Install every skill from this repository$ npx skills add nvidia/skills
Repository directory
Skills
199 of 199
#Skill8W activityInstalls
1rag-blueprintNVIDIA RAG Blueprint — deploy, configure, troubleshoot, and manage. Handles any RAG action: deploy, install, start, enable, disable, toggle, change, configure, troubleshoot, debug, fix, shutdown, stop, or tear down any RAG feature or service (Agentic RAG, VLM, guardrails, query rewriting, models, search, ingestion, observability, summarization, reasoning, and more).3.1K2accelerated-computing-cudfOfficial 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.3K3aiq-research|2.9K4cuopt-developerModify, build, test, debug, and contribute to NVIDIA cuOpt (C++/CUDA, Python, server, CI). Use for solver internals, PRs, DCO, and code conventions.2.7K5aiq-deploy|2.6K6cuopt-installInstall cuOpt for Python, C, or server via pip, conda, or Docker; verify the install. For building cuOpt from source, see cuopt-developer.2.5K7cudaq-guideCUDA-Q onboarding guide for installation, test programs, GPU simulation, QPU hardware, and quantum applications.2.5K8data-designerUse when the user wants to create a dataset, generate synthetic data, or build a data generation pipeline.2.4K9deepstream-devNVIDIA DeepStream SDK development with Python pyservicemaker API. Use when building video analytics pipelines, GStreamer-based video processing, TensorRT inference integration, object detection/tracking, or Kafka/message broker integration.2.4K10rag-eval>-2.3K11nemo-retrieverUse when the user wants to search, query, extract, transcribe, describe, quote, filter, or aggregate across documents — PDFs, scanned forms / images (`.jpg` `.png` `.tiff`), Office (`.docx` `.pptx`), text (`.html` `.txt`), audio (`.mp3` `.wav` `.m4a`), or video (`.mp4` `.mov`). Prefer this over native Read / Grep for multi-file or non-PDF corpora. Not for: editing files, web browsing, single-file plain-text lookups, fine-tuning.2.3K12omniverse-cad-to-simreadyCoordinate the end-to-end CAD/source-asset to SimReady workflow. Use for broad requests such as CAD to SimReady, source asset to simulation-ready USD, or prop packaging that require conversion, material/physics assignment, SimReady conformance, validation, and optional package creation; deploy or verify Content Agents services first when property assignment is enabled; route single-stage work through nested references.2.3K13deepstream-import-vision-model>2.3K14omniverse-usd-performance-tuningTop-level workflow skill for USD performance diagnosis and optimization. Handles slow loading, high memory, low FPS, and broad scene-optimization requests; delegates auth/runtime setup to Phase 0 owners.2.3K15cuopt-routing-api-pythonVehicle routing (VRP, TSP, PDP) with cuOpt — Python API only. Use when the user is building or solving routing in Python.2.2K16skill-card-generatorUse only to generate or update a governance skill card for a specified existing agent skill directory. Do not use for explaining, listing, comparing, or discussing skill capabilities.2.2K17omniverse-realtime-viewerUse as the top-level router for Omniverse Realtime Viewer USD app requests and focused viewer reference documents.2.2K18rag-perf>-2.1K19cuopt-server-api-pythoncuOpt REST server — start server, endpoints, Python/curl client examples. Use when the user is deploying or calling the REST API.2.1K20cuopt-numerical-optimization-formulationLP, MILP, QP — concepts, problem-text parsing, and formulation patterns (parameters, constraints, decisions, objective). Concepts only; no API.2.1K21physical-ai-neural-reconstructionRouter for NVIDIA NuRec/NRE: USDZ rendering, NCore conversion, 3DGS, gRPC sensor sim, carline adaptation, PhysicalAI HF datasets. Do NOT use for SimReady or infra setup.2.1K22nemotron-customizePlan, configure, and chain repo-native Nemotron customization steps into single-step or multi-step pipelines: curation, translation, SFT/PEFT (AutoModel or Megatron-Bridge), pretraining/CPT, RL alignment (DPO/RLVR/GRPO/RLHF), BYOB/MCQ benchmarks, checkpoint conversion, ModelOpt optimization, env profiles, and evaluation of trained checkpoints or existing/hosted endpoints. Use when a request names a Nemotron step or workflow, or asks to clean, translate, train, fine-tune, align, convert, optimize, evaluate, or compose these into a pipeline. Do NOT use for frontend/dashboard/visualization work, generic ML advice, billing/access, or non-Nemotron coding tasks.2.1K23dali-dynamic-modeDALI imperative dynamic mode (`nvidia.dali.experimental.dynamic`, ndd): use when working on ndd code or migrating pipelines; skip pipeline-only tasks.2.1K24dynamo-recipe-runnerSelect, validate, patch, and deploy existing NVIDIA Dynamo Kubernetes recipes. Use for model/backend/GPU/deployment-mode recipe bring-up; use router-starter for router-only mode work and troubleshoot for broken deployments.2.1K25nemo-rl-session-memoryManage durable working-session memory for coding agents. Use when a user asks to preserve or recover agent context across disconnects, VS Code restarts, long-running work, handoffs, or any session where important state should be written periodically under the repo's session directory. Do NOT use for: simple questions, short tasks, one-off commands, linting, or code review.2K26physical-ai-infrastructure-setup-and-resilient-scaling>-2K27nemotron-speechRoutes NVIDIA Nemotron Speech (Riva) NIM tasks — deploys, runs, and tests ASR, TTS, and NMT NIMs on build.nvidia.com or self-hosted.2K28nemotron-retrieval-recipesUse when planning, debugging, tuning, evaluating, exporting, or deploying public Nemotron `embed`/`rerank` retrieval recipes.2K29cupynumeric-installInstall and verify cuPyNumeric for Python — requirements, commands, verification. Source builds are out of scope.2K30physical-ai-video-data-augmentation>-2K31physical-ai-defect-image-generation>-2K32nemo-automodel-distributed-trainingGuide for selecting and configuring distributed training strategies in NeMo AutoModel, including FSDP2, Megatron FSDP, DDP, and parallelism settings.2K33cupynumeric-migration-readinessPre-migration readiness assessor for porting NumPy to cuPyNumeric. Use BEFORE substantial porting work begins when the user asks whether code will scale on GPU, whether they should migrate to cuPyNumeric, which NumPy patterns transfer cleanly, what must be refactored before porting, or mentions pre-port assessment, scaling analysis, or refactor planning. Inspect the user's source code, look up NumPy usage, cross-reference the cuPyNumeric API support manifest, and distinguish distributed-scaling-friendly patterns from blockers such as unsupported APIs, scalar synchronization, host round-trips, Python/object-heavy control flow, shape/data-dependent branching, and in-place mutation hazards. Produce a verdict of READY, LIGHT REFACTOR, SIGNIFICANT REFACTOR, or NOT RECOMMENDED, with concrete refactor pointers.2K34nemo-automodel-recipe-developmentCreate and modify NeMo AutoModel training and evaluation recipes, including YAML structure, builders, and execution flow.2K35vss-summarize-videoUse to summarize a recorded video via the LVS summarization microservice (HITL-gated) with a VLM fallback. Not for report generation or live RTSP captioning.2K36cupynumeric-parallel-data-loadLoad a sharded, on-disk dataset (sharded .npy, Parquet/Arrow, raw binary, sharded HDF5, custom layouts) into a distributed cuPyNumeric ndarray via a manual partition + leaf @task launch with CPU/OMP/GPU variants. Use when no single-call loader fits, including when per-shard row counts differ across files. Prefer cupynumeric.load or legate.io.hdf5.from_file when they apply.1.9K37nemo-automodel-model-onboardingGuide for onboarding new model architectures into NeMo AutoModel, including architecture discovery, implementation patterns, registration, and validation.1.9K38dynamo-router-starterStart or patch Dynamo router modes and run router endpoint smoke checks. Use for round-robin, KV-aware, least-loaded, or device-aware routing setup; use recipe-runner for recipe deployment and troubleshoot for failure diagnosis.1.9K39nemo-automodel-launcher-configConfigure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.1.9K40dynamo-troubleshootDiagnose failed or unhealthy Dynamo deployments. Use when pods, model-cache jobs, PVCs, workers, frontend/router health, endpoints, or benchmark jobs fail; use recipe-runner/router-starter before this for normal bring-up.1.9K41nemo-rl-auto-researchAutonomous NeMo-RL research agent workflow for directed hypothesis testing and open-ended discovery. Guides agents through the full experiment lifecycle: understanding recipes and environments, wiring RL or NeMo-gym runs, launching reproducible baselines and iterations, analyzing results, preserving human oversight, and using git plus TSV logs as the research ledger. Do NOT use for: bug fixes, code review, documentation, refactoring, dependency updates, or single-file changes.1.9K42physicsnemo-discoverOfficial NVIDIA-authored guidance for navigating PhysicsNeMo — pick the model, datapipe, or example for a SciML/AI4Science task (surrogates, forecasting, downscaling, physics-informed, inverse, generative). Points at existing files via live repo search; never writes code. Do NOT use for installation or environment setup, training-loop or other code authoring/scaffolding, contributor/CI/packaging questions, repo-specific questions in physicsnemo-sym/-cfd/-curator, or general (non-physics) ML/PyTorch.1.9K43vss-ask-videoUse this skill to ask the VSS agent's video_understanding tool a fresh visual