Used for finetuning NV-Generate-CTMR MR-brain diffusion UNet from a NIfTI datalist. Not for clinical or production data approval.
Two-step image grounding pipeline: extracts referring expressions from (image, caption) pairs and grounds them
Install Holoscan SDK natively on Ubuntu via apt. Use for C++ installs on Ubuntu; pair with /holoscan-install-wheel for Python.
Call the vss agent to run video understanding on video to answer a text question. Use when the user asks about video content, or about visual details that cannot be answered from…
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 carrier derivation.
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.
SECURITY: the DOCA UROM Service binary has NO standalone access control (no authz list / `allowed_host` / `allowed_users` / `auth_token`) — access is governed ENTIRELY by DOCA…
Official NVIDIA-authored guidance for PhysicsNeMo ShardTensor domain parallelism — integrate domain parallelism into training/inference scripts (new or existing) with DDP or…
Route open-ended or ambiguous NVFLARE requests by inspecting the local project and recommending one specific workflow skill without editing files; do not use for an explicit…
Given a GitHub issue number, plan and implement the work described in the issue. Supports direct user requests and unattended queue processing through the `agent:*` workflow…
Code instrumentation for timing workloads. Two scenarios: (1) Training loop — inject manual timing to report per-iteration latency, throughput (samples/sec), and data load time.
Enable and interpret TensorRT-LLM AutoDeploy FX graph text dumps via AD_DUMP_GRAPHS_DIR. Use when you need before/after graphs per transform, to locate subgraphs, or to confirm a…
Sparse4D for multi-camera temporal 3D object detection and tracking. Uses sparse queries with deformable
ONLY for OpenAI Triton (@triton.jit) kernel development. NEVER use for CUDA C++ kernels, TileIR, or profiling tools (ncu, nsys). The user's request must involve Triton explicitly.
Runs the daytime maintainer loop for NemoClaw, prioritizing items labeled with the current version target.
Guide for adding a new sink to PhysicsNeMo Curator. Covers discovery questions, implementation patterns (simple writer, append-based, split-based), output naming, parallel…
Systematic approach to exploring the TensorRT-LLM codebase before implementing new features or optimizations.
Guide for adding a new filter to PhysicsNeMo Curator. Covers discovery questions, implementation patterns (pass-through, stateful, in-place), generator semantics, flush/artifacts,…
Multi-step video annotation pipeline that turns raw videos into Chain-of-Thought training data — multi-level captions, structured descriptions, and QA pairs (MCQ, binary,…
Analyze host/CPU overhead in TensorRT-LLM inference from nsys traces. Detect whether host overhead is the bottleneck using GPU idle ratio, host prep exposed ratio, and per-phase…
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 recipe-runner for recipe…
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).
Top-level workflow skill for USD performance diagnosis and optimization. Use for slow loading, high memory, low FPS, or 'optimize my scene' requests; delegates auth/runtime setup…
Deploy and operate the RTVI-CV-3D stack (also known as MV3DT, Multi-View 3D Tracking, or RTVI-CV-MV3DT) — per-camera DeepStream perception plus BEV Fusion over multiple calibrated…
Used for extracting selected metadata from one DICOM file and flagging standard-tag PHI presence. Not for anonymization or clinical use.
Validate and use selective and full activation recompute in Megatron Bridge to reduce GPU memory usage at the cost of extra compute. — from NVIDIA/skills
Validate and use CUDA graph capture in Megatron Bridge, including local full-iteration graphs and Transformer Engine scoped graphs for attention, MLP, and MoE modules.
Build Holoscan SDK from source via the in-tree ./run script. Use only when published packages don't meet the user's needs.
Query and browse evaluation results stored in MLflow. Use when the user wants to look up runs by invocation ID, compare metrics across models, fetch artifacts (configs, logs,…
Long-context MoE training guidance for Megatron Bridge. Covers CP sizing, selective recompute, dispatcher choices, and practical patterns from DSV3, Qwen3, and Qwen3-Next — from…
Used for converting one CT DICOM series folder to a HU NIfTI volume with affine evidence. Not for multi-frame DICOM or clinical use.
