Dev & Engineering ✓ NVIDIA · Official holoscan-sensor-bridgehardware-devkitsensor-applicationsssh-automationdocker-container

Holoscan Sensor Bridge App Runner

Find, run, and debug HSB examples on a connected development kit.

FollowSkills review · FSRS-2.0
Not recommended
45/ 100 5-point scale 2.3 / 5
Trust10 / 25 · 2.0/5

The skill requires explicit invocation, phase confirmations, confirmation before code edits, and describes SSH, Docker, device mappings, and cleanup. However, it permits Bash, configurable sudo, privileged containers, host networking, /dev mounts, xhost, and remote writes; it also prints resolved environment variables. Least privilege, sensitive-value protection, complete data-flow disclosure, and robust rollback are incomplete, so points are deducted.

Reliability8 / 20 · 2.0/5

The phased workflow, watchdog, log summaries, failure taxonomy, re-verification, and container cleanup are useful. However, this is a static review with no committed tests covering key paths, and heredoc handling, variable interpolation, remote paths, and container prerequisites remain unverified assumptions. The score stays within the static-review ceiling.

Adaptability8 / 15 · 2.7/5

The audience, hardware variants, sensor filtering, explicit invocation, and platform constraints are clearly described. Chinese interaction and mainland-China reachability are not documented, and the workflow depends on a preconfigured remote devkit, SSH, Docker, GPU runtime, and physical hardware; boundary evidence is therