NVIDIA
- 1.2k skills
- 0 followers
- 23k repo stars
- 561 verified
- 11 hours ago last updated
- ▌ Deploy Runai Kai · nvidiaUse when adding or removing the optional KAI Scheduler or Run:ai platform on a DeepOps Kubernetes GPU cluster, enabling fractional-GPU scheduling, or diagnosing KAI/Run:ai install or scheduling failures.
- ▌ Manage Model Cache · nvidiaUse when preparing a shared Hugging Face model cache for Slurm or Kubernetes, pinning downloaded model revisions, validating offline model files, or diagnosing missing snapshots and cache permissions.
- ▌ Provision With Maas · nvidiaProvision or reinstall bare-metal servers and test VMs through Canonical MAAS, map deployed machines into DeepOps Ansible inventory with MAAS tags, validate access, or release them safely. Use when operating DeepOps with a MAAS-owned machine lifecycle.
- ▌ Deploy Slurm Cluster · nvidiaDeploy a Slurm GPU cluster with DeepOps and prove it works. Use when asked to deploy, install, or rebuild Slurm on one or more GPU servers with this repository.
- ▌ Validate Gpu Cluster · nvidiaCheck whether a DeepOps-deployed Slurm or Kubernetes GPU cluster is healthy and report a machine-readable verdict. Use for health checks, post-deploy verification, "is the cluster working?" questions, and after any node or driver change.
- ▌ Deploy K8S Gpu Cluster · nvidiaDeploy a Kubernetes GPU cluster with DeepOps (Kubespray + GPU Operator) and prove it schedules GPU pods. Use when asked to deploy or rebuild Kubernetes on GPU servers with this repository.
- ▌ Diagnose Driver Install · nvidiaDiagnose NVIDIA driver installation failures on DeepOps-managed nodes — nvidia-smi errors, "No devices were found", DKMS build failures, or GPU pods crash-looping. Use before reinstalling anything.
- ▌ Modelopt Compare Results · nvidia bundleEstablish baseline-vs-candidate evaluation plans, delegate missing evaluations, compare validated results, and decide quantization feasibility. Use when the user asks to compare baseline vs quantized runs, explain an accuracy drop/regression, verify whether a quantized checkpoint is acceptable, or compare NEL/MLflow evaluation outputs. Do NOT use for generic single-model evaluation without comparison intent (use modelopt-evaluation), live NEL status/debugging (use modelopt-launching-evals), or generic MLflow browsing without a comparison goal (use modelopt-accessing-mlflow).
- ▌ Modelopt Launching Evals · nvidia bundleRun, monitor, analyze, and debug LLM evaluations via nemo-evaluator-launcher. Covers running evaluations, checking status and live progress, debugging failed runs, exporting artifacts and logs, and analyzing results. ALWAYS triggers on mentions of running evaluations, checking progress, debugging failed evals, analyzing or analysing runs or results, run directories or artifact paths on clusters, Slurm job issues, invocation IDs, or inspecting logs (client logs, server logs, SSH to cluster, tail logs, grep logs). Do NOT use for creating or modifying evaluation configs.
- ▌ Modelopt Quant Recipe Search · nvidia bundleUse when the user asks to find, search for, or optimize the best quantization recipe for a model, including direct requests like "find the best quantization recipe and generate a PTQ checkpoint." Guides the multi-candidate loop: choose compute-vs-memory success metrics, select ModelOpt recipe baselines, design AutoQuant/manual recipe deltas, interpret sensitivity, and decide next candidates. Do NOT use for a single known PTQ recipe run (use modelopt-ptq), serving (use modelopt-deployment), creating/running evals (use modelopt-evaluation or modelopt-launching-evals), monitoring jobs (use modelopt-monitor), MLflow browsing (use modelopt-accessing-mlflow), or comparing completed baseline-vs-candidate scores only (use modelopt-compare-results).
- ▌ Modelopt Release Cherry Pick · nvidiaAudit merged bug-fix PRs and release NVBugs for missing cherry-pick labels, then cherry-pick labeled PRs into a release branch and open a PR. Use when asked to "cherry-pick PRs for release/X.Y.Z", "pick PRs to release branch", "verify cherry-pick labels", or "cherry-pick labeled PRs".
- ▌ Modelopt Speculative Decoding · nvidia bundleTrain, debug, and validate a speculative decoding draft model (EAGLE3, DFlash, DSpark, Domino) through the ModelOpt launcher pipeline. Use when the user wants to add a new model to a draft-training pipeline, asks why a pipeline run failed, wants experiment logs reviewed, or wants to check whether a run's acceptance rate meets threshold. Triggers on "EAGLE3", "DFlash", "DSpark", "Domino", "draft model", "acceptance rate", "speculative decoding pipeline". Do NOT use for quantizing a model (use modelopt-ptq) or serving a checkpoint (use modelopt-deployment).