Configuration conventions for NeMo-RL. YAML is the single source of truth for defaults. Covers TypedDict usage, exemplar YAML updates, and forbidden default patterns.
Guides human users' AI agents to the NemoClaw docs MCP server and canonical Fern documentation in Markdown form.
DINO (DETR with Improved DeNoising Anchor Boxes) for 2D object detection. Transformer-based detector with
Generate sandbox security policies from plain-language requirements and optional REST API documentation.
Audit Warp CHANGELOG.md before a release: recover lost entries, sort by user impact, refine entry language, line-wrap, and (release-branch mode) bump compare refs.
NVPanoptix3D for panoptic 3D scene reconstruction from posed RGB images. Produces 3D panoptic segmentation
Scan recent git commits for changes that affect user-facing behavior, then draft or update the corresponding documentation pages and refresh generated user skills for release…
Converts cuTile GPU kernels (@ct.kernel) to Triton (@triton.jit). Handles standard in-repo conversion, debugging (cudaErrorIllegalAddress, shape mismatch, numerical misma — from…
CenterPose for keypoint / pose estimation. Detects object centers and regresses keypoint locations for 6-DoF
Operational guide for enabling hierarchical context parallelism in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification. — from NVIDIA/skills
Adds, removes, or modifies allowed endpoints in the sandbox policy. Use when customizing network policy, changing egress rules, or configuring sandbox endpoint access.
Fetch weather/climate data via Earth2Studio data sources for specific variables and times. Do NOT use for inference pipelines, model discovery, or installation.
Explains how to run NemoClaw on a remote GPU instance, including the deprecated Brev compatibility path and the preferred installer plus onboard flow.
Upgrade flashinfer-python version in TensorRT-LLM. Fetches the latest releases from GitHub (stable and nightly), compares with the current pinned version, lets the user pick a…
Docker conventions for running NVIDIA GPU container workloads — NGC authentication, --gpus flag, mount patterns,
External verl end-to-end validation workflow for Megatron-Bridge model/provider changes. Covers running a small verl Megatron backend job from a Bridge checkout, choosing LoRA/DDP…
Entry skill for Jetson / IGX BSP customization. Asks one core click-to-select setup questionnaire and passes prefilled answers to downstream setup skills.
Base rules for end users calling NVIDIA cuOpt (routing/LP/MILP/QP/install/server). Not for cuOpt internals — use cuopt-developer for those.
Playbook for launching, monitoring, stopping, and debugging NeMo-RL recipes on a Kubernetes cluster via the nrl-k8s CLI.
OneFormer for universal image segmentation. Unifies panoptic, instance, and semantic segmentation with a
Generate video summary reports using the VSS video_search_frag extension with Long Video Summarization (LVS), Enterprise RAG knowledge retrieval, and human-in-the-loop parameter…
Use to diagnose and fix incorrect gradients in differentiable Warp programs. Anything trained, optimized, calibrated, or fit through Warp kernels depends on wp.Tape gradients, so…
Guide for adding a new data source to PhysicsNeMo Curator. Use when adding a remote dataset (HuggingFace, S3, etc.) or local file-based source.
Mask2Former for universal image segmentation (panoptic, instance, and semantic). Transformer-based with
Practical guidance for training MoE VLMs in Megatron Bridge. Compares FSDP and 3D-parallel approaches, using rounded lessons from Qwen3-VL, Qwen3-Next, and other multimod — from…
GitHub Actions CI guidance for validation gates, workflow permissions, matrices, caching, path filters, artifacts, and workflow verification.
How to swap the DeepStream CV detection model in the VSS Alerts Blueprint verification (2d_cv) mode - covers ONNX export, custom bbox parsers, compose mount gotchas, nvinfer…
Bootstrap a custom carrier board by forking carrier files and scaffolding a DT overlay from the reference devkit.
Use to run AutoMagicCalib on local MP4s, RTSP, or the bundled sample dataset, and to deploy vss-auto-calibration when needed. Not for non-AMC calibration or runtime analytics.