@skills · Owner

Agent skills by NVIDIA

350 skills indexed from github.com/NVIDIA. Reference any of them in AdaL, Claude Code, Cursor or any coding agent — nothing to install.

• accelerated-computing-cudf · Skill

  Official NVIDIA-authored guidance for NVIDIA cuDF GPU DataFrames, pandas acceleration, dask-cuDF, ETL, joins, groupby, CSV/Parquet I/O, nullable semantics,

• aiq-deploy · Skill

  Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.

• aiq-research · Skill

  Use when asked to run deep research or AI-Q research through a reachable NVIDIA AI-Q Blueprint backend.

• amc-run-rtsp-calibration · Skill

  Calibrate a new dataset from live RTSP camera streams via the AutoMagicCalib REST API. Use when the user provides RTSP URLs or asks to calibrate live camer

• amc-run-sample-calibration · Skill

  Run end-to-end calibration on the shipped sample dataset (sdg_08_2_sample_data_010926.zip) against a running AMC microservice. Use when user says 'test sam

• amc-run-video-calibration · Skill

  Calibrates pre-recorded `cam_*.mp4` datasets through the AutoMagicCalib REST API. Use for user-supplied local MP4s; route live RTSP streams to `amc-run-rts

• amc-setup-calibration-stack · Skill

  Launch AutoMagicCalib microservice and web UI from NGC release images via Docker Compose. Use when user says 'deploy auto calibration', 'launch auto calibr

• cudaq-guide · Skill

  CUDA-Q onboarding guide for installation, test programs, GPU simulation, QPU hardware, and quantum applications.

• cuopt-developer · Skill

  Modify, build, test, debug, and contribute to NVIDIA cuOpt (C++/CUDA, Python, server, CI). Use for solver internals, PRs, DCO, and code conventions.

• cuopt-install · Skill

  Install cuOpt for Python, C, or server via pip, conda, or Docker; verify the install. For building cuOpt from source, see cuopt-developer.

• cuopt-multi-objective-exploration · Skill

  Trace, complete, and interpret the Pareto frontier across competing objectives using repeated single-objective cuOpt solves (weighted-sum and ε-constraint)

• cuopt-numerical-optimization-api · Skill

  LP, MILP, and QP (beta) with cuOpt — Python, C, and CLI. Use when the user is solving LP, MILP, or QP with any cuOpt interface.

• cuopt-numerical-optimization-formulation · Skill

  LP, MILP, QP — concepts, problem-text parsing, and formulation patterns (parameters, constraints, decisions, objective). Concepts only; no API.

• cuopt-routing-api-python · Skill

  Vehicle routing (VRP, TSP, PDP) with cuOpt — Python API only. Use when the user is building or solving routing in Python.

• cuopt-server-api-python · Skill

  cuOpt REST server — start server, endpoints, Python/curl client examples. Use when the user is deploying or calling the REST API.

• cupynumeric-hdf5 · Skill

  Read and write large cuPyNumeric arrays to HDF5 with Legate's parallel, distributed HDF5 I/O (legate.io.hdf5: to_file, from_file, from_file_batched). Use w

• cupynumeric-install · Skill

  Install and verify cuPyNumeric for Python — requirements, commands, verification. Source builds are out of scope.

• cupynumeric-migration-readiness · Skill

  Pre-migration readiness assessor for porting NumPy to cuPyNumeric. Use BEFORE substantial porting work begins when the user asks whether code will scale on

• cupynumeric-parallel-data-load · Skill

  Load a sharded, on-disk dataset (sharded .npy, Parquet/Arrow, raw binary, sharded HDF5, custom layouts) into a distributed cuPyNumeric ndarray via a manual

• dali-dynamic-mode · Skill

  DALI imperative dynamic mode (`nvidia.dali.experimental.dynamic`, ndd): use when working on ndd code or migrating pipelines; skip pipeline-only tasks.

• data-designer · Skill

  Use when the user wants to create a dataset, generate synthetic data, or build a data generation pipeline.

• deepstream-dev · Skill

  NVIDIA DeepStream SDK development with Python pyservicemaker API. Use when building video analytics pipelines, GStreamer-based video processing, TensorRT i

• deepstream-generate-pipeline · Skill

  Build DeepStream GStreamer pipelines interactively. Use when the user asks about pipelines for video/image inference, detection, tracking, or streaming — i

• deepstream-import-vision-model · Skill

  Use this skill to bring a supported object-detection vision model from HuggingFace or NVIDIA NGC into an NVIDIA DeepStream pipeline with end-to-end automat

• deepstream-profile-pipeline · Skill

  Profile a DeepStream pipeline with Nsight Systems and derive its configs from the measurement. Use when the user asks for an efficient, performant, or prof

• deepstream-run-mv3dt · Skill

  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

• deepstream-sop · Skill

  Use this skill when building, deploying, evaluating, debugging, or measuring latency for the DeepStream SOP Inference Microservice — a GPU-accelerated Fast

• dicom-metadata-extract · Skill

  Used for extracting selected metadata from one DICOM file and flagging standard-tag PHI presence. Not for anonymization or clinical use.

• dicom-series-preflight · Skill

  Used for header-only preflight of one DICOM series folder before conversion or inference. Not for de-identification or clinical clearance.

• dicom-series-to-volume · Skill

  Used for converting one CT DICOM series folder to a HU NIfTI volume with affine evidence. Not for multi-frame DICOM or clinical use.

• digital-health-clinical-asr-build · Skill

  Stage 2 of the Clinical ASR Flywheel. Use when curating clinical terms, tagging IPA, and synthesizing a NeMo manifest. NOT for scoring (use /digital-health

• digital-health-clinical-asr-eval · Skill

  Stage 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).

• digital-health-clinical-asr-finetune · Skill

  Stage 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 ge

• digital-health-clinical-asr-setup · Skill

  Stage 1 of Clinical ASR Flywheel. Use when bootstrapping a cycle: NVCF+MW disclosure, NVIDIA_API_KEY check, deps install, TTS+ASR smoke test.

• doca-aes-gcm · Skill

  Use this skill when the user is doing hands-on DOCA AES-GCM work on a BlueField DPU or ConnectX NIC — configuring `doca_aes_gcm_task_encrypt` / `_task_decr

• doca-argp · Skill

  Use this skill for hands-on DOCA Arg Parser CLI work on a shipped sample or new DOCA-using app — adding / removing / renaming flags; wiring `doca_argp_init

• doca-argus · Skill

  Use this skill when the user is deploying or operating the DOCA Argus Service — the packaged BlueField-side runtime-security container that watches the Blu

• doca-bare-metal-deployment · Skill

  Use this skill for launching, supervising, debugging, OR platform lifecycle on a BlueField — BFB install, RShim/TMFIFO, host PF rebind, post-BFB recovery —

• doca-bench · Skill

  Run `doca_bench` (DOCA 2.7.0 or newer) to measure throughput, bulk latency, precision latency, or maximum bandwidth for RDMA, Compress, AES-GCM, SHA, DMA,

• doca-bench-extension · Skill

  Use this skill when the operator is authoring, building, loading, or debugging a custom doca-bench plug-in — a versioned shared library with DOCA_EXPERIMEN

• doca-bf3-deployment · Skill

  Use 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 bf

• doca-bf4-deployment · Skill

  WARNING: guides potentially IRREVERSIBLE BlueField-4 hardware operations (PLDM firmware burns, ISO reflashes, power cycles, BMC factory resets) that can br

• doca-caps · Skill

  Use this skill when the user wants to invoke the read-only doca_caps CLI to ask what DOCA sees on this host — listing DOCA devices and PCIe addresses, list