- ▌ Modelopt Benchmark Model Kernels · nvidia bundleInspect Hugging Face decoder layers on meta tensors and plan or run per-rank BF16, FP8, and NVFP4 GEMM or fused-MoE microbenchmarks with the bundled scripts and a local FlashInfer checkout. Use when choosing a model, GPU, TP, EP, or M/token-concurrency sweep; deriving common fused QKV and gate/up shapes without loading checkpoint weights; or using FlashInfer benchmark utilities. Do not use for end-to-end server throughput or request latency.
- ▌ Modelopt Qad · nvidiaRun explicitly requested ModelOpt Quantization-Aware Distillation (QAD) through Megatron Bridge (recommended) or Hugging Face Transformers to recover a measured BF16-to-PTQ accuracy gap. Use only when the user explicitly asks for QAD, including its topology, data preparation, launch, resume, checkpoint export, or recovery decisions.
- ▌ Modelopt Debug · nvidiaRun commands inside a remote Docker container via the file-based command relay (tools/debugger). Use when the user says "run in Docker", "run on GPU", "debug remotely", "run test in container", "check nvidia-smi", "run pytest in Docker", or needs to execute any command inside a Docker container that shares the repo filesystem. Requires the user to have started server.sh inside the container first.
- ▌ Modelopt Monitor · nvidiaMonitor submitted jobs (PTQ, evaluation, deployment) on SLURM clusters. Use when the user asks "check job status", "is my job done", "monitor my evaluation", "what's the status of the PTQ", "check on job" followed by a SLURM job ID, or after any skill submits a long-running job. Also triggers on "nel status", "squeue", or any request to check progress of a previously submitted job.
- ▌ Modelopt Deployment · nvidia bundleServe a quantized or unquantized LLM checkpoint as an OpenAI-compatible API endpoint using vLLM, SGLang, or TRT-LLM. Use when user says "deploy model", "serve model", "start vLLM server", "launch SGLang", "TRT-LLM deploy", "AutoDeploy", "benchmark throughput", "serve checkpoint", or needs an inference endpoint from a HuggingFace or ModelOpt-quantized checkpoint. Do NOT use for quantizing models (use modelopt-ptq) or evaluating accuracy (use modelopt-evaluation).
- ▌ Modelopt Evaluation · nvidia bundleEvaluates accuracy of quantized or unquantized LLMs using NeMo Evaluator Launcher (NEL). Triggers on "evaluate model", "benchmark accuracy", "run MMLU", "evaluate quantized model", "run nel". Handles deployment, config generation, and evaluation execution. Always load when validating completed runs or reporting evaluation scores, including standalone modelopt-launching-evals analysis. Not for quantizing models (use modelopt-ptq), standalone serving (use modelopt-deployment), or comparing completed baseline-vs-quantized results (use modelopt-compare-results).
- ▌ Modelopt Run Agentx · nvidiaRun the AgentX agentic serving benchmark with the SemiAnalysis harness. Use for AgentX setup, dataset downloads, concurrency sweeps, or serving latency and throughput comparisons.
- ▌ Modelopt Day0 Release · nvidia bundleDeterministic end-to-end driver for day-0 quantized-checkpoint releases — chains PTQ → evaluation → comparison with enforced gates between stages (the evaluation stage deploys the checkpoint itself), and returns a publish decision (ACCEPT / REGRESSION / INDETERMINATE / ANOMALOUS / INFEASIBLE). Use when the user asks to "release a model at day-0", "quantize and validate model X is within N% of baseline and tell me if it's publishable", or "run the full day-0 workflow". Do NOT use for single-stage requests — quantizing only (use modelopt-ptq), serving only (use modelopt-deployment), evaluating only (use modelopt-evaluation), or comparing two existing runs (use modelopt-compare-results).
- ▌ Modelopt Ptq · nvidia bundleUse when the user asks to "quantize a model", "run PTQ", "post-training quantization", "NVFP4 quantization", "FP8 quantization", "INT8 quantization", "INT4 AWQ", "quantize LLM", "quantize MoE", "quantize VLM", or needs to produce a quantized HuggingFace checkpoint from a pretrained model using ModelOpt. Do NOT use for multi-candidate recipe exploration or optimization (use modelopt-quant-recipe-search).
- ▌ Cuequivariance Torch 2 · nvidia bundleUse when building or debugging cuEquivariance PyTorch tensor products or equivariant GNNs, especially CPU references, layouts, indexing, and gradients.
- ▌ Cuequivariance 2 · nvidia bundleUse when defining cuEquivariance groups and irreps, building equivariant tensor products, or validating segmented polynomials with NumPy.
- ▌ Cuequivariance Jax 2 · nvidia bundleUse when writing or debugging cuEquivariance JAX polynomials, irrep dictionaries, Flax NNX layers, or CPU/CUDA execution and layouts.
- ▌ Nvflare Convert Huggingface 2 · nvidia bundleConvert existing Hugging Face Transformers Trainer or TRL SFTTrainer training code into an NVFLARE federated job using flare.patch(trainer), local validation, and job export; use when the user names Hugging Face or preliminary source inspection identifies one Hugging Face owner, and not for manual PyTorch loops, Lightning, inference-only pipelines, deployment, or experiment workflows.