question about a recorded clip. Not for prior tool output, search hits, or metadata-answerable questions.1.9K44dicom-metadata-extractUsed for extracting selected metadata from one DICOM file and flagging standard-tag PHI presence. Not for anonymization or clinical use.1.9K45cupynumeric-hdf5>-1.9K46nemo-mbridge-perf-cuda-graphsValidate and use CUDA graph capture in Megatron Bridge, including local full-iteration graphs and Transformer Engine scoped graphs for attention, MLP, and MoE modules.1.9K47nemo-mbridge-perf-memory-tuningTechniques for reducing peak GPU memory in Megatron Bridge — expandable segments, PEFT + SP input re-gather, parallelism resizing, activation recompute, CPU offloading constraints, and common OOM fixes.1.9K48dynamo-interconnect-checkValidate that a Dynamo deployment's NIXL/UCX/NCCL interconnect is ready for disaggregated serving over RDMA/NVLink. Use after recipe-runner brings a deployment up (especially disagg/multi-node) to confirm the KV transport is correct; use troubleshoot for diagnosing already-failed pods.1.9K49dicom-series-preflightUsed for header-only preflight of one DICOM series folder before conversion or inference. Not for de-identification or clinical clearance.1.9K50digital-health-clinical-asr-setupStage 1 of Clinical ASR Flywheel. Use when bootstrapping a cycle: NVCF+MW disclosure, NVIDIA_API_KEY check, deps install, TTS+ASR smoke test.1.9K51vss-search-archiveUse this skill to run top-level VSS fusion search on archived video, or to ingest video files / RTSP streams for search. Do NOT use for ad-hoc visual Q&A (use vss-ask-video), live captioning (use vss-deploy-dense-captioning), or video summarization and reports (use vss-summarize-video).1.9K52earth2studio-install>1.9K53digital-health-clinical-asr-buildStage 2 of the Clinical ASR Flywheel. Use when curating clinical terms, tagging IPA, and synthesizing a NeMo manifest. NOT for scoring (use /digital-health-clinical-asr-eval).1.9K54digital-health-clinical-asr-finetuneStage 4 of the Clinical ASR Flywheel. Use when priority KER is above 0.3 to run stock NeMo SFT on Parakeet TDT v2 and offline cycle N+1 re-eval. NOT for generic word boosting (use /finetune-asr).1.9K55digital-health-clinical-asr-evalStage 3 of Clinical ASR Flywheel. Score a NeMo manifest, produce the five-section KER leaderboard (by-ipa_source diagnostic). Not for ASR auth (/riva-asr).1.9K56launch-nemo-rlPlaybook for launching, monitoring, stopping, and debugging NeMo-RL recipes on a Kubernetes cluster via the nrl-k8s CLI. Covers ephemeral vs long-lived RayCluster modes, iterating on runs, and debugging hung or failed training jobs.1.9K57dicom-series-to-volumeUsed for converting one CT DICOM series folder to a HU NIfTI volume with affine evidence. Not for multi-frame DICOM or clinical use.1.9K58nemo-mbridge-recipe-recommenderRecommend and customize Megatron Bridge library and benchmark recipes for a user's model, GPU count, hardware, sequence length, and pretrain/SFT/PEFT goal. Use when selecting a starting recipe, comparing library and benchmark configs, resizing parallelism for a GPU allocation, or distinguishing convergence changes, semantics-preserving execution tuning, and benchmark-only shortcuts.1.9K59vss-generate-video-reportUse this skill when producing a VSS analysis report — Mode A per-clip VLM, Mode B incident-range via video-analytics. Not for standalone video summarization, real-time alerts or ad-hoc Q&A.1.9K60earth2studio-discover>1.9K61earth2studio-data-fetch>1.9K62earth2studio-deterministic-forecast>1.9K63vss-deploy-detection-tracking-3d>1.9K64nemo-mbridge-perf-activation-recompute>-1.9K65nemo-mbridge-perf-moe-vlm-trainingPractical guidance for training MoE VLMs in Megatron Bridge. Compares FSDP and 3D-parallel approaches, using rounded lessons from Qwen3-VL, Qwen3-Next, and other multimodal experiments.1.9K66vss-deploy-profileUse to select, configure, deploy, verify, debug, or tear down a VSS profile (base, search, lvs, warehouse, edge). Not for standalone microservices — use the vss-deploy-* skill.1.9K67nemo-mbridge-mlm-bridge-trainingRun Megatron-LM (MLM) and Megatron Bridge training with mock or real data. Covers correlation testing, available recipes, and multi-GPU examples.1.9K68nemo-mbridge-perf-sequence-packingValidate and use packed sequences and long-context training in Megatron-Bridge, including offline LLM packing, collate-time VLM packing, Energon online packing, and CP constraints.1.9K69nemotron-policy-generatorGenerates BYO custom safety policies for NVIDIA Nemotron content-safety guardrails — Nemotron-Content-Safety-Reasoning-4B (text) and multimodal Nemotron-3-Content-Safety. Produces a Markdown policy, JSON taxonomy, and drop-in inference prompts. Maps rough words or an existing policy to V2 categories, adding