• doca-collectx-deployment · Skill

  Use this skill to deploy and operate a CollectX (clx) based DOCA telemetry collector on a host or BlueField — wiring providers / counters into the collecto

• doca-comch · Skill

  Use this skill when the user is doing hands-on DOCA Comch work on a host + BlueField pair — bringing up host ↔ DPU PCIe control-plane messaging, picking se

• doca-comm-channel-admin · Skill

  Use this skill to enumerate host↔DPU DOCA comch (formerly Comm Channel) servers and connections via the shipped doca_comm_channel_admin binary — listing co

• doca-common · Skill

  Use this skill whenever the user is doing hands-on DOCA programming on a BlueField DPU or ConnectX NIC and needs the foundation primitives every per-librar

• doca-compress · Skill

  Use this skill for hands-on DOCA Compress programming on a BlueField DPU, ConnectX NIC, or host with DOCA — enabling compress-deflate, decompress-deflate,

• doca-container-deployment · Skill

  Use this skill when the user is hands-on deploying an in-bundle DOCA service container (Argus, DMS, Firefly, or UROM service) on a BlueField — kubelet stan

• doca-debug · Skill

  Use this skill when the user is debugging any DOCA symptom — a build that won't compile, a link step that can't resolve a doca_* symbol, a runtime call ret

• doca-devemu · Skill

  Use this skill when the user is doing hands-on DOCA Device Emulation on a BlueField DPU — exposing a custom emulated PCIe device the host sees as a real pe

• doca-dma · Skill

  Use this skill when the user is doing hands-on DOCA DMA programming — bringing up a doca_dma context, configuring the single doca_dma_task_memcpy task type

• doca-dms · Skill

  Operate NVIDIA DOCA Management Service (`dmsd` + `dmspe`) on a BlueField, Arm/x86 host, or Kubernetes pod: choose deployment and authentication, configure

• doca-dpa · Skill

  Use this skill when the user is doing hands-on DOCA DPA host-side work on a BlueField — creating the `doca_dpa` Core context, loading a DPACC-compiled DPA

• doca-dpa-hl-tracer · Skill

  Use this skill when the user runs doca_dpa_hl_tracer to capture/decode DPA-side traces at the programming-events layer (kernel entry/exit, sync points, com

• doca-dpdk-bridge · Skill

  Use this skill when the user has an existing DPDK application and is adding DOCA capabilities in-place — most commonly DOCA Flow hardware steering — withou

• doca-erasure-coding · Skill

  Use this skill when the user is doing hands-on DOCA Erasure Coding programming on a BlueField DPU, ConnectX NIC, or host — bringing up a doca_ec context, p

• doca-eth · Skill

  Use this skill for hands-on DOCA Ethernet packet-queue work on a BlueField DPU or ConnectX NIC — bringing up a `doca_eth_rxq` or `doca_eth_txq` on a port /

• doca-firefly · Skill

  Use this skill when the user is operating the DOCA Firefly Service container on BlueField — picking the four PTP configuration axes (role / profile / domai

• doca-flow · Skill

  Build and debug DOCA Flow applications on supported NVIDIA NICs/DPUs: define match/action pipes, initialize ports and representors, choose forwarding targe

• doca-flow-dpa-perf · Skill

  Use this skill when the user is invoking doca_flow_dpa_perf on DPA-capable hardware (ConnectX-7 minimum supported, ConnectX-8 recommended, or BlueField-3)

• doca-flow-dpa-provider · Skill

  Use this skill when the user is doing hands-on DOCA Flow DPA Provider work — exporting a `doca-flow` pipe or external resource (index-selector/memory) into

• doca-flow-grpc-server · Skill

  PLAINTEXT-ONLY: the shipped `doca_flow_grpc` server uses `grpc::InsecureServerCredentials()` with NO TLS / mTLS / token-auth knob on the binary — transport

• doca-flow-perf · Skill

  Use this skill when the user is measuring the host or DPU-CPU control-plane rate of a DOCA Flow pipeline with doca_flow_perf — picking a JSON policy from c

• doca-flow-tune · Skill

  Use this skill when the user is tuning a live or captured `doca-flow` pipeline with `doca_flow_tune` — snapshotting pipe / counter / KPI state, picking a t

• doca-gpi · Skill

  Use this skill for hands-on DOCA GPI programming — wiring a GPU-Packet-Initiator context so a CUDA kernel drives RDMA queues directly from GPU memory witho

• doca-gpunetio · Skill

  Use this skill when the user is doing hands-on DOCA GPUNetIO programming — wiring a CUDA kernel on an NVIDIA GPU to a doca-eth queue via doca_gpu_eth_rxq /

• doca-gpunetio-ib-write-bw · Skill

  Use this skill when the user is building, running, or interpreting the doca/tools/gpunetio_ib_write_bw client+server benchmark — a CUDA kernel on the clien

• doca-gpunetio-ib-write-lat · Skill

  Use this skill when the user is measuring GPU-kernel-initiated RDMA WRITE latency through doca-gpunetio — building and running the `gpunetio_ib_write_lat`

• doca-hardware-safety · Skill

  Use 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-param

• doca-mgmt · Skill

  Use this skill when the user is doing hands-on DOCA Management programming against BlueField / ConnectX devices — standing up a management or representor c

• doca-pcc · Skill

  Use this skill when the user is doing hands-on host-side DOCA PCC work to load a CUSTOM Programmable Congestion Control algorithm onto a BlueField DPU — cr

• doca-pcc-counters · Skill

  Use this skill when the user is invoking the DOCA PCC Counters tool — the `pcc_counters.sh` bash script under the DOCA tools directory — to arm and read th

• doca-pcc-ztr-rttcc-algo · Skill

  Use this skill when the user is doing hands-on deployment, tuning, or evaluation of the DOCA-shipped Zero-Touch RoCE RTT-based Congestion Control (ZTR RTTC

• doca-programming-guide · Skill

  Use 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 modif

• doca-public-knowledge-map · Skill

  Use this skill when the user needs to locate authoritative information about NVIDIA DOCA without access to the source tree — finding the right docs.nvidia.

• doca-rdma · Skill

  Use this skill when the user is doing hands-on DOCA RDMA programming on a BlueField DPU, ConnectX NIC, or DOCA host — bringing up an RDMA context on a doca

• doca-rdmi · Skill

  Use this skill when the user is doing hands-on DOCA RDMI (RDMA Initiator) programming — picking doca-rdmi vs doca-rdma for an accelerator-initiated one-sid

• doca-rmax · Skill

  Use this skill when the user is doing hands-on DOCA Rivermax work on a BlueField DPU or ConnectX host — standing up `doca_rmax_in_stream` (receive) session

• doca-setup · Skill

  Use this skill when the user is dealing with the DOCA environment around their workload — verifying an install is healthy, preparing the build env (pkg-con

• doca-sha · Skill

  Use this skill when the user is doing hands-on DOCA SHA programming — offloading SHA-1, SHA-256, or SHA-512 hashing onto a BlueField DPU or ConnectX accele

• doca-sha-offload-engine · Skill

  Use 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

• doca-socket-relay · Skill

  Use this skill when the operator is driving the DOCA Socket Relay to bridge a socket-oriented host application onto a BlueField DPU peer without rewriting

• doca-spcx-cc · Skill

  Use this skill when the user is invoking `doca_spcx_cc` (the host-side CLI under /opt/mellanox/doca/tools/) to load, parameterize, start, observe, or stop