- ▌ Skillevaluator · nvidia bundleRun and interpret SkillEvaluator for agent skills. Use for any skill benchmark or eval - validating or scoring a SKILL.md, writing evals.json test cases or evals config, explaining benchmark reports, verdicts and exit codes, or gating skills in CI.
- ▌ Torch Optimize · nvidia bundleThree-role optimization loop for a PyTorch training workload: the Main Loop Agent owns the end-to-end run, authoritative profiling, and hotspot selection; a disposable Optimizer subagent reworks one hotspot per round in an isolated context; a Judge subagent renders an evidence-based verdict driving iterate/pass. All cross-role state lives in files.
- ▌ Torch Perf Analysis · nvidia bundleAnalyze an existing nsys trace of a PyTorch workload (training or inference, NVTX-instrumented) and produce an end-to-end performance diagnosis. Uses the bundled `veloq` CLI for evidence extraction.
- ▌ Udf Gen Test · nvidia bundleAssists with generating a unit test for an Apache Spark UDF. This is step 1 of 3 in the UDF conversion workflow (udf-gen-test -> udf-convert-to-* -> udf-benchmark). Use this skill when you have a CPU UDF and need to create a unit test for the UDF before converting it into a GPU-compatible implementation.
- ▌ Udf Optimize Cudf · nvidia bundleIteratively optimizes a cuDF RapidsUDF implementation for GPU performance. Use after testing and benchmarking with udf-benchmark. Runs a loop of profiling, optimizing, testing, and benchmarking until performance converges or the iteration budget is exhausted.
- ▌ Udf Convert To SQL · nvidia bundleAssists with converting an Apache Spark UDF to a functionally equivalent Spark SQL expression. This is step 2 of 3 in the UDF conversion workflow (udf-gen-test -> udf-convert-to-sql -> udf-benchmark). Use this skill when you have a CPU UDF with a unit test and need to convert it to SQL for GPU acceleration.
- ▌ Udf Convert To Cuda · nvidia bundleAssists with converting a non-aggregating Apache Spark UDF to a native CUDA RapidsUDF using JNI and libcudf. This is step 2 of 3 in the UDF conversion workflow (udf-gen-test -> udf-convert-to-cuda -> udf-benchmark). Use this skill when you have a CPU UDF with a unit test and need to convert it to a native CUDA implementation. Prefer udf-convert-to-cudf unless a CUDA implementation is necessary for performance or correctness, or if requested by the user.
- ▌ Udf Convert To Cudf · nvidia bundleAssists with converting an Apache Spark UDF to a GPU-accelerated RapidsUDF using cuDF Java APIs. This is step 2 of 3 in the UDF conversion workflow (udf-gen-test -> udf-convert-to-cudf -> udf-benchmark). Use this skill when you have a CPU UDF with a unit test and need to convert it to a RapidsUDF.
- ▌ Udf Judge Conversion · nvidiaReviews generated UDF tests and GPU/SQL implementations for robustness, anti-cheating, and GPU execution integrity. Use when the user requests a judge/review-agent pass, or when manually reviewing a completed conversion.
- ▌ Test Writing · nvidiaGuide for writing tests in torch-harmonics. Use this skill whenever the user asks to add tests, check or extend test coverage, write a new test class, test a new layer or kernel, or add distributed tests. Also use when the user asks about tolerance values, how to compare tensors, how to structure distributed test infrastructure, or how to test gradients. Covers serial tests in test_convolution.py and distributed tests in test_distributed_*.py.
- ▌ Disco Kernel Dev · nvidiaExpert guide for adding, optimizing, or debugging DISCO convolution kernels in torch-harmonics. Use this skill whenever the user mentions: adding a new kernel variant or GPU arch path, wiring up a kernel through the Python dispatch layer, fixing bf16/fp16/autocast issues in DISCO, profiling or benchmarking the CSR or kpacked forward/backward, propagating kernel changes to the distributed DISCO path, or writing tests that cover all dispatch branches. Also use for debugging correctness issues in the DISCO fwd/bwd path or distributed a2a collectives.
- ▌ Ncore Data Conversion · nvidia bundleUse when getting sensor data into an NCore V4 store: running a built-in converter (PAI, Waymo, COLMAP/ScanNet++, KITTI, nuScenes, Argoverse 2), adapting the nearest built-in converter for a dataset NCore does not ship one for, or checking a converted store before something downstream reads it. Covers V4 conventions for poses, camera and lidar models, cuboids and timestamps. Do NOT use to train reconstructions or to extract per-object 3D assets.
- ▌ Migrate Mongodb To Percona · nvidiaExecute the NVSentinel MongoDB backend migration from Bitnami to the Percona Operator after readiness is confirmed: data dump (default preserve path), removal of the installation, cleanup of surviving objects, values preparation, and the Percona deployment. Destructive: on the opt-out clean path all health event data is wiped. Use only after check-mongodb-migration-readiness reports READY and the operator confirmed the decisions.