custom categories or topic-following rules.1.9K70nemo-mbridge-multi-node-slurmConvert single-node scripts to multi-node Slurm sbatch jobs and debug common multi-node failures. Covers srun-native vs uv run torch.distributed approaches, container setup, NCCL timeouts, OOM sizing for MoE models, and interactive allocation.1.9K71nemo-mbridge-perf-parallelism-strategiesOperational guide for choosing and combining parallelism strategies in Megatron Bridge, including sizing rules, hardware topology mapping, and combined parallelism configuration.1.9K72tilegym-adding-cutile-kernelAdd a new cuTile GPU kernel operator to TileGym. Covers dispatch registration in ops.py, cuTile backend implementation, __init__.py exports, test creation, and benchmark in tests/benchmark. Use when adding, creating, or implementing a new cuTile operator/kernel in TileGym, or when asking how to register a new cuTile op.1.9K73nemo-mbridge-perf-expert-parallel-overlapValidate and use MoE expert-parallel communication overlap in Megatron-Bridge, including overlap_moe_expert_parallel_comm, delay_wgrad_compute, and flex dispatcher backends such as DeepEP and HybridEP.1.9K74nemo-mbridge-perf-cpu-offloadingValidate and use CPU offloading in Megatron Bridge, including layer-level activation offloading and fractional optimizer state offloading with HybridDeviceOptimizer.1.9K75nemo-mbridge-perf-moe-optimization-workflowEvidence-gated workflow for MoE performance optimization in Megatron Bridge. Covers measurement contracts, the Three Walls framework, parallel folding, profiling, matched A/B tuning, and final validation.1.9K76nemo-mbridge-perf-moe-long-contextLong-context MoE training guidance for Megatron Bridge. Covers CP sizing, selective recompute, dispatcher choices, and practical patterns from DSV3, Qwen3, and Qwen3-Next long-context experiments.1.9K77vss-deploy-detection-tracking-2dUse this skill when the user wants to deploy, run, debug, tear down, or call the REST API of the RTVI-CV 2D detection / tracking microservice. Trigger when the user says things like 'deploy rtvi-cv', 'start warehouse 2d', 'add a stream', 'check rtvi-cv health', or 'stop the perception container'. Not for VLM, embedding, or analytics — use the matching vss-* skill.1.9K78vss-deploy-video-embedding>1.9K79vss-deploy-dense-captioningUse this skill when deploying standalone RT-VLM dense captioning or calling its REST API (uploads, captions, streams, chat-completions, Kafka). Not for VSS profile deploy or video-search ingestion.1.9K80nemo-mbridge-perf-moe-comm-overlapMoE expert-parallel communication overlap in Megatron Bridge. Covers dispatch/combine overlap, flex dispatcher backends, and expert wgrad scheduling.1.9K81vss-generate-video-calibrationUse to run AutoMagicCalib on local MP4s, RTSP, or the bundled sample dataset, and to deploy vss-auto-calibration when needed. Do not use for non-AMC calibration or runtime analytics.1.9K82nemo-mbridge-perf-hierarchical-context-parallelOperational guide for enabling hierarchical context parallelism in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification.1.9K83nemo-mbridge-perf-moe-hardware-configsRepresentative, point-in-time MoE training playbooks by hardware and model family. Use them as candidate seeds, then revalidate the exact runtime, semantics, topology, and steady-state throughput.1.9K84vss-manage-video-io-storageUse to call the VIOS REST API (sensor list, timelines, clip extraction, snapshots, add/delete sensors and streams). Not for VLM inference or search.1.9K85vss-setup-video-analytics-apiUse to deploy the vss-video-analytics-api REST service standalone (config-source, data-log bind, Elasticsearch, optional Kafka). Not for full warehouse deploy.1.9K86nemo-mbridge-resiliencyResiliency features in Megatron Bridge including fault tolerance, straggler detection, in-process restart, preemption, and re-run state machine.1.9K87vss-manage-alertsUse for VSS alert workflows — real-time monitoring, Alert-Bridge subscriptions, Slack notifications, incident queries, camera onboarding. Not for non-alert analytics.1.8K88vss-query-analyticsUse this skill when reading video-analytics metrics, incidents, alerts, and sensor data via the VA-MCP server (port 9901). Not for live VLM or incident-range narrative reports.1.8K89nemo-mbridge-perf-megatron-fsdpOperational guide for enabling Megatron FSDP in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification.1.8K90nemo-mbridge-perf-moe-dispatcher-selectionChoose the right MoE token dispatcher (`alltoall`, DeepEP, or HybridEP) for the hardware, EP degree, and optimization stage. Summarizes patterns from DSV3, Qwen3, Qwen3-Next, and VLM bring-up work.1.8K91holoscan-setupGuides Holoscan SDK installation: inspects the host, assesses platform compatibility, recommends