• doca-sta · Skill

  Use this skill when the user is doing hands-on NVMe-over-Fabrics storage-target work on a BlueField DPU or ConnectX NIC with DOCA STA — standing up a doca_

• doca-structured-tools-contract · 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 t

• doca-telemetry · Skill

  Use this skill to read DOCA hardware-counter events from a `doca_dev` through the per-domain Telemetry reader libraries: `doca_telemetry_pcc`, `_dpa`, `_di

• doca-telemetry-exporter · Skill

  Use 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_s

• doca-telemetry-utils · Skill

  Use this skill when the user is invoking `doca_telemetry_utils` on a host with DOCA installed — discovering the diagnostic-counter schema, translating coun

• doca-upgrade · Skill

  Use this skill when the user is contemplating a DOCA upgrade or downgrade — moving a host to a newer DOCA release, refreshing the BlueField BFB, bumping th

• doca-urom · Skill

  Use this skill when the user is doing hands-on DOCA UROM library work from the host side — wiring doca-urom under an HPC / UCX / MPI stack to OFFLOAD remot

• doca-urom-svc · Skill

  Operate the DOCA UROM Service container on BlueField Arm for remote memory operations (puts, gets, atomics, collectives) enqueued by a paired host using `d

• doca-verbs · Skill

  Use this skill when the user is dropping below the higher-level DOCA libraries (doca-rdma / doca-eth / doca-rmax) into the raw-verbs escape hatch — managin

• doca-version · Skill

  Use this skill when the user is doing DOCA version handling — detecting the installed release, validating the four-way match across pkg-config doca-common,

• dynamo-interconnect-check · Skill

  Validate that a Dynamo deployment's NIXL/UCX/NCCL interconnect is ready for disaggregated serving over RDMA/NVLink. Use after recipe-runner brings a deploy

• dynamo-recipe-runner · Skill

  Select, validate, patch, and deploy existing NVIDIA Dynamo Kubernetes recipes. Use for model/backend/GPU/deployment-mode recipe bring-up; use router-starte

• dynamo-router-starter · Skill

  Start 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 re

• dynamo-troubleshoot · Skill

  Diagnose failed or unhealthy Dynamo deployments. Use when pods, model-cache jobs, PVCs, workers, frontend/router health, endpoints, or benchmark jobs fail;

• earth2studio-create-datasource · Skill

  Create and validate Earth2Studio data source wrappers (DataSource, ForecastSource, DataFrameSource, ForecastFrameSource) from remote stores. Do NOT use for

• earth2studio-create-diagnostic · Skill

  Create Earth2Studio diagnostic model wrappers for single-step data transformations, including simple derived diagnostics, packaged AutoModel diagnostics, a

• earth2studio-create-prognostic · Skill

  Create Earth2Studio prognostic (time-stepping forecast) model wrappers. Do NOT use for diagnostic models, data sources, or installation.

• earth2studio-data-fetch · Skill

  Fetch weather/climate data via Earth2Studio data sources for specific variables and times. Do NOT use for inference pipelines, model discovery, or installa

• earth2studio-deterministic-forecast · Skill

  Build deterministic forecast scripts with Earth2Studio (model, data source, IO, inference). Do NOT use for ensemble, diagnostics, data-only fetch, or insta

• earth2studio-discover · Skill

  Find Earth2Studio models, data sources, and examples for a weather/climate use case. Do NOT use for writing inference code, downloading data, or installati

• earth2studio-install · Skill

  Guide installing Earth2Studio via uv or pip, selecting model extras, and configuring the environment. Do NOT use for writing inference code, choosing model

• holohub-app-lifecycle · Skill

  Use for non-failing HoloHub app work with ./holohub: scaffold, build, run, test, visual evidence, lint, and flow benchmarking.

• holohub-debug-build-run · Skill

  Use when a concrete ./holohub command fails, hangs, regresses, or returns wrong output and needs reproducible diagnosis and verification.

• holohub-module-lifecycle · Skill

  Use for reusable Holoscan Module work with ./holohub: scaffold, tests, editable install, DEB/WHEEL packaging, and clean-consumer proof.

• holoscan-install-conda · Skill

  Install Holoscan SDK v4.3+ via Conda in a CUDA 13 environment. Use for Conda installs; redirect CUDA 12 hosts to container/wheel.

• holoscan-install-container · Skill

  Install Holoscan SDK via the NGC Docker container. Use for container-based installs; not for native apt/pip/Conda installs.

• holoscan-install-debian · Skill

  Install Holoscan SDK natively on Ubuntu via apt. Use for C++ installs on Ubuntu; pair with /holoscan-install-wheel for Python.

• holoscan-install-source · Skill

  Build Holoscan SDK from source via the in-tree ./run script. Use only when published packages don't meet the user's needs.

• holoscan-install-wheel · Skill

  Install Holoscan SDK Python wheel via pip into a venv. Use for Python installs; not for native C++/apt or Conda installs.

• holoscan-setup · Skill

  Guides Holoscan SDK installation: inspects the host, assesses platform compatibility, recommends an install method, and delegates to the matching install s

• hsb-app · Skill

  Discover and run Holoscan Sensor Bridge example applications on a connected devkit. Filters available apps by the user's platform, HSB software version, bo

• hsb-flash · Skill

  Flash 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

• hsb-ip-create-top · Skill

  Create or explain fixed-format HSB FPGA_top.sv wrappers from validated HOLOLINK_def.svh files. Do not use for def generation or validation.

• hsb-ip-def · Skill

  Generate, validate, compare, or explain HSB HOLOLINK_def.svh macros. Do not use for FPGA_top.sv wrappers or packetizer-only derivation. Generation runs bun

• hsb-ip-packetizer · Skill

  Choose or explain HSB Sensor RX packetizer fields for HOLOLINK_def.svh. Do not use for full defs, validation, or runtime APB programming.

• hsb-setup · Skill

  Clone the latest NVIDIA Holoscan Sensor Bridge repo, ask which supported devkit is being used, configure the host per platform, build the correct demo cont

• hsb-test · Skill

  Execute QA test plans on Holoscan Sensor Bridge hardware. Reads a user-provided test document, filters tests by the user's setup, determines which tests ca

• i4h-lerobot-viz · Skill

  Serve and visually inspect a converted LeRobot dataset in the browser. Use for videos and state/action timelines; do not use for raw workflow HDF5 or incom

• i4h-workflow · Skill

  Orient users to the i4h workflow runtime and route them to the correct stage skill. Use for architecture, support, or where-to-start questions; do not exec

• i4h-workflow-create · Skill

  Create a minimal blank Workflow scaffold with a Scene containing ground and light plus an idle run mode. Use for fast new Workflow scaffolding.

• i4h-workflow-dataset-annotate · Skill

  Grade or filter workflow HDF5 episodes with an OpenAI-compatible vision model. Use for visual success labels; do not use for replay, policy evaluation, or

• i4h-workflow-dataset-convert · Skill

  Convert workflow HDF5 recordings to LeRobot datasets for training or browser inspection. Use for conversion; do not use for replay, augmentation, or raw-da

• i4h-workflow-dataset-mimic · Skill

  Expand workflow HDF5 demonstrations with action jitter, optionally scoped to node segments. Use for synthetic variants; do not use to collect data, alter s

• i4h-workflow-dataset-replay · Skill

  Replay a workflow HDF5 episode through its original Scene. Use for visual trajectory and recording verification; do not use for policy evaluation or LeRobo

• i4h-workflow-dataset-teleop · Skill

  Record 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

• i4h-workflow-e2e · Skill

  Run the maintained workflow data-to-policy pipeline from recording through checkpoint validation. Use for full end-to-end requests; do not use for one indi

• i4h-workflow-finetune · Skill

  Fine-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 rol

• i4h-workflow-scene-edit · Skill

  Edit 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 w

• i4h-workflow-setup · Skill

  Preflight and set up the root-level workflow runtime. Use for installation, missing component environments, or third-party failures; do not use for rollout

• i4h-workflow-train-rl · Skill

  Use when training, evaluating, or exporting Workflow policies with online RSL-RL or RLinf, including RL checkpoint and Workflow handoff.