- ▌ Verify Mongodb Percona Migration · nvidiaPost-migration verification for the NVSentinel Percona MongoDB backend: waits on the five install gates, performs the optional data restore and consumer restarts, and walks the operator through the aftermath expectations. Use after migrate-mongodb-to-percona completes, or on its own to health-check an existing Percona-backed installation.
- ▌ Check Mongodb Migration Readiness · nvidiaReadiness check for migrating an NVSentinel installation from the Bitnami MongoDB backend to the Percona Operator backend. Runs the preflight script, interprets the verdict table, and captures the operator decisions (data handling, quarantined nodes, GitOps) that the migration skill requires. Use this first, before any migration step.
- ▌ Trtllm Modeling V2 Entry · nvidiaLoad before implementing, modifying or certifying a TensorRT-LLM ModelingV2 catalog entry (OpWrapper) or its GPU test, including verification at new shapes/dtypes/architectures and target-local variants. Not needed for catalog discovery, assembling a target from unchanged entries, running model-level gates, or tuning serving settings.
- ▌ Trtllm Modeling V2 Features · nvidia bundleLoad when working on, testing or debugging a TensorRT-LLM engine feature around a ModelingV2 target -- KV-cache block (prefix) reuse, disaggregated serving, chunked prefill, KV-cache dtype (fp8/nvfp4 and their scales), CUDA graphs and the overlap scheduler, multimodal inputs, speculative decoding, or attention-DP/TP/EP where it changes one of those -- including a check that fails on a feature field, or a number that moves only with a feature on. Says which engine stage owns the feature and what the target must honour, where its code and tests are in the tree, and how to prove it from a CPU check up to the flow's accuracy, disagg or acceptance gate. Not needed to write or certify one catalog entry (trtllm-modeling-v2-entry) or to debug core accuracy with every feature at its default.
- ▌ Nv Segment Ct Finetune 2 · nvidia bundleRuns 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.
- ▌ Cuopt Server API Python 2 · nvidia bundlecuOpt REST server — start server, endpoints, Python/curl client examples. Use when the user is deploying or calling the REST API.
- ▌ Holoscan Install Debian 2 · nvidia bundleInstall Holoscan SDK 5.x native C++ Debian packages on Ubuntu 24.04, choosing the full SDK or matching modular packages.
- ▌ Holoscan Install Source 2 · nvidia bundleBuild and verify Holoscan SDK 5.x from an explicit source ref with the repository run wrapper; use for Early Access or SDK development.
- ▌ Holoscan Install Container 2 · nvidia bundleInstall and verify the Holoscan SDK 5.x CUDA 13 development container after checking release availability and host compatibility.
- ▌ Nv Generate Mr Brain Finetune 2 · nvidia bundleUsed 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.
- ▌ Cuopt Multi Objective Exploration 2 · nvidia bundleTrace, complete, and interpret the Pareto frontier across competing objectives using repeated single-objective cuOpt solves (weighted-sum and ε-constraint).
- ▌ Nemo Automodel Recipe Development 2 · nvidia bundleCreate and modify NeMo AutoModel training and evaluation recipes, including YAML structure, builders, and execution flow.
- ▌ Cudaq Guide 2 · nvidia bundleUse for CUDA-Q setup, simulation targets, QPU access, and @cudaq.kernel authoring guidance.
- ▌ Holoscan Setup 2 · nvidia bundleAssess a host and choose a Holoscan SDK 5.x installation method using release availability, CUDA 13 compatibility, and component needs.
- ▌ Nemo Retriever 2 · nvidia bundleUse when searching, extracting, ingesting, or querying a document collection with the NeMo Retriever 26.8.1 CLI, including local LanceDB indexes and deployed Retriever services. Use for PDFs, images, Office files, HTML, text, audio, and video; not for editing documents or web search.
- ▌ Nemotron Speech 2 · nvidia bundleRoutes NVIDIA Nemotron Speech (Formerly Riva) NIM tasks — deploys, runs, and tests ASR, TTS, and NMT NIMs on build.nvidia.com or self-hosted.
- ▌ Nv Segment Ctmr 2 · nvidia bundleUsed for running NV-Segment-CTMR on CT or MRI NIfTI volumes and recording label-map evidence. Not for clinical interpretation.
- ▌ Nv Generate Mr Brain 2 · nvidia bundleUsed for generating synthetic T1, T2, FLAIR, SWI, or MRA brain MRI volumes with NV-Generate-CTMR MR-Brain v1. Not for production training data.
- ▌ Dicom Metadata Extract 2 · nvidia bundleUsed for extracting selected metadata from one DICOM file and flagging standard-tag PHI presence. Not for anonymization or clinical use.
- ▌ Dicom Series Preflight 2 · nvidia bundleUsed for header-only preflight of one DICOM series folder before conversion or inference. Not for de-identification or clinical clearance.
- ▌ Dicom Series To Volume 2 · nvidia bundleUsed for converting one CT DICOM series folder to a HU NIfTI volume with affine evidence. Not for multi-frame DICOM or clinical use.
- ▌ Holoscan Install Conda 2 · nvidia bundleHandle Holoscan SDK 5.x Conda installation requests: explain 5.0 package unavailability, check later releases, and select a supported installation method.
- ▌ Holoscan Install Wheel 2 · nvidia bundleInstall Holoscan SDK 5.x CUDA 13 Python wheels in an isolated environment, choosing the full stack or version-matched components.
- ▌ Doca Bare Metal Deployment · nvidia bundleRunning a DOCA-linked binary directly on the hardware, with no container or kubelet: on host x86 with a BlueField NIC over PCIe, or bare-metal on BlueField Arm. Use this skill for launching and supervising it (direct, tmux, systemd), and when it "exits 1 right after launch", "systemd keeps restarting it", reports "no matching device on the BF", or the BlueField itself is unhealthy: "bfb-install exited 0 but DPU is dead", "ping 192.168.100.2 works but ssh fails", "host PFs aren't showing netdevs". Refuse service containers under kubelet (doca-container-deployment). Covers PCI/NUMA/CPU/IRQ binding, co-tenant isolation (cgroup-v2/netns/numactl), a seven-layer error taxonomy, platform lifecycle (BFB install, RShim/TMFIFO, host PF rebind, post-BFB recovery) and a six-state BlueField lifecycle classifier. Destructive firmware burn / mlxconfig set requires explicit confirmation via doca-hardware-safety; library APIs, env prep, and build use other skills.
- ▌ I4h Workflow Dataset Mimic · nvidia bundleExpand workflow HDF5 demonstrations with action jitter, optionally scoped to node segments. Use for synthetic variants; do not use to collect data, alter state directly, or generate new images.
- ▌ I4h Workflow Dataset Replay · nvidia bundleReplay a workflow HDF5 episode through its original Scene. Use for visual trajectory and recording verification; do not use for policy evaluation or LeRobot data.
- ▌ I4h Workflow Dataset Teleop · nvidia bundleRecord demonstrations through a workflow's teleop Task into workflow HDF5. Use for keyboard, leader, VR, or bus input; do not use for policy evaluation or autonomous rule-based Tasks.
- ▌ I4h Workflow Dataset Convert · nvidia bundleConvert workflow HDF5 recordings to LeRobot datasets for training or browser inspection. Use for conversion; do not use for replay, augmentation, or raw-data repair.
- ▌ I4h Workflow Dataset Annotate · nvidia bundleGrade or filter workflow HDF5 episodes with an OpenAI-compatible vision model. Use for visual success labels; do not use for replay, policy evaluation, or recordings without frames.
- ▌ Medtech Model Evidence Export · nvidia bundleExports sanitized metadata, parameters, reproducibility details, quality metrics, and optional review artifacts from Medical AI inference runs or evidence packs to MLflow. Use after inference, including NV-Generate runs; not for live training tracking, model registration, or clinical use.
- ▌ Doca Structured Tools Contract · nvidia bundleNot for writing library code (name that library skill). Use this skill whenever another DOCA skill says "prefer the structured tool per doca-structured-tools-contract", or when the user wants a one-shot answer consolidating what multiple manual commands would produce — DOCA env / version / devices / capabilities / validate / host vs DPU state. Trigger even without the phrases "structured tool" or "doca-env --json" — typical phrasings include "is there one command that tells me everything about my DOCA install", "what version is X capability available since", "every PF/VF/SF visible on this BlueField with PCIe address", or "will this pipe pass validate before commit". Refuse and route elsewhere for general DOCA orientation, library API how-to, or install-from-scratch guidance — those belong to the per-library skill, doca-public-knowledge-map, or doca-setup.
- ▌ Tao Generate Referring Expressions · nvidia bundleFour-step image referring-expression pipeline: turns images plus KITTI bounding-box labels into region descriptions, scene captions, grounded referring expressions, and (optionally) verified expressions via VLM distillation. Use when the user wants to generate referring-expression annotations from images with KITTI labels, build region descriptions, produce grouped grounding phrases tied to bboxes, run a double-check verification pass on grounding expressions, auto-label traffic / scene images for referring datasets, or run the image_referring_expression pipeline. Triggers include 'referring expression', 'region description', 'KITTI labels', 'spatial relationship annotation', 'auto-label image referring expression', 'image_referring_expression'.
- ▌ Tao Route Visual Changenet Samples · nvidia bundleRoutes the weakest VCN samples (output of `tao-analyze-gaps-visual-changenet`) into per-augmentation-module subsets based on each module's label eligibility. Use when the user asks to "route VCN gap samples", "split AOI gaps for k-NN mining and AnomalyGen", or prepare the immediate next step after DEFT gap analysis in a VCN AOI SDA iteration.
- ▌ Tilegym Converting Cutile To Julia · nvidia bundleConverts 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.
- ▌ Tilegym Converting Cutile To Triton · nvidia bundleConverts 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.
- ▌ Tao Train Metric Learning Recognition · nvidia bundleMetric-learning recognition (ml-recog) for fine-grained visual recognition. Learns embeddings for retrieval-based matching (e.g., retail product recognition) using triplet / contrastive losses. Use when training, evaluating, exporting, or running inference for a TAO metric-learning recognition model. Trigger phrases include "train metric learning", "ml-recog", "retrieval embeddings", "triplet loss recognition", "fine-grained matching".