an install method, and delegates to the matching install skill.1.8K92nemo-mbridge-perf-tp-dp-comm-overlapOperational guide for enabling TP, DP, and PP communication overlap in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification.1.8K93vss-setup-behavior-analyticsUse to deploy the vss-behavior-analytics service standalone (entrypoint, config-source, optional calibration). Not for the full warehouse deploy.1.8K94nemo-rl-docsDocumentation conventions for NeMo-RL. Covers docs/index.md updates and docstring format. Do NOT use for: bug fixes, test fixes, dependency bumps, refactoring, CI/CD changes, performance tuning, or any task that does not involve writing or updating documentation.1.8K95mcore-linting-and-formattingLinting and formatting for Megatron-LM. Covers running autoformat.sh, tools (ruff, black, isort, pylint, mypy), and code style rules.1.8K96holoscan-install-containerInstall Holoscan SDK via the NGC Docker container. Use for container-based installs; not for native apt/pip/Conda installs.1.8K97holoscan-install-condaInstall Holoscan SDK v4.3+ via Conda in a CUDA 13 environment. Use for Conda installs; redirect CUDA 12 hosts to container/wheel.1.8K98holoscan-install-wheelInstall Holoscan SDK Python wheel via pip into a venv. Use for Python installs; not for native C++/apt or Conda installs.1.8K99holoscan-install-sourceBuild Holoscan SDK from source via the in-tree ./run script. Use only when published packages don't meet the user's needs.1.8K100holoscan-install-debianInstall Holoscan SDK natively on Ubuntu via apt. Use for C++ installs on Ubuntu; pair with /holoscan-install-wheel for Python.1.8K101mcore-run-on-slurmHow to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules across hardware and parallelism modes, container conventions, monitoring, and per-rank failure diagnosis.1.8K102nv-generate-mr-brainUsed for generating synthetic T1, T2, FLAIR, SWI, or MRA brain MRI volumes with NV-Generate-CTMR MR-Brain v1. Not for production training data.1.8K103mcore-create-issueInvestigate a failing GitHub Actions run or job and create a GitHub issue for the failure.1.8K104nemo-rl-brev-etiquetteBrev instance operating guidance for NeMo-RL agents working in /home/ubuntu/RL with limited workspace disk, a larger /ephemeral volume, and optional /home/ubuntu/RL/.env secrets. Use when running nemo-rl-auto-research campaigns, experiments, training jobs, model or dataset downloads, shared cache-heavy commands, log-producing runs, checkpoint generation, W&B or Hugging Face authenticated workflows, or any workflow that may create large files on Brev.1.8K105mcore-testingTest system for Megatron-LM. Covers test layout, recipe YAML structure, adding and running unit and functional tests, golden values, marker filters, and CI parity.1.8K106mcore-split-prSplit a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.1.8K107nv-generate-mr-brain-finetuneUsed for finetuning NV-Generate-CTMR MR-Brain v1 for T1, T2, FLAIR, SWI, or MRA data from a NIfTI datalist. Not for clinical or production data approval.1.8K108nemo-data-designer-pluginUse when the user wants to create a dataset, generate synthetic data, or build a data generation pipeline.1.8K109nv-generate-mrUsed for generating synthetic body MRI volumes with NV-Generate-CTMR rflow-mr. Not for paired masks or production training data.1.8K110nv-segment-ctUsed for running NV-Segment-CT VISTA3D on CT NIfTI volumes and recording label-map evidence.1.8K111nv-generate-ct-rflowUsed for generating synthetic CT volumes and masks with NV-Generate-CTMR rflow-ct. Not for production training data without review.1.8K112nv-segment-ctmrUsed for running NV-Segment-CTMR on CT or MRI NIfTI volumes and recording label-map evidence. Not for clinical interpretation.1.8K113nv-reason-cxrUsed for command-shape or live NV-Reason-CXR chest X-ray reasoning smoke tests. Not for diagnosis or clinical reporting.1.8K114nv-segment-ct-finetuneRuns standard or fixed-channel softmax finetuning of NV-Segment-CT VISTA3D on CT NIfTI image/label datasets, with optional MONAI-native MLflow tracking and checkpoint evidence. Uses softmax for predefined, mutually exclusive classes; keeps the standard workflow when point prompts or runtime-variable classes are needed. Not for clinical validation.1.8K115nv-generate-vae-finetuneUsed for finetuning the NV-Generate-CTMR MAISI VAE from CT/MRI NIfTI datalists. Not for clinical or production data approval.1.8K116nemo-evaluator-pluginEvaluate models, datasets, and agents with the NeMo Evaluator plugin. Use for metric selection, SDK checks, platform jobs, and result retrieval.1.8K117hsb-setupClone the latest NVIDIA Holoscan Sensor Bridge repo, ask which supported devkit is being used, configure the host per platform, build the correct demo