Skills from NVIDIA’s repositories that passed our trust, licence and security gates. Each keeps its author’s attribution.
| Repository | Tier | Verdict | Authorship | Checked | Skills |
|---|---|---|---|---|---|
| NVIDIA/skills | Vendor | Approved · created 2026-02-25, pushed 2026-11-06, 3888 stars | Written in-house | 2026-11-07 | 431 |
Files are mirrored only where the licence allows redistribution; otherwise we link to the source.
NVIDIAOfficial
Configure and validate CPU offloading mechanisms in Megatron Bridge to optimize GPU memory usage during model training.
NVIDIAOfficial
A guide for selecting and tuning MoE token dispatchers (alltoall, DeepEP, HybridEP) based on hardware, EP degree, and workload.
NVIDIAOfficial
A utility for converting single-series CT DICOM folders into Hounsfield Unit NIfTI volumes with affine metadata.
NVIDIAOfficial
Official guidance for integrating PhysicsNeMo ShardTensor domain parallelism into PyTorch training and inference workflows.
NVIDIAOfficial
The i4h-workflow skill acts as a central router and orientation guide for the Isaac for Healthcare (i4h) robotics development environment.
NVIDIAOfficial
Converts cuTile GPU kernels (@ct.kernel) to Triton (@triton.jit). Handles standard in-repo conversion, debugging (cudaErrorIllegalAddress, shape mismatch, numerical mismatch), and mapping cuTile idioms (ct.load/ct.store, ct.Constant, ct.launch) to Triton equivalents. Covers dual-kernel layout flags (e.g. transpose=True/False + autotune grid via META) per translations/advanced-patterns.md. Use when converting, porting, or translating cuTile kernels to Triton, or debugging existing Triton translations.
NVIDIAOfficial
Bootstrap and orchestrate the TAO skill bank for individual skill installations.
NVIDIAOfficial
Use when turning a Jetson encoder use case into one surface-neutral recipe with native Video Codec SDK and PyNvVideoCodec projections.
NVIDIAOfficial
Automated visual grading and filtering of HDF5 workflow recordings using OpenAI-compatible vision models.
NVIDIAOfficial
Diagnose and resolve incorrect gradients in differentiable NVIDIA Warp programs using systematic verification and taping pattern analysis.
NVIDIAOfficial
A structured workflow for maintaining durable, human-readable session state to enable seamless agent continuity across disconnects and long-running tasks.
NVIDIAOfficial
Orchestrates the customization and fine-tuning pipeline for NVIDIA Nemotron Speech (Riva) ASR models.
NVIDIAOfficial
Automates an iterative smart-data-augmentation loop for NVIDIA TAO Grounding DINO object detection models.
NVIDIAOfficial
Configures PCIe controllers for Jetson Thor and Orin custom carriers by generating kernel-DT overlays.
NVIDIAOfficial
Prepare and execute LoRA fine-tuning for AnomalyGenNext models using validated datasets and canonical recipes.
NVIDIAOfficial
Analyze VLM binary classification predictions to identify and report false-positive and false-negative failure cases.
NVIDIAOfficial
Guidance for implementing hardware-offloaded Reed-Solomon erasure coding using the NVIDIA DOCA SDK.
NVIDIAOfficial
Run and operate the DeepStream Multi-View 3D Tracking reference app, also known as MV3DT. Use when the user asks to set up prerequisites, run shipped MV3DT samples, run Multi-View 3D Tracking on custom synchronized MP4 datasets, import camera calibration, delegate missing calibration to AutoMagicCalib, inspect OSD or BEV visualization, consume MV3DT Kafka metadata, or clean up MV3DT run state in the DeepStream MV3DT app directory.
NVIDIAOfficial
Generates a snapshot markdown knowledge base for a Jetson target platform by scanning BSP and source directories.
NVIDIAOfficial
Manage the lifecycle of DINO transformer-based 2D object detection models within the TAO ecosystem.
NVIDIAOfficial
Configure and tune Mixture-of-Experts (MoE) communication overlap in Megatron Bridge for improved training throughput.
NVIDIAOfficial
Execute NVIDIA TAO Toolkit training and inference containers on provisioned NVIDIA Brev GPU instances.
NVIDIAOfficial
Initializes the Clinical ASR Flywheel environment by verifying API access, installing dependencies, and running a speech-to-text smoke test.
NVIDIAOfficial
Evaluates clinical ASR performance using NeMo manifests and generates a detailed KER-focused leaderboard for model assessment.