- ▌ Nemo Fabric Integrate · nvidia bundleUse this skill when integrating NVIDIA NeMo Fabric into a consumer application, service, evaluation harness, or platform through the typed Python SDK — translating the consumer's own application, job, or deployment config into an in-memory FabricConfig, choosing the single-invocation convenience API or an explicitly started runtime, validating with plan and doctor, and consuming normalized results, artifacts, and telemetry.
- ▌ Nemotron Asr Finetune · nvidia bundleOrchestration skill for NVIDIA Nemotron Speech (Riva) / NeMo ASR domain and language adaptation. Given a goal like "improve/fine-tune ASR for my domain or language", it scopes the task, picks the cheapest sufficient path (word boosting → n-gram LM → fine-tuning), delegates each stage to the right sub-skill (data generation, training, evaluation, deployment), and answers cost/time/data questions along the way.
- ▌ Doca Programming Guide · nvidia bundleUse this skill when the user is writing their first DOCA app or asking a library-agnostic programming question — picking a shipped sample to copy and modify, wiring pkg-config + meson (or FFI from Rust / Go / Python against the public C ABI), walking the Core lifecycle (create → configure → start → connect → use → stop → destroy), or decoding a DOCA_ERROR_*. Trigger on implicit phrasing: "write my first DOCA program", "meson line for doca_rdma_*", "DOCA_ERROR_BAD_STATE on first call", "call DOCA from Rust", "built clean but nothing on the wire". Refuse install / hugepages / pkg-config (doca-setup), docs or version lookup (doca-public-knowledge-map), and library-internal API work (matching library skill).
- ▌ Jetson Video Benchmark · nvidia bundleUse when measuring Jetson Video Codec SDK or PyNvVideoCodec encode/decode throughput, comparing presets or surfaces, testing codec-worker capacity with authenticated samples and user media, or producing a clearly labeled documentation-derived planning estimate when representative media is absent. Also use for a video request limited to PSNR or SSIM, to apply the terminal scope response.
- ▌ Nemo Relay Get Started · nvidia bundleUse this skill when first-time NeMo Relay users want help selecting and following an appropriate quick-start guide or verifying their first working integration. Do not use it for production rollout or broad integration development.
- ▌ Portfolio Optimization · nvidia bundleUse when a user asks to build, optimize, backtest, rebalance, or analyze a stock portfolio with Mean-CVaR, Mean-Variance/SOCP variance caps, efficient frontiers, scenario generation, or NVIDIA cuOpt.
- ▌ Doca Comm Channel Admin · nvidia bundleUse this skill to enumerate host↔DPU DOCA comch (formerly Comm Channel) servers and connections via the shipped doca_comm_channel_admin binary — listing comch-capable devices and decoding the per-device server / connection table (server name, PID, in-use / max, PCIe address). The shipped binary is a SINGLE-SHOT SCAN-AND-PRINT tool with no registered arguments — NO list / inspect / drain / restart subcommands; one inventory pass over every comch-capable doca_dev on this side. Channel reset / drain / restart go to doca-comch (program side), doca-setup / doca-hardware-safety (driver reload), or BFB / RShim — NOT to this binary. Trigger on phrasings like "list comch servers", "which channels are active on this BlueField", or "verify admin tool sees same channel as program." Refuse and route elsewhere for the comch programming API, library install, protocol design, channel reset, or general orientation.
- ▌ Doca Sha Offload Engine · nvidia bundleUse this skill when wiring the DOCA SHA Offload Engine (an OpenSSL ENGINE) into an existing OpenSSL pipeline to offload one-shot SHA-1, SHA-256, or SHA-512 (EVP_Digest) onto DOCA SHA hardware without rewriting against doca-sha. Covers engine load mechanics (`openssl engine dynamic`, `set_pci_addr` ctrl, `-engine_impl`), the SHA-224 negative test that proves offload engaged, the message-size window where offload beats CPU SHA, and engine-vs-library selection. Trigger even when the user does not say "DOCA SHA Offload Engine" or "OpenSSL ENGINE" — typical implicit phrasings: "speed up openssl SHA on BlueField", "offload SHA without code changes", "is openssl using the accelerator or falling back to software", "prove DOCA SHA actually ran", "openssl dgst hashed but I'm not sure it was offloaded". Refuse and route elsewhere for new SHA pipelines (use doca-sha), MD5 / SHA-3 / SHA-224 / HMAC-SHA offload, incremental hashing via chained `EVP_DigestUpdate`, or OpenSSL PROVIDER authoring.