• i4h-workflow-validate · Skill

  Run the root-level workflow runtime policy or rule-based rollouts and verify simulator success. Use for evaluation, checkpoints, or local controllers; do n

• jetson-build-source · Skill

  Use when you need to rebuild the BSP overlay — DT, OOT modules, or kernel — from changes under bsp_sources/. Triggers: build bsp, rebuild dtb, rebuild kern

• jetson-customize-camera · Skill

  Enable MIPI/GMSL camera sensors on a Jetson Thor or Orin custom carrier by rendering a kernel-DT overlay from the in-tree sensor DTSI. Do NOT use for UPHY

• jetson-customize-clocks · Skill

  Use to lock/cap Jetson CPU/GPU/EMC clocks, toggle EMC/CPU DVFS, or change cpufreq governors by editing BPMP DTB and nvpower.sh pre-flash. Do NOT use for li

• jetson-customize-fan · Skill

  Use when you need to add, remove, edit, list, or change the boot default of an nvfancontrol fan profile on a Jetson/Tegra (Orin, Thor) target. Triggers: ed

• jetson-customize-mgbe · Skill

  Enable Jetson Thor 25G/10G/1G MGBE QSFP via kernel-DT overlay. Do NOT use for UPHY lane allocation or ODMDATA edits.

• jetson-customize-nvpmodel · Skill

  Use when you need to add, remove, edit, list, or change the boot default of an nvpmodel power mode on a Jetson/Tegra (Orin, Thor) target. Triggers: edit po

• jetson-customize-pcie · Skill

  Per-controller PCIe enable / disable / lanes / link-speed for a Jetson Thor or Orin custom carrier via ODMDATA + kernel-DT overlay. Do NOT use for UPHY lan

• jetson-customize-pinmux · Skill

  Per-pin SFIO / direction / initial-state configurator for a Jetson Orin or Thor custom carrier from the pinmux XLSM. Do NOT use for kernel-DT overlay or OD

• jetson-customize-uphy · Skill

  Configure Jetson UPHY lane allocation (uphy0/uphy1-config) on Orin/Thor custom carriers. Do NOT use for pinmux or PCIe-only edits.

• jetson-customize-usb · Skill

  Enable/disable Jetson USB2/USB3 SS ports via kernel-DT overlay. Do NOT use for UPHY lane allocation or ODMDATA edits.

• jetson-derive-carrier · Skill

  Bootstrap a custom carrier board by forking carrier files and scaffolding a DT overlay from the reference devkit. Use after jetson-init-source; not for mod

• jetson-diagnostic · Skill

  Read-only Jetson health snapshot for identity, memory, GPU, thermal, power, storage, services, and top processes.

• jetson-download-bsp · Skill

  Download NVIDIA Jetson Linux BSP artifacts (BSP tarball, sample rootfs, public_sources, x-tools, guides) for the active target. Use for Auto Setup; not for

• jetson-flash-image · Skill

  Use to flash a promoted BSP image to a Jetson DUT in RCM mode via flash.sh or l4t_initrd_flash.sh. Do NOT use for BSP customization, image promotion, or ca

• jetson-generate-kb · Skill

  Build a per-target knowledge-base markdown next to the active profile by walking the BSP root and source tree. Use after init-image / init-source; not for

• jetson-headless-mode · Skill

  Plan and apply safe Jetson headless-mode changes to reclaim GUI and daemon memory.

• jetson-inference-mem-tune · Skill

  Pick the serving stack and per-runtime memory flags (vLLM, SGLang, llama.cpp, TensorRT Edge-LLM) for an LLM/VLM workload on any NVIDIA Jetson.

• jetson-init-image · Skill

  Extract Jetson Linux + sample-rootfs tarballs and run apply_binaries.sh for the active target, then record bsp_image in the profile. Use after jetson-init-

• jetson-init-source · Skill

  Set up the BSP source workspace: Linux_for_Tegra overlay tracker, bsp_sources, Crosstool-NG toolchain. Use after jetson-init-image; not for fetching inputs

• jetson-init-target · Skill

  Author a new Jetson target-platform profile (reference_devkit + optional custom_carrier) and update the active pointer. Use to create a target; not for swi

• jetson-link-docs · Skill

  Bind pre-downloaded Jetson reference docs (developer guide, design guide, pinmux, schematics) into the active profile documents block. Use after staging do

• jetson-llm-benchmark · Skill

  Benchmark Jetson LLM/VLM serving performance across vLLM, llama.cpp, and Ollama with structured JSON output.

• jetson-llm-serve · Skill

  Stand up vLLM or SGLang serving on Jetson, using upstream vLLM on Thor and Orin JetPack 7.2+, and NVIDIA-AI-IOT vLLM on older Orin.

• jetson-memory-audit · Skill

  Measure Jetson DRAM/NvMap usage and verify before/after memory reclamation with live audit data.

• jetson-optimize-memory · Skill

  Reclaim DRAM by disabling unused subsystems across MB1 BCT, MB2 BCT, kernel reserved-memory, and SWIOTLB. Use for headless or no-camera Jetson deployments;

• jetson-package · Skill

  Pick Jetson-compatible containers, vLLM runtime images, and Jetson AI Lab PyPI indexes; maps Orin SM 8.7 vs Thor SM 11.0 and JetPack-specific package choic

• jetson-print-bsp-info · Skill

  Use when you need to print Jetson BSP info (L4T version, board configs, rootfs state) from a Linux_for_Tegra root on the host PC. This is an example skill.

• jetson-print-device-info · Skill

  Use when you need to print Jetson device info (module model, L4T version, kernel, OS version, current power mode) from a running Jetson target. This is an

• jetson-promote-image · Skill

  Use to promote overlay files and built artifacts into the staged BSP image. Do NOT use to flash or build. Triggers: promote bsp image.

• jetson-quick-start · Skill

  Entry skill for Jetson / IGX BSP customization. Asks one core click-to-select setup questionnaire and passes prefilled answers to downstream setup skills.

• jetson-set-target · Skill

  Switch the active Jetson target-platform pointer to an existing profile YAML. Use before customize/build/flash to change target; not for authoring profiles

• jetson-speculative-decoding · Skill

  Add EAGLE-3 or draft-model speculative decoding to a Jetson vLLM server when TPOT is the bottleneck.

• jetson-validate-image · Skill

  Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers:

• jetson-video-benchmark · Skill

  Use when measuring Jetson Video Codec SDK or PyNvVideoCodec encode/decode throughput, comparing presets or surfaces, testing codec-worker capacity with aut

• jetson-video-capability · Skill

  Use when Jetson codec, profile, chroma, bit-depth, dimension, engine-count, or operational support must be reconciled using live SDK APIs, authenticated NV

• jetson-video-pipeline · Skill

  Use when executing and verifying Jetson Video Codec SDK or PyNvVideoCodec encode/decode, transcode, segmentation, container decode, AV1, or acceptance work

• jetson-video-recipe · Skill

  Use when turning a Jetson encoder use case into one validated surface-neutral recipe with native and PyNvVideoCodec projections for codec, preset, rate con

• jetson-video-setup · Skill

  Use when installing, repairing, probing, or verifying native NVIDIA Video Codec SDK or PyNvVideoCodec on Jetson with official encode-to-decode samples, inc

• launch-nemo-rl · Skill

  Playbook for launching, monitoring, stopping, and debugging NeMo-RL recipes on a Kubernetes cluster via the nrl-k8s CLI. Covers ephemeral vs long-lived Ray

• mcore-create-issue · Skill

  Investigate a failing GitHub Actions run or job and create a GitHub issue for the failure.

