gr.State` semantics across the worker boundary, no `torch.compile` (use AoTI instead), CUDA wheel-only builds (no `nvcc` at build or runtime), large vs xlarge sizing, and dynamic duration callables. Make sure to use this skill whenever the user mentions ZeroGPU, `@spaces.GPU`, or the `spaces` Python package, or hits ZeroGPU-specific code errors like `PicklingError` across the worker boundary, `illegal duration`, or `flash-attn` wheel-build failures — even when the user does not explicitly ask for ZeroGPU coding guidance. Trigger on `import spaces` or `@spaces.GPU` in code.

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.

Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands.

Sync chat-ui's model config with the HuggingFace router — add descriptions for new models, flag reasoning-capable ones, enable artifacts for models with 32B+ parameters, and prune deprecated models the router no longer serves. Use when models are released or removed on the router and prod.yaml/dev.yaml need syncing. Triggers on requests like "add new model descriptions", "update models from router", "sync models", "remove deprecated models", "prune models no longer on the router", or when explicitly invoking /sync-models.

Generate Hugging Face Hub (huggingface_hub) release notes from cached PR JSON files. Use when asked to draft release notes from PR files.

Use when adding or migrating non-thumbnail images for a Hugging Face Blog post. Uploads body images to the Hugging Face documentation-images dataset, updates the Markdown, and verifies the new links.

Review code changes for bugs and alignment with OpenEnv principles and RFCs. Use when reviewing PRs, checking code before commit, or when asked to review changes. Implements two-tier review model.

Deploy an OpenEnv environment to Hugging Face Spaces. Use when asked to deploy, push to Hugging Face, or update a space.

Generate OpenEnv environments from a concrete use case (for example, "generate an env for the library textarena"). Use when asked to design or implement a new environment under envs/ by researching a target library/API, selecting matching OpenEnv examples, asking key implementation questions, and building models/client/server/openenv.yaml. Do not use for model training or evaluation tasks.

Hugging Face Hub CLI (`hf`) for downloading, uploading, and managing repositories, models, datasets, and Spaces on the Hugging Face Hub. Replaces now deprecated `huggingface-cli` command.

Diagnose and recover failing or stuck Hugging Face Space deployments for OpenEnv environments. Use when deploying envs from `envs/` to the Hub (`openenv` namespace with version suffixes), when Spaces are in `BUILDING`/`APP_STARTING`/`RUNTIME_ERROR`, or when release collections need to be reconciled after targeted redeploys.

Make tests pass. Invoke after /write-tests produces failing tests.

OpenEnv CLI (`openenv`) for scaffolding, validating, building, and pushing OpenEnv environments.

Validate changes before submitting a pull request. Run comprehensive checks including lint, tests, alignment review, and RFC analysis. Use before creating a PR, when asked if code is ready for review, or before pushing for PR.

Release workflow for deploying OpenEnv environments to Hugging Face Spaces and keeping canonical references in sync.

Determine if proposed changes require an RFC. Use when planning significant changes, before starting major work, or when asked whether an RFC is needed.

Refactor code after tests pass. The "Refactor" phase of Red-Green-Refactor.

Work on a batch of GitHub issues in parallel using Agent Teams. Creates one worktree per issue with TDD enforcement, coordinates via a lead agent, then produces stacked PRs.

Update documentation across the repo after API changes. Finds stale references in docs, examples, docstrings, and fixes them.

Monitor a PR's CI checks and Greptile code review after submission. Polls CI status, auto-fixes failures via ralph-loop, waits for Greptile review, addresses comments, and iterates until green.

Start work on a GitHub issue. Extracts requirements, creates worktree, sets up TDD workflow.

Write failing tests from requirements. Invoke for each todo before /implement.

Provides guidance for writing, optimizing, and benchmarking C++ CPU kernels with SIMD intrinsics (AVX2/AVX512) for the Hugging Face kernels ecosystem. Includes a two-phase workflow: Phase 1 correctness (generic → AVX2) and Phase 2 performance exploration (AVX512 with branching trial loop), runtime CPU dispatch, OpenMP threading, and brgemm integration for GEMM-heavy kernels.

Provides guidance for writing and benchmarking optimized CUDA kernels for NVIDIA GPUs (H100, A100, T4) targeting HuggingFace diffusers and transformers libraries. Kernels must be kernel-builder/ABI3-compliant: no pybind11, no setup.py, TORCH_LIBRARY_EXPAND bindings only. Supports models like LTX-Video, Stable Diffusion, LLaMA, Mistral, and Qwen. Includes integration with HuggingFace Kernels Hub (get_kernel) for loading pre-compiled kernels. Includes benchmarking scripts to compare kernel performance against baseline implementations.

Provides guidance for writing and benchmarking optimized Triton kernels for AMD GPUs (MI355X, R9700) on ROCm, targeting HuggingFace diffusers (LTX-Video, SD3, FLUX) and transformers. Core kernels: RMSNorm, RoPE 3D, GEGLU, AdaLN. Includes XCD swizzle, autotune, diffusers integration patterns, and LTX-Video pipeline injection.

Provides guidance for writing, optimizing, and benchmarking Triton kernels for Intel XPU GPUs (Battlemage/Arc Pro B50) using the Xe-Forge optimization framework. Includes an LLM-driven trial-loop workflow (analyze, validate, benchmark, profile, finalize), XPU-specific patterns (tensor descriptors, GRF mode, tile swizzling), KernelBench fused kernels, and Flash Attention.

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`.

Use Hermes delegate_task cleanly in this repo for planner, reviewer, researcher, reporter, experiment-worker, and memory-keeper roles.

Run one Autolab benchmark experiment safely on Hugging Face Jobs. Use when a planner, reviewer, or experiment worker is preparing, auditing, launching, or reviewing a single train.py hypothesis against the current local promoted master.

Operate the local Trackio reporter for Autolab HF Jobs. Use when a reporter or planner needs to inspect scores, active jobs, worker anomalies, duplicate launches, or the overall experiment board.

Hugging Face Hub CLI (`hf`) for downloading, uploading, and managing repositories, models, datasets, and Spaces on the Hugging Face Hub. Replaces now deprecated `huggingface-cli` command.

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.

Convert Hugging Face transformers language models to MLX format for Apple Silicon inference. Use when asked to convert a model from transformers to MLX, port a model to MLX, create an MLX implementation of a model, or test/validate an MLX model conversion. Triggers include mentions of "convert to MLX", "MLX implementation", "port to MLX", "mlx-lm", or requests involving both transformers and MLX frameworks.

Investigates an Autophysicist workspace run. Use to understand what went wrong and could be improved in the single-agent iterative research process.

Investigates PhysicsIntern workspace run. Use to understand what went wrong and could be improved in the multi-agent research process.

Papers With Code CLI (`pwc`) for searching and reading AI/ML papers, discovering recent and trending research, finding related work and paper lineage, browsing tasks, methods, conferences, and benchmark leaderboards through the public Papers With Code catalog. Use whenever the user asks to find papers, survey literature, compare research, inspect an arXiv paper, explore AI/ML taxonomy or conferences, discover benchmarks or state-of-the-art models, or mentions Papers With Code, `pwc`, or the `pwc-cli`. Prefer this skill for grounded AI/ML research even when the user does not explicitly ask for a CLI command.

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.

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.

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.

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.