- ▌ Doca Telemetry Exporter · nvidia bundleUse this skill when the user is doing hands-on DOCA Telemetry Exporter programming on a host where DOCA is installed — defining a doca_telemetry_exporter_schema and event types, creating sources, picking a publish surface (typed events / opaque events / the metrics counter-gauge-histogram API / OTLP logs / NetFlow), walking the schema-then-source lifecycle, or debugging DOCA_ERROR_* failures from the exporter API. Trigger even when the user does not explicitly mention "DOCA Telemetry Exporter" or "doca_telemetry_exporter_*" — typical implicit phrasings include "publishing counters from my DOCA app", "BAD_STATE when I report an event", "consumer/DTS sees nothing but my report succeeded", "how do I export NetFlow/IPFIX records", or "should I link the exporter or the telemetry service". Refuse and route elsewhere for the receiving DOCA Telemetry Service (DTS), plain stdout logging via doca_log, or real-time event subscription back into the app via doca-comch — those belong to other skills.
- ▌ I4h Workflow Scene Edit · nvidia bundleEdit an existing workflow Scene or task contract. Use for assets, layout, cameras, randomization, task text, or success rules; do not use to create a new workflow.
- ▌ Jetson Video Capability · nvidia bundleUse when Jetson codec, profile, chroma, bit-depth, dimension, engine-count, or operational support must be reconciled from live APIs, authenticated NVIDIA samples, and NVIDIA documentation; also applies the content-DRM scope.
- ▌ Doca Container Deployment · nvidia bundleRunning an in-bundle DOCA service as a container on a BlueField (Argus, DMS, Firefly, or UROM service): kubelet standalone watching a static-pod manifests directory, YAML pod-spec drop, kubelet status / ENTRYPOINT logs / per-service liveness. Trigger on "how do I run my built service on the BlueField?", "where do I drop the pod-spec YAML?", "pod stuck in Pending / ImagePullBackOff / CrashLoopBackOff", "container Running but service isn't ready", "pod restart-loops after edit", or "DMS and Firefly together", even without naming container deployment. A DOCA-linked binary launched directly on the host or Arm belongs to doca-bare-metal-deployment. Covers smoke-before-bulk and the layered error taxonomy (pod-spec, scheduling, image pull, runtime, mount, network, version, host). Refuse and route elsewhere for per-service config schemas, DOCA install, library-API questions, external NVIDIA services (BlueMan, HBN, SNAP, Virtio-net), or full Kubernetes-cluster ops.
- ▌ Doca Public Knowledge Map · nvidia bundleUse this skill to locate authoritative NVIDIA DOCA information without the source tree: the docs.nvidia.com page for a library/service/tool, which DOCA libraries are installed and at what version, a sample on disk or GitHub, a path under /opt/mellanox/doca, or a 404'd/renamed doc URL. Activate on implicit phrasing only when the conversation already names DOCA, BlueField, ConnectX, or /opt/mellanox. Do not activate for generic 'which version is installed' or 'where is the sample' questions with none of that context. Refuse and route elsewhere for hands-on programming, env prep, API tutorials, or hardware/firmware mutation — those belong to doca-programming-guide, doca-setup, the per-library skills, and doca-hardware-safety.
- ▌ Nemo Fabric Build Adapter · nvidia bundleBuild, migrate, review, and maintain third-party NVIDIA NeMo Fabric adapters against the public adapter contract. Use when creating adapter or target descriptors, mapping AgentConfig into an agent harness or custom-agent runtime, implementing start/invoke/stop, declaring schemas and capabilities, packaging discovery metadata, or assessing adapter conformance. Do not use for consumer applications that only call the NVIDIA NeMo Fabric SDK.
- ▌ Nemo Relay Install · nvidia bundleUse this skill when choosing or running NeMo Relay installation for an agent runtime, application language, framework integration, or plugin bundle, or checking a runtime's native Relay readiness. Do not use it for first-run instrumentation or export configuration.
- ▌ Doca Bf3 Deployment · nvidia bundleUse this skill for BlueField-3 (BF3) day-1 platform bring-up via the classic RShim/BFB path: pushing a BlueField bundle (BFB) to the DPU over RShim with bfb-install from the host, the host-to-DPU TMFIFO management channel (tmfifo_net0, the 192.168.100.x convention), RShim daemon state and console-over-rshim, DPU mode selection (DPU/embedded-function vs separated-host/NIC mode) via mlxconfig, post-BFB recovery, a six-state BlueField-state classifier, and verifying the install (cat /etc/mlnx-release plus version checks). Trigger even when the user does not say "BF3" — typical phrasings include {push a BFB to my BlueField-3}, {bfb-install exited 0 but the DPU never came back}, {ping 192.168.100.2 works but ssh fails}, or {is DOCA on the host or the Arm side?}. BFB reflash, mlxconfig set, mode changes, and firmware burns are destructive: require explicit target-bound confirmation and load doca-hardware-safety. App launch, container deploy, env install, and the BF4 BMC-Redfish path route elsewhere.