• mcore-linting-and-formatting · Skill

  Linting and formatting for Megatron-LM. Covers running autoformat.sh, tools (ruff, black, isort, pylint, mypy), and code style rules.

• mcore-run-on-slurm · Skill

  How to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.

• mcore-split-pr · Skill

  Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.

• mcore-testing · Skill

  Test system for Megatron-LM. Covers test layout, recipe YAML structure, adding and running unit and functional tests, golden values, marker filters, and CI

• medtech-model-evidence-export · Skill

  Exports sanitized metadata, parameters, reproducibility details, quality metrics, and optional review artifacts from Medical AI inference runs or evidence

• nemo-automodel-distributed-training · Skill

  Guide for selecting and configuring distributed training strategies in NeMo AutoModel, including FSDP2, Megatron FSDP, DDP, and parallelism settings.

• nemo-automodel-launcher-config · Skill

  Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.

• nemo-automodel-model-onboarding · Skill

  Guide for onboarding new model architectures into NeMo AutoModel, including architecture discovery, implementation patterns, registration, and validation.

• nemo-automodel-recipe-development · Skill

  Create and modify NeMo AutoModel training and evaluation recipes, including YAML structure, builders, and execution flow.

• nemo-fabric-build-adapter · Skill

  Build, migrate, review, and maintain third-party NVIDIA NeMo Fabric adapters against the public adapter contract. Use when creating adapter or target descr

• nemo-fabric-integrate · Skill

  Use this skill when integrating NVIDIA NeMo Fabric into a consumer application, service, evaluation harness, or platform through the typed Python SDK — tra

• nemo-mbridge-mlm-bridge-training · Skill

  Run Megatron-LM (MLM) and Megatron Bridge training with mock or real data. Covers correlation testing, available recipes, and multi-GPU examples.

• nemo-mbridge-multi-node-slurm · Skill

  Convert single-node scripts to multi-node Slurm sbatch jobs and debug common multi-node failures. Covers srun-native vs uv run torch.distributed approaches

• nemo-mbridge-perf-activation-recompute · Skill

  Validate and use selective and full activation recompute in Megatron Bridge to reduce GPU memory usage at the cost of extra compute. Use for activation mem

• nemo-mbridge-perf-cpu-offloading · Skill

  Validate and use CPU offloading in Megatron Bridge, including layer-level activation offloading and fractional optimizer state offloading with HybridDevice

• nemo-mbridge-perf-cuda-graphs · Skill

  Validate and use CUDA graph capture in Megatron Bridge, including local full-iteration graphs and Transformer Engine scoped graphs for attention, MLP, and

• nemo-mbridge-perf-expert-parallel-overlap · Skill

  Validate and use MoE expert-parallel communication overlap in Megatron-Bridge, including overlap_moe_expert_parallel_comm, delay_wgrad_compute, and flex di

• nemo-mbridge-perf-hierarchical-context-parallel · Skill

  Operational guide for enabling hierarchical context parallelism in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification.

• nemo-mbridge-perf-megatron-fsdp · Skill

  Operational guide for enabling Megatron FSDP in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification.

• nemo-mbridge-perf-memory-tuning · Skill

  Techniques for reducing peak GPU memory in Megatron Bridge — expandable segments, PEFT + SP input re-gather, parallelism resizing, activation recompute, CP

• nemo-mbridge-perf-moe-comm-overlap · Skill

  MoE expert-parallel communication overlap in Megatron Bridge. Covers dispatch/combine overlap, flex dispatcher backends, and expert wgrad scheduling.

• nemo-mbridge-perf-moe-dispatcher-selection · Skill

  Choose the right MoE token dispatcher (`alltoall`, DeepEP, or HybridEP) for the hardware, EP degree, and optimization stage. Summarizes patterns from DSV3,

• nemo-mbridge-perf-moe-hardware-configs · Skill

  Representative, point-in-time MoE training playbooks by hardware and model family. Use them as candidate seeds, then revalidate the exact runtime, semantic

• nemo-mbridge-perf-moe-long-context · Skill

  Long-context MoE training guidance for Megatron Bridge. Covers CP sizing, selective recompute, dispatcher choices, and practical patterns from DSV3, Qwen3,

• nemo-mbridge-perf-moe-optimization-workflow · Skill

  Evidence-gated workflow for MoE performance optimization in Megatron Bridge. Covers measurement contracts, the Three Walls framework, parallel folding, pro

• nemo-mbridge-perf-moe-vlm-training · Skill

  Practical guidance for training MoE VLMs in Megatron Bridge. Compares FSDP and 3D-parallel approaches, using rounded lessons from Qwen3-VL, Qwen3-Next, and

• nemo-mbridge-perf-parallelism-strategies · Skill

  Operational guide for choosing and combining parallelism strategies in Megatron Bridge, including sizing rules, hardware topology mapping, and combined par

• nemo-mbridge-perf-sequence-packing · Skill

  Validate and use packed sequences and long-context training in Megatron-Bridge, including offline LLM packing, collate-time VLM packing, Energon online pac

• nemo-mbridge-perf-tp-dp-comm-overlap · Skill

  Operational guide for enabling TP, DP, and PP communication overlap in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification.

• nemo-mbridge-recipe-recommender · Skill

  Recommend and customize Megatron Bridge library and benchmark recipes for a user's model, GPU count, hardware, sequence length, and pretrain/SFT/PEFT goal.

• nemo-mbridge-resiliency · Skill

  Resiliency features in Megatron Bridge including fault tolerance, straggler detection, in-process restart, preemption, and re-run state machine.

• nemo-relay-debug-runtime-integration · Skill

  Use this skill when NeMo Relay is installed or imported but application-side runtime behavior is missing or incorrect, including load failures, inactive sc

• nemo-relay-get-started · Skill

  Use this skill when first-time NeMo Relay users want to try Relay, choose the least-complex supported quick start, or verify initial value through the CLI,

• nemo-relay-install · Skill

  Use this skill when choosing or running NeMo Relay installation for the CLI, Python, Node.js, Rust, OpenClaw, Hermes, or maintained framework integrations

• nemo-relay-instrument-calls · Skill

  Use this skill when an application owns tool or LLM/provider call sites and needs to wrap them with NeMo Relay scopes and managed execution APIs for lifecy

• nemo-relay-instrument-context-isolation · Skill

  Use this skill when concurrent requests, async tasks, threads, workers, goroutines, or agents need independent NeMo Relay scope stacks and correct ancestry

• nemo-relay-instrument-typed-wrappers · Skill

  Use this skill when adding NeMo Relay typed wrappers, domain types, or provider codecs while preserving JSON middleware semantics and caller-visible behavi

• nemo-relay-migrate-from-flow · Skill

  Use this skill when migrating applications, examples, integrations, documentation, manifests, or repository code from NeMo Flow to NeMo Relay across Python