container, run it, and verify HSB connectivity by pinging 192.168.0.2. Use for Holoscan Sensor Bridge setup, build, container launch, and first-connectivity bring-up.1.7K118hsb-testExecute QA test plans on Holoscan Sensor Bridge hardware. Reads a user-provided test document, filters tests by the user's setup, determines which tests can run automatically, executes them with pass/fail evaluation, and produces a structured test results report.1.7K119hsb-flashFlash the FPGA on an HSB board connected to an NVIDIA devkit. Supports HSB Lattice boards (FPGA versions 2407, 2412, 2507, 2510) and Leopard Imaging VB1940 "all-in-one" cameras (FPGA versions 2507, 2510). Uses release-specific YAML manifests and board-type-specific program commands. Lattice and VB1940 commands must never be mixed.1.7K120hsb-appDiscover and run Holoscan Sensor Bridge example applications on a connected devkit. Filters available apps by the user's platform, HSB software version, board type, and sensors. Supports timed execution, failure analysis, code-edit suggestions, and iterative re-runs.1.7K121tao-setup-nvidia-gpu-host>-1.7K122cufolioUse when a user asks to build, optimize, backtest, rebalance, or analyze a stock portfolio with Mean-CVaR, efficient frontiers, scenario generation, or NVIDIA cuOpt.1.7K123tao-finetune-huggingface-model>1.7K124cuopt-numerical-optimization-api-pythonSolve LP, MILP, QP (beta) with cuOpt Python API — linear/quadratic objectives, integer variables, scheduling, portfolio, least squares.1.7K125tilegym-cutile-pythonExpert cuTile programming assistant. Write high-performance GPU kernels using cuTile's tile-based programming model with proper validation and optimization. Supports deep agent orchestration for complex multi-kernel tasks.1.7K126tao-finetune-cosmos-reasonCosmos3-Nano video QA supervised fine-tuning with FSDP parallelism across variable-width GPU configurations. Use when training or evaluating video1.7K127tao-finetune-cosmos-embed>-1.7K128tao-launch-workflow>-1.7K129tao-port-huggingface-model>1.6K130tao-run-automlRun container-backed AutoML / hyperparameter optimization (HPO) for NVIDIA TAO networks using AutoMLRunner. Handles algorithm1.6K131tao-finetune-clipCLIP vision-language model for image-text retrieval, zero-shot classification, embedding extraction, ONNX1.6K132tao-generate-video-reasoning-annotations>-1.6K133tao-list-capabilities>-1.6K134tao-convert-dataset-formatRun `tao-daft convert` to convert NVIDIA TAO DAFT datasets between supported formats. Do not use for non-DAFT data.1.6K135tao-train-ocrnetOCRNet for scene text recognition. Recognizes text content from cropped text-region images and supports CTC1.6K136tao-train-image-classificationPyTorch-based TAO image classification. Supports a wide range of backbones (FAN, EfficientNet, ResNet, etc.)1.6K137tao-generate-image-groundingTwo-step image grounding pipeline: extracts referring expressions from (image, caption) pairs and grounds them1.6K138tao-run-on-kubernetesKubernetes execution platform — submits TAO container jobs as single-pod k8s Jobs with NVIDIA GPU scheduling.1.6K139tao-run-inference-service>1.6K140tao-analyze-changenet-rcaPerforms deep Root Cause Analysis (RCA) on NVIDIA TAO Visual ChangeNet classification experiments with1.6K141tao-run-on-local-dockerLocal or remote Docker execution for TAO SDK job containers using a Docker daemon with NVIDIA GPU runtime. Use1.6K142tao-train-optical-inspectionOptical Inspection for defect detection using Siamese networks. Compares image pairs to detect manufacturing1.6K143tao-train-action-recognitionAction recognition from video sequences. Supports RGB, optical flow, and joint (multi-stream) input types for1.6K144tao-train-grounding-dinoGrounding DINO for open-set object detection. Combines DINO-style detection with a BERT text encoder for1.6K145tao-train-pose-classificationPose classification using ST-GCN (Spatial Temporal Graph Convolutional Network). Classifies skeleton sequences1.6K146tao-analyze-gaps-visual-changenetPerforms gap analysis on NVIDIA TAO VCN Classify (Visual Component Net) experiments by invoking the pinned TAO data-services container directly via `docker run … gap_analysis vcn_aoi …` — picks the optimal decision threshold, ranks per-sample weakness, and emits a top-K weakest parquet expanded per-lighting for downstream augmentation. Use when analyzing VCN classification failures, picking SDA augmentation targets, or auditing PASS/NO_PASS boundary cases.1.6K147tao-train-nvpanoptix3dNVPanoptix3D for panoptic 3D scene reconstruction from posed RGB images. Produces 3D panoptic segmentation1.6K148tao-train-reidPerson re-identification (ReID). Learns discriminative embeddings to match the same person across different1.6K149tao-train-depth-anything-v2Monocular depth