Trusted skills for AI agents from official vendors and established maintainers. Every skill is attributed to its author, scanned, and comes with install steps for your AI tool.
706 skills
NVIDIAOfficial
Train, evaluate, and deploy lightweight SegFormer transformer models for real-time semantic segmentation tasks.
NVIDIAOfficial
Train, evaluate, and deploy TAO action recognition models for video sequences using RGB, optical flow, or joint inputs.
NVIDIAOfficial
Train, evaluate, and deploy OCDNet models for arbitrary-oriented scene text detection.
NVIDIAOfficial
Generate image embeddings from file paths using CLIP, SigLIP, or TAO checkpoints for downstream data mining.
NVIDIAOfficial
Fine-tunes clinical ASR models using NeMo and validates performance via offline evaluation.
NVIDIAOfficial
Manage the full lifecycle of NVIDIA TAO OCRNet models for scene text recognition.
NVIDIAOfficial
A guide for fetching weather and climate data using Earth2Studio data source APIs.
NVIDIAOfficial
Fine-tune manifest-backed GR00T or openpi remote tasks using compatible LeRobot datasets.
NVIDIAOfficial
Train, evaluate, and deploy PointPillars 3D object detection models for LiDAR point clouds.
NVIDIAOfficial
Orchestrates containerized continued pretraining for KERMT models on custom SMILES corpora.
NVIDIAOfficial
Train, evaluate, and export Depth Anything v2 monocular depth estimation models using the NVIDIA TAO toolkit.
NVIDIAOfficial
Assists with developing, debugging, and upgrading Python code using NVIDIA cuML for GPU-accelerated machine learning.
NVIDIAOfficial
A conceptual guide for translating business problems into linear, mixed-integer, and quadratic optimization formulations.
NVIDIAOfficial
Official NVIDIA-authored guidance for NVIDIA cuDF GPU DataFrames, pandas acceleration, dask-cuDF, ETL, joins, groupby, CSV/Parquet I/O, nullable semantics, and multi-GPU DataFrame workloads.
NVIDIAOfficial
A workflow for querying and validating grounded answers from the Auto Ontology system using MCP or REST.
NVIDIAOfficial
Routes ambiguous NVFLARE requests to the appropriate specialized workflow skill based on project evidence.
NVIDIAOfficial
Practical performance optimization and training strategies for Mixture-of-Experts (MoE) Vision Language Models (VLMs).
NVIDIAOfficial
A two-step pipeline that uses vision-language models to extract referring expressions from captions and ground them to pixel-space bounding boxes.
NVIDIAOfficial
Orchestrate Cosmos Data Factory (CDF) pipelines for video and image augmentation, generation, and evaluation.
NVIDIAOfficial
Discover Earth2Studio models, data sources, and workflow examples for weather and climate research.
NVIDIAOfficial
Orchestrates the end-to-end finetuning of KERMT encoder models on labeled chemical datasets.
NVIDIAOfficial
NVIDIA TAO skill for self-supervised visual representation learning using NVDINOv2.
NVIDIAOfficial
A multi-phase pipeline for fine-tuning AnomalyGen models and generating synthetic anomaly images.
NVIDIAOfficial
A workflow for training, evaluating, and deploying Grounding DINO models for open-vocabulary object detection.