• nemo-relay-plugin-adaptive-tuning · Skill

  Use this skill when baseline NeMo Relay instrumentation exists and the user wants to configure or evaluate adaptive plugin behavior, including telemetry, s

• nemo-relay-plugin-build · Skill

  Use this skill when building or packaging reusable NeMo Relay runtime behavior as an embedded configuration component or a manifest-backed `rust_dynamic` n

• nemo-relay-plugin-observability · Skill

  Use this skill when choosing or configuring NeMo Relay 0.6 or 0.7 observability through the built-in plugin, subscribers, or exporters, including raw ATOF

• nemo-retriever · Skill

  Use when the user wants to search, query, extract, transcribe, describe, quote, filter, or aggregate across documents — PDFs, scanned forms / images (`.jpg

• nemo-rl-auto-research · Skill

  Autonomous NeMo-RL research agent workflow for directed hypothesis testing and open-ended discovery. Guides agents through the full experiment lifecycle: u

• nemo-rl-brev-etiquette · Skill

  Brev instance operating guidance for NeMo-RL agents working in /home/ubuntu/RL with limited workspace disk, a larger /ephemeral volume, and optional /home/

• nemo-rl-docs · Skill

  Documentation conventions for NeMo-RL. Covers docs/index.md updates and docstring format. Do NOT use for: bug fixes, test fixes, dependency bumps, refactor

• nemo-rl-session-memory · Skill

  Manage durable working-session memory for coding agents. Use when a user asks to preserve or recover agent context across disconnects, VS Code restarts, lo

• nemoclaw-user-guide · Skill

  Guides human users' AI agents to the NemoClaw docs MCP server and canonical Fern documentation in Markdown form. Use when users ask how to install, configu

• nemotron-asr-finetune · Skill

  Orchestration skill for NVIDIA Nemotron Speech (Riva) / NeMo ASR domain and language adaptation. Given a goal like "improve/fine-tune ASR for my domain or

• nemotron-customize · Skill

  Plan, configure, and chain repo-native Nemotron customization steps into single-step or multi-step pipelines: curation, translation, SFT/PEFT (AutoModel or

• nemotron-policy-generator · Skill

  Generates BYO custom safety policies for NVIDIA Nemotron content-safety guardrails — Nemotron-Content-Safety-Reasoning-4B (text) and multimodal Nemotron-3-

• nemotron-retrieval-recipes · Skill

  Use when planning, debugging, tuning, evaluating, exporting, or deploying public Nemotron `embed`/`rerank` retrieval recipes.

• nemotron-speech · Skill

  Routes NVIDIA Nemotron Speech (Riva) NIM tasks — deploys, runs, and tests ASR, TTS, and NMT NIMs on build.nvidia.com or self-hosted.

• nv-generate-ct-rflow · Skill

  Used for generating synthetic CT volumes and masks with NV-Generate-CTMR rflow-ct. Not for production training data without review.

• nv-generate-mr · Skill

  Used for generating synthetic body MRI volumes with NV-Generate-CTMR rflow-mr. Not for paired masks or production training data.

• nv-generate-mr-brain · Skill

  Used for generating synthetic T1, T2, FLAIR, SWI, or MRA brain MRI volumes with NV-Generate-CTMR MR-Brain v1. Not for production training data.

• nv-generate-mr-brain-finetune · Skill

  Used 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.

• nv-generate-vae-finetune · Skill

  Used for finetuning the NV-Generate-CTMR MAISI VAE from CT/MRI NIfTI datalists. Not for clinical or production data approval.

• nv-reason-cxr · Skill

  Used for command-shape or live NV-Reason-CXR chest X-ray reasoning smoke tests. Not for diagnosis or clinical reporting.

• nv-segment-ct · Skill

  Used for running NV-Segment-CT VISTA3D on CT NIfTI volumes and recording label-map evidence.

• nv-segment-ct-finetune · Skill

  Runs standard or fixed-channel softmax finetuning of NV-Segment-CT VISTA3D on CT NIfTI image/label datasets, with optional MONAI-native MLflow tracking and

• nv-segment-ctmr · Skill

  Used for running NV-Segment-CTMR on CT or MRI NIfTI volumes and recording label-map evidence. Not for clinical interpretation.

• nvidia-skill-finder · Skill

  Use 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,

• nvidia-skill-finder · Skill

  Use 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,

• nvidia-skills · Collection
• omniverse-cad-to-simready · Skill

  Coordinate 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 p

• omniverse-realtime-viewer · Skill

  Use as the top-level router for Omniverse Realtime Viewer USD app requests and focused viewer reference documents.

• omniverse-usd-performance-tuning · Skill

  Top-level workflow skill for USD performance diagnosis and optimization. Handles slow loading, high memory, low FPS, and broad scene-optimization requests;

• paidf-anomalygen · Skill

  Full PAIDF AnomalyGen pipeline — fine-tune on a new anomaly dataset, generate synthetic anomaly images (SDG), evaluate quality (nn_score), and search per-s

• paidf-augmentation · Skill

  Use when authoring or validating PAIDF augmentation YAML configs, or running remote Cosmos Transfer/Predict, image-edit, or image-to-video inference.

• paidf-auto-labeling · Skill

  Use when a user needs to get started with PAIDF Auto-Labeling, plan a scenario, run or debug a shipped cookbook, author prompts or cookbooks, migrate a pip

• paidf-curation-and-retrieval · Skill

  Use when operating PAIDF Curation and Retrieval or NVIDIA Cosmos Curator pipelines (split, filter, caption, embed, dedup, shard, image annotate) or PAIDF D

• paidf-orchestration-setup · Skill

  Audit, prepare, and deploy PAIDF Orchestration on a Kubernetes GPU cluster - single-GPU H100/L40S hosts, managed Kubernetes, kubeadm, and similar. Select f

• paidf-orchestration-write-dag · Skill

  Use when a user describes a custom PAIDF Orchestration pipeline — a specific ordered combination of stages such as augmentation only, auto-labeling only, d

• physical-ai-defect-image-generation · Skill

  Use when the user wants to orchestrate defect image generation with NVIDIA Cosmos AnomalyGen (Cosmos-Predict2-derived) on OSMO for PCBA, metal surface, and

• physical-ai-event-video-generation · Skill

  Run the PAIDF Orchestration Event Video Generation DAG on Kubernetes - image-to-video anomaly generation, auto-labeling, and anomaly dataset generation. Se

• physical-ai-image-attribute-augmentation · Skill

  Run the PAIDF Orchestration Image Attribute Augmentation DAG on Kubernetes - person-crop clothing augmentation, attribute search, and augmented dataset gen

• physical-ai-infrastructure-setup-and-resilient-scaling · Skill

  Use when the user wants to set up, scale, validate, or harden NVIDIA physical AI infrastructure for synthetic data generation workflows across local MicroK

• physical-ai-neural-reconstruction · Skill

  Router for NVIDIA NuRec/NRE: USDZ rendering, NCore conversion, 3DGS, gRPC sensor sim, carline adaptation, PhysicalAI HF datasets. Do NOT use for SimReady o

• physical-ai-video-data-augmentation · Skill

  Use when running video data augmentation and auto-labeling workflows on OSMO: flow selection, preflight, submit-time interpolation, monitoring, and output

• physicsnemo-discover · Skill

  Official NVIDIA-authored guidance for navigating PhysicsNeMo — pick the model, datapipe, or example for a SciML/AI4Science task (surrogates, forecasting, d

• physicsnemo-shard-tensor · Skill

  Official NVIDIA-authored guidance for PhysicsNeMo ShardTensor domain parallelism — integrate domain parallelism into training/inference scripts (new or exi