estimation using Metric Depth Anything v2 or Relative Depth Anything architectures. Predicts1.6K150tao-train-dinoDINO (DETR with Improved DeNoising Anchor Boxes) for 2D object detection. Transformer-based detector with1.6K151tao-train-mask2formerMask2Former for universal image segmentation (panoptic, instance, and semantic). Transformer-based with1.6K152tao-analyze-gaps-vlm-bcqExtract false-positive and false-negative gaps from VLM binary-classification-question (BCQ, yes/no) predictions.1.6K153tao-train-metric-learning-recognitionMetric-learning recognition (ml-recog) for fine-grained visual recognition. Learns embeddings for1.6K154tao-train-nvdinov2NVDINOv2 for self-supervised visual representation learning. Trains vision transformers via self-distillation1.6K155tao-run-on-brevBrev managed GPU instances with Docker support. Use when running TAO training, evaluation, or inference on1.6K156tao-train-rtdetrRT-DETR (Real-Time DEtection TRansformer) for 2D object detection. Designed for real-time inference with1.6K157tao-train-visual-changenetVisual ChangeNet for binary image classification and segmentation in AOI defect detection. Use when training,1.6K158tao-generate-referring-expressionsFour-step image referring-expression pipeline: turns images plus KITTI bounding-box labels into region1.6K159tao-run-automl-deft-pipeline>1.6K160tao-run-deft-aoi>1.6K161tao-train-centerposeCenterPose for keypoint / pose estimation. Detects object centers and regresses keypoint locations for 6-DoF1.6K162tao-train-segformerSegFormer for semantic segmentation. Lightweight transformer-based architecture with hierarchical feature1.6K163tao-train-single-stepStandard single-step train/eval/export workflow for any TAO model. Use when training a TAO model on a dataset1.6K164tao-train-pointpillarsPointPillars for 3D object detection from LiDAR point clouds. Encodes point clouds into a pseudo-image via a1.6K165tao-train-mask-auto-encoderMasked Auto-Encoder (MAE) for self-supervised pretraining and fine-tuning. Masks random patches and reconstructs1.6K166tao-train-mask-auto-labelMAL (Mask Auto-Label) for weakly-supervised segmentation. Produces segmentation masks from minimal annotations1.6K167tao-train-ocdnetOCDNet for scene text detection. Detects arbitrary-oriented text regions in natural images using a1.6K168tao-train-bevfusionBEVFusion for multi-sensor 3D object detection. Fuses LiDAR point clouds and camera images in bird's-eye-view1.6K169tao-validate-dataset-formatRun `tao-daft validate` to check NVIDIA TAO DAFT datasets for structure, schema, and cross-reference errors. Do1.6K170tao-train-mask-grounding-dinoMask Grounding DINO for grounded instance segmentation. Extends Grounding DINO with a mask-prediction head for1.6K171tao-train-oneformerOneFormer for universal image segmentation. Unifies panoptic, instance, and semantic segmentation with a1.6K172tilegym-monkey-patch-kernels-to-transformersIntegrate TileGym kernels into Hugging Face `transformers` models by replacing the library's submodule(s) and certain class(es)' implementations, and patching certain class(es)' init/forward/load weight methods prior to instantiating models. Used when the user requires integrating TileGym kernels into `transformers` models.1.6K173tao-train-deformable-detrDeformable DETR for 2D object detection. Uses deformable attention for efficient multi-scale feature processing,1.6K174tao-mine-aoi-imagesRuns the DEFT embed-then-mine workflow for VCN AOI iterations — embeds the gap-analysis target parquet, embeds a source pool, and mines nearest-neighbour source images for downstream augmentation. Use as the immediate next step after `tao-route-visual-changenet-samples` when expanding a real-image augmentation queue from the mining subset.1.6K175tao-run-on-slurmRemote SLURM GPU cluster execution over SSH with sbatch/srun, Pyxis/Enroot containers, and Lustre-backed1.6K176tao-train-sparse4dSparse4D for multi-camera temporal 3D object detection and tracking. Uses sparse queries with deformable1.6K177tao-train-foundation-stereoStereo depth estimation using FoundationStereo. Predicts disparity maps from stereo image pairs for 3D1.6K178tao-train-fast-foundation-stereoReal-time stereo depth estimation using FastFoundationStereo (FFS), the distilled bp2 commercial variant of1.6K179tao-route-visual-changenet-samplesRoutes the weakest VCN samples (output of `tao-analyze-gaps-visual-changenet`) into per-augmentation-module1.6K180tilegym-improve-cutile-kernel-perfIteratively optimize cuTile kernel performance through systematic profiling, bottleneck analysis, IR comparison, and targeted tuning. Covers tile sizes, occupancy, autotune configs, TMA, latency hints, persistent scheduling, num_ctas, flush_to_zero, and IR-level debugging. Use when asked to "optimize cutile kernel", "improve kernel perf", "tune cutile performance", "make kernel faster", or iteratively benchmark and refine a cuTile GPU kernel in the TileGym project.1.6K181tao-run-platformTAO Execution SDK for submitting and monitoring GPU training jobs on supported platforms (Brev, SLURM,1.6K182tilegym-cutile-autotuningUse when adding, modifying, optimizing, or debugging CuTile autotuning code. Trigger signals: `exhaustive_search` / `replace_hints` / `hints_fn` / `cuda.tile.tune` in code, `autotune` in filenames, or correctness/performance issues in autotuned CuTile kernels. Covers: tune-once/cache/launch pattern, per-architecture configs (sm80–sm120), parameter space design (tile sizes, occupancy, num_ctas), and 7 common pitfalls with solutions.1.5K183tilegym-converting-cutile-to-tritonConverts 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.1.5K184tilegym-converting-cutile-to-juliaConverts cuTile Python GPU kernels (@ct.kernel) to cuTile.jl Julia equivalents. Handles kernel syntax translation, 0-indexed to 1-indexed conversion, broadcasting differences, memory layout (row-major to column-major), type system mapping, and launch API differences. Use when converting, porting, or translating cuTile Python kernels to Julia cuTile.jl, or debugging/optimizing existing Julia cuTile translations.1.5K185cuopt-user-rulesBase rules for end users calling NVIDIA cuOpt (routing/LP/MILP/QP/install/server). Not for cuOpt internals — use cuopt-developer for those.1.4K186cuopt-server-commoncuOpt REST server — what it does and how requests flow. Domain concepts; no deploy or client code.1.3K187cuopt-skill-evolutionAfter solving a non-trivial problem, detect generalizable learnings and propose skill updates. Always active — applies to every interaction.1.3K188cuopt-numerical-optimization-api-cliLP, MILP, and QP (beta) with cuOpt — CLI only (MPS/LP/QPS files, cuopt_cli). Use when the user is solving LP, MILP, or QP from an MPS or LP file via command line.1.3K189nemo-retriever-mcpUse when a task needs to search or add documents through NeMo Retriever MCP.9190cuopt-multi-objective-explorationcuopt-multi-objective-exploration is published by nvidia/skills. Review its source instructions before using it in an agent workflow.8191nemoclaw-user-guideGuides human users' AI agents to the NemoClaw docs MCP server and canonical Fern documentation in Markdown form. Use when users ask how to install, configure, operate, troubleshoot, secure, or learn NemoClaw with an AI coding assistant. Trigger keywords - nemoclaw docs, use nemoclaw with ai agent, nemoclaw mcp docs, nemoclaw install help, nemoclaw quickstart, nemoclaw markdown docs, llms.txt, agent skills.8192nemotron-voice-agent-builderCreate, refine, or fix NVIDIA voice agents (Cascaded or Omni) with Pipecat or LiveKit. Use when building, scaffolding, or iterating on a real-time voice-agent pipeline, including speech (ASR/TTS) customization and cloud or local deployment. Not for offline/batch speech-to-text, text-only chat or RAG, or generic Docker or CUDA work unrelated to a voice agent.8193warp-eval>8194warp-debug-gradients>-7195nvidia-skill-findernvidia-skill-finder is published by nvidia/skills. Review its source instructions before using it in an agent workflow.6196warp-compile-time-optimizer>-6197cuopt-numerical-optimization-apiLP, MILP, and QP (beta) with cuOpt — Python, C, and CLI. Use when the user is solving LP, MILP, or QP with any cuOpt interface.5198isaac-mission-control-showcaseRun and validate an end-to-end Mission Control showcase with a locally installed Isaac Sim launched in its GUI window, driven through the isaac-sim-remote Python server, with Nova Carter SIL. Use for demos, showcase replays, Mission Control driving a simulated robot, or diagnosing the integrated small-warehouse scenario. Detect existing Isaac Sim installations without modifying them, automatically select a usable runtime without prompting whenever compatibility can be confirmed, and delegate requested installation or version changes to isaac-sim-installation. Defaults to a canonical Isaac 6.1 warehouse and deterministic circular route when the user does not specify another scenario.1199tao-run-on-virtualenvRun a Python training/eval script directly in an existing local virtualenv — no docker, no container. Implements the four-verb consumer contract (submit/status/logs/cancel) over a vendored process-lifecycle runner with durable on-disk state, PID-reuse-safe identity, and process-group cleanup. Use for docker-free local execution, plain-Python model scripts, fast HPO/AutoML trial smokes, or hosts where containers are unavailable. Trigger phrases include "run in my venv", "no docker", "virtualenv execution", "local python training", "run this training script directly".1