- ▌ Doca Bf4 Deployment · nvidia bundleWARNING: guides IRREVERSIBLE BlueField-4 hardware operations (PLDM firmware burns, ISO reflashes, power cycles, BMC factory resets); every mutating step is governed by doca-hardware-safety (window, rollback). BlueField-4 only: BlueField-3 bring-up is doca-bf3-deployment. Use this skill for BlueField-4 (BF4) day-1 platform bring-up from the BMC: installing the BlueField/DOCA bundle ISO onto the DPU (Grace, the Arm complex) over UEFI HTTP Boot, PXE, or Redfish Virtual Media; the PLDM platform-firmware update via the Redfish UpdateService and pldmtool; and a Grace Ubuntu image with optional cloud-init. Trigger on BlueField-4/BF4 bring-up phrasings even without "BF4": {bring up my new BlueField-4}, {the BlueField ISO will not boot over HTTP from the BMC}, {attach BF4 virtual media via Redfish}, {BF4 firmware Task stuck at Running}, {prepare a BlueField-4 host: platform support, version alignment, safest first smoke test}. Route application launch and library APIs to other skills.
- ▌ Doca Erasure Coding · nvidia bundleHardware-offloaded erasure coding with DOCA Erasure Coding (Reed-Solomon, doca_ec) on a BlueField DPU, ConnectX NIC, or host: generating parity for data blocks, rebuilding lost blocks from the survivors, and refreshing parity after a change. Trigger on "one data block changed, how do I refresh parity without re-encoding", "a disk failed and 2 parity blocks are gone, can I rebuild", "RAID-6 resilience across 12 disks", "my doca_ec_task_create returns NOT_PERMITTED", or "is this N+K layout still recoverable", even without naming DOCA Erasure Coding. Covers bringing up a doca_ec context, the create / recover / update tasks, matrix type / N / K / block size, doca_ec_cap_* queries before sizing, doca_mmap src/dst permissions, and DOCA_ERROR_* from doca_ec_task_*. Refuse and route elsewhere for non-Reed-Solomon codes (fountain / LDPC / raptor), pure-replication designs, network FEC, or other DOCA accelerator libraries (SHA / Compress / AES-GCM / DMA).
- ▌ I4h Workflow Create · nvidia bundleCreate a minimal blank Workflow scaffold with a Scene containing ground and light plus an idle run mode. Use for fast new Workflow scaffolding.
- ▌ Jetson Video Recipe · nvidia bundleUse when turning a Jetson encoder use case into one surface-neutral recipe with native Video Codec SDK and PyNvVideoCodec projections.
- ▌ Nvidia Skill Finder · nvidia bundleUse for NVIDIA-related requests where an NVIDIA skill might help, even if the user did not ask for a skill. Trigger on NVIDIA products, hardware, software, SDKs, GPUs, Jetson/JetPack/L4T/BSP/SDK Manager/driver/flashing/setup, CUDA, NIM, NeMo, Omniverse/OpenUSD/SimReady, Isaac, RAPIDS/cuDF, cuPyNumeric, cuOpt, Dynamo, Holoscan, TensorRT, DeepStream, VSS, TAO, NGC/NVCF. Do not use for generic non-NVIDIA route, optimize, deploy, AI, video, data, or infrastructure tasks.
- ▌ Deepstream Run Mv3dt · nvidia bundleRun 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.
- ▌ Doca Hardware Safety · nvidia bundleUse this skill whenever the agent is about to recommend or apply a change that touches DPU / NIC hardware state on a live system — mlxconfig firmware-parameter write, NIC firmware burn, BFB reflash, NIC ↔ DPU mode flip, SR-IOV or device-emulation slot enable, kernel boot-parameter change (IOMMU, hugepages, VFIO), PCIe rebind / rescan / link-state flip, BlueField cold reboot, or remediation of a BlueField thermal shutdown. Wrap the change in pre-flight inventory, OOB reachability, a maintenance window, replica rehearsal, and rollback. Trigger on implicit phrasing: "flip BlueField mode over SSH", "enable SR-IOV and reboot", "burned firmware but mlxconfig shows old value", "BlueField thermal shutdown". Not for read-only inventory or version queries (doca-version, doca-caps). Refuse orientation (doca-public-knowledge-map), install or env debug (doca-setup), and program-side debug (doca-debug, doca-programming-guide).
- ▌ Doca Telemetry Utils · nvidia bundledoca_telemetry_utils command line on a host with DOCA: discover the diagnostic-counter schema, translate counter names to binary Data IDs and back, and check per-device counter support before committing a DOCA Telemetry exporter config. Use this skill when telemetry counters vanish between the exporter and the collector: "my exporter ships but the collector sees nothing", "this metric silently drops downstream", "which counters does this BlueField expose", "translate this 0x... back to a counter name", "what do node / pcie_index / depth mean here", or "is this counter supported on this device", even without naming the tool. Route exporter or collector code to doca-telemetry-exporter / doca-telemetry, DTS deployment to doca-dts, and DOCA install or repair to doca-setup.
- ▌ I4h Workflow Finetune · nvidia bundleFine-tune a manifest-backed GR00T or openpi remote Task on compatible LeRobot data. Use for training; do not use for inference-only Tasks or checkpoint rollout.
- ▌ I4h Workflow Train Rl · nvidia bundleUse when training, evaluating, or exporting Workflow policies with online RSL-RL or RLinf, including RL checkpoint and Workflow handoff.