• plugins · Collection
• portfolio-optimization · Skill

  Use when a user asks to build, optimize, backtest, rebalance, or analyze a stock portfolio with Mean-CVaR, Mean-Variance/SOCP variance caps, efficient fron

• rag-blueprint · Skill

  NVIDIA RAG Blueprint — deploy, configure, troubleshoot, and manage. Handles any RAG action: deploy, install, start, enable, disable, toggle, change, config

• rag-eval · Skill

  Filesystem RAG benchmarks: corpus/, train.json, evaluate_rag.py (RAGAS quality). Not for prod monitoring, latency/throughput benchmarking (use rag-perf), o

• rag-perf · Skill

  Performance benchmarking for a deployed NVIDIA RAG Blueprint server: profiling pass + aiperf load test driven by a single YAML config. Not for accuracy / R

• rtvi-cv-customize-model · Skill

  How to swap the DeepStream CV detection model in the VSS Alerts Blueprint verification (2d_cv) mode - covers ONNX export, custom bbox parsers, compose moun

• rtvi-cv-scaffold-vss-service · Skill

  Scaffold a standalone RTVI CV microservice that plugs into VSS Search and Alerts profiles via Kafka mdx-raw. The shipped scaffold script is a YOLO26 refere

• rtvi-vlm-customize-model · Skill

  How to swap the VLM in the VSS Alerts Blueprint — covers RTVI-VLM microservice deployment methods, all three VLM consumers (rtvi-vlm, vlm-as-verifier, vss-

• skill-card-generator · Skill

  Use only to generate or update a governance skill card for a specified existing agent skill directory. Do not use for explaining, listing, comparing, or di

• skills · Collection
• skills · Collection
• skills · Collection
• tao-analyze-changenet-rca · Skill

  Performs deep Root Cause Analysis (RCA) on NVIDIA TAO Visual ChangeNet classification experiments with

• tao-analyze-gaps-visual-changenet · Skill

  Performs gap analysis on NVIDIA TAO VCN Classify (Visual Component Net) experiments by invoking the pinned TAO data-services container directly via `docker

• tao-analyze-gaps-vlm-bcq · Skill

  Extract false-positive and false-negative gaps from VLM binary-classification-question (BCQ, yes/no) predictions.

• tao-convert-dataset-format · Skill

  Run `tao-daft convert` to convert NVIDIA TAO DAFT datasets between supported formats. Do not use for non-DAFT data.

• tao-finetune-clip · Skill

  CLIP vision-language model for image-text retrieval, zero-shot classification, embedding extraction, ONNX

• tao-finetune-cosmos-embed · Skill

  Cosmos-Embed1 video-text embedding for text-to-video retrieval, video-to-video search, semantic deduplication, and fine-tuning. Use when the user asks to "

• tao-finetune-cosmos-reason · Skill

  Cosmos3-Nano video QA supervised fine-tuning with FSDP parallelism across variable-width GPU configurations. Use when training or evaluating video

• tao-finetune-huggingface-model · Skill

  Fine-tune any HuggingFace CV / VLM / LLM model on local NVIDIA GPUs inside an NGC PyTorch container when no dedicated TAO model skill matches. Use when the

• tao-generate-image-grounding · Skill

  Two-step image grounding pipeline: extracts referring expressions from (image, caption) pairs and grounds them

• tao-generate-referring-expressions · Skill

  Four-step image referring-expression pipeline: turns images plus KITTI bounding-box labels into region

• tao-generate-video-reasoning-annotations · Skill

  Multi-step video annotation pipeline that turns raw videos into Chain-of-Thought training data — multi-level captions, structured descriptions, and QA pair

• tao-launch-workflow · Skill

  Shared launch intake for any TAO workflow or action. Use when the user wants to run TAO AutoML, train, evaluate, infer, export, generate TensorRT engines,

• tao-list-capabilities · Skill

  Answer what the TAO Skill Bank plugin can do by generating the response from packaged application, data, model, AutoML, and platform manifests. Use when th

• tao-mine-aoi-images · Skill

  Runs the DEFT embed-then-mine workflow for VCN AOI iterations — embeds the gap-analysis target parquet, embeds a source pool, and mines nearest-neighbour s

• tao-port-huggingface-model · Skill

  Integrate a HuggingFace Computer Vision model into the NVIDIA TAO Toolkit ecosystem (tao-core config, tao-pytorch trainer, tao-deploy TensorRT pipeline). U

• tao-route-visual-changenet-samples · Skill

  Routes the weakest VCN samples (output of `tao-analyze-gaps-visual-changenet`) into per-augmentation-module

• tao-run-automl · Skill

  Run container-backed AutoML / hyperparameter optimization (HPO) for NVIDIA TAO networks using AutoMLRunner. Handles algorithm

• tao-run-automl-deft-pipeline · Skill

  Run the canonical NVIDIA AOI three-phase training pipeline — Phase 1 AutoML baseline (HPO), Phase 2 DEFT loop (RCA → SDG → mining → plain-train retrain), P

• tao-run-deft-aoi · Skill

  Run the full DEFT AOI improvement loop for NVIDIA TAO VisualChangeNet / ChangeNet PCB inspection models: baseline evaluate, RCA, Cosmos AnomalyGen / AMP sy

• tao-run-inference-service · Skill

  Start, query, and stop a network-specific TAO inference microservice ({network_arch}-inference-microservice) by delegating container execution to the appro

• tao-run-on-brev · Skill

  Brev managed GPU instances with Docker support. Use when running TAO training, evaluation, or inference on

• tao-run-on-docker · Skill

  Docker conventions for running NVIDIA GPU container workloads — NGC authentication, --gpus flag, mount patterns,

• tao-run-on-kubernetes · Skill

  Kubernetes execution platform — submits TAO container jobs as single-pod k8s Jobs with NVIDIA GPU scheduling.

• tao-run-on-local-docker · Skill

  Local or remote Docker execution for TAO SDK job containers using a Docker daemon with NVIDIA GPU runtime. Use

• tao-run-on-slurm · Skill

  Remote SLURM GPU cluster execution over SSH with sbatch/srun, Pyxis/Enroot containers, and Lustre-backed

• tao-run-platform · Skill

  TAO Execution SDK for submitting and monitoring GPU training jobs on supported platforms (Brev, SLURM,

• tao-setup-nvidia-gpu-host · Skill

  Host setup for TAO GPU backends. Checks and, after user approval, installs minimum-compatible NVIDIA driver, CUDA Toolkit, and NVIDIA Container Toolkit ver

• tao-train-action-recognition · Skill

  Action recognition from video sequences. Supports RGB, optical flow, and joint (multi-stream) input types for

• tao-train-bevfusion · Skill

  BEVFusion for multi-sensor 3D object detection. Fuses LiDAR point clouds and camera images in bird's-eye-view

• tao-train-centerpose · Skill

  CenterPose for keypoint / pose estimation. Detects object centers and regresses keypoint locations for 6-DoF

• tao-train-deformable-detr · Skill

  Deformable DETR for 2D object detection. Uses deformable attention for efficient multi-scale feature processing,

• tao-train-depth-anything-v2 · Skill

  Monocular depth estimation using Metric Depth Anything v2 or Relative Depth Anything architectures. Predicts

• tao-train-dino · Skill

  DINO (DETR with Improved DeNoising Anchor Boxes) for 2D object detection. Transformer-based detector with

• tao-train-fast-foundation-stereo · Skill

  Real-time stereo depth estimation using FastFoundationStereo (FFS), the distilled bp2 commercial variant of

• tao-train-foundation-stereo · Skill

  Stereo depth estimation using FoundationStereo. Predicts disparity maps from stereo image pairs for 3D

• tao-train-grounding-dino · Skill

  Grounding DINO for open-set object detection. Combines DINO-style detection with a BERT text encoder for

• tao-train-image-classification · Skill

  PyTorch-based TAO image classification. Supports a wide range of backbones (FAN, EfficientNet, ResNet, etc.)

• tao-train-mask-auto-encoder · Skill

  Masked Auto-Encoder (MAE) for self-supervised pretraining and fine-tuning. Masks random patches and reconstructs

• tao-train-mask-auto-label · Skill

  MAL (Mask Auto-Label) for weakly-supervised segmentation. Produces segmentation masks from minimal annotations

• tao-train-mask-grounding-dino · Skill

  Mask Grounding DINO for grounded instance segmentation. Extends Grounding DINO with a mask-prediction head for

• tao-train-mask2former · Skill

  Mask2Former for universal image segmentation (panoptic, instance, and semantic). Transformer-based with

• tao-train-metric-learning-recognition · Skill

  Metric-learning recognition (ml-recog) for fine-grained visual recognition. Learns embeddings for

• tao-train-nvdinov2 · Skill

  NVDINOv2 for self-supervised visual representation learning. Trains vision transformers via self-distillation

• tao-train-nvpanoptix3d · Skill

  NVPanoptix3D for panoptic 3D scene reconstruction from posed RGB images. Produces 3D panoptic segmentation

• tao-train-ocdnet · Skill

  OCDNet for scene text detection. Detects arbitrary-oriented text regions in natural images using a

• tao-train-ocrnet · Skill

  OCRNet for scene text recognition. Recognizes text content from cropped text-region images and supports CTC

• tao-train-oneformer · Skill

  OneFormer for universal image segmentation. Unifies panoptic, instance, and semantic segmentation with a

• tao-train-optical-inspection · Skill

  Optical Inspection for defect detection using Siamese networks. Compares image pairs to detect manufacturing

• tao-train-pointpillars · Skill

  PointPillars for 3D object detection from LiDAR point clouds. Encodes point clouds into a pseudo-image via a

• tao-train-pose-classification · Skill

  Pose classification using ST-GCN (Spatial Temporal Graph Convolutional Network). Classifies skeleton sequences

• tao-train-reid · Skill

  Person re-identification (ReID). Learns discriminative embeddings to match the same person across different

• tao-train-rtdetr · Skill

  RT-DETR (Real-Time DEtection TRansformer) for 2D object detection. Designed for real-time inference with

• tao-train-segformer · Skill

  SegFormer for semantic segmentation. Lightweight transformer-based architecture with hierarchical feature

• tao-train-single-step · Skill

  Standard single-step train/eval/export workflow for any TAO model. Use when training a TAO model on a dataset

• tao-train-sparse4d · Skill

  Sparse4D for multi-camera temporal 3D object detection and tracking. Uses sparse queries with deformable

• tao-train-visual-changenet · Skill

  Visual ChangeNet for binary image classification and segmentation in AOI defect detection. Use when training,

• tao-validate-dataset-format · Skill

  Run `tao-daft validate` to check NVIDIA TAO DAFT datasets for structure, schema, and cross-reference errors. Do

• tilegym-adding-cutile-kernel · Skill

  Add a new cuTile GPU kernel operator to TileGym. Covers dispatch registration in ops.py, cuTile backend implementation, __init__.py exports, test creation,

• tilegym-converting-cutile-to-julia · Skill

  Converts cuTile Python GPU kernels (@ct.kernel) to cuTile.jl Julia equivalents. Handles kernel syntax translation, 0-indexed to 1-indexed conversion, broad

• tilegym-converting-cutile-to-triton · Skill

  Converts cuTile GPU kernels (@ct.kernel) to Triton (@triton.jit). Handles standard in-repo conversion, debugging (cudaErrorIllegalAddress, shape mismatch,

• tilegym-cutile-autotuning · Skill

  Use when adding, modifying, optimizing, or debugging CuTile autotuning code. Trigger signals: `exhaustive_search` / `replace_hints` / `hints_fn` / `cuda.ti

• tilegym-cutile-python · Skill

  Expert cuTile programming assistant. Write high-performance GPU kernels using cuTile's tile-based programming model with proper validation and optimization

• tilegym-improve-cutile-kernel-perf · Skill

  Iteratively optimize cuTile kernel performance through systematic profiling, bottleneck analysis, IR comparison, and targeted tuning. Covers tile sizes, oc

• tilegym-monkey-patch-kernels-to-transformers · Skill

  Integrate TileGym kernels into Hugging Face `transformers` models by replacing the library's submodule(s) and certain class(es)' implementations, and patch

• vss-ask-video · Skill

  Use 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 me

• vss-deploy-dense-captioning · Skill

  Use this skill when deploying standalone RT-VLM dense captioning or calling its REST API (uploads, captions, streams, chat-completions, Kafka). Not for VSS

• vss-deploy-detection-tracking-2d · Skill

  Use 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

• vss-deploy-detection-tracking-3d · Skill

  Deploy and operate the RTVI-CV-3D microservice as MV3DT (`MODE=mv3dt`): per-camera DeepStream perception plus BEV Fusion over calibrated cameras. Supports

• vss-deploy-profile · Skill

  Use to select, configure, deploy, verify, debug, or tear down a VSS profile (base, search, lvs, warehouse, edge). Not for standalone microservices — use th

• vss-deploy-video-embedding · Skill

  Use this skill when deploying, operating, or integrating the VSS 3.2 GA RT-Embed Video Embedding microservice. Covers Docker Compose bring-up, GPU and stor

• vss-generate-video-calibration · Skill

  Use 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 calibr

• vss-generate-video-report · Skill

  Use this skill when producing a VSS analysis report — Mode A per-clip VLM, Mode B incident-range via video-analytics. Not for standalone video summarizatio

• vss-manage-alerts · Skill

  Use for VSS alert workflows — real-time monitoring, Alert-Bridge subscriptions, Slack notifications, incident queries, camera onboarding. Not for non-alert

• vss-manage-video-io-storage · Skill

  Use to call the VIOS REST API (sensor list, timelines, clip extraction, snapshots, add/delete sensors and streams). Not for VLM inference or search.

• vss-query-analytics · Skill

  Use 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-ran

• vss-search-archive · Skill

  Use 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 (

• vss-setup-behavior-analytics · Skill

  Use to deploy the vss-behavior-analytics service standalone (entrypoint, config-source, optional calibration). Not for the full warehouse deploy.

• vss-setup-video-analytics-api · Skill

  Use to deploy the vss-video-analytics-api REST service standalone (config-source, data-log bind, Elasticsearch, optional Kafka). Not for full warehouse dep

• vss-summarize-video · Skill

  Use to summarize a recorded video via the LVS summarization microservice (HITL-gated) with a VLM fallback. Not for report generation or live RTSP captionin

• warp-compile-time-optimizer · Skill

  Use when compile time or startup time is the problem in code that uses Warp: a request to improve, optimize, or cut compile times; an app that is slow to s

• warp-debug-gradients · Skill

  Use to diagnose and fix incorrect gradients in differentiable Warp programs. Anything trained, optimized, calibrated, or fit through Warp kernels depends o

• warp-eval · Skill

  Evaluate whether an existing hot path is a credible NVIDIA Warp candidate. Use for irregular or spatial queries, particle or geometry simulation, branch-he

Browse the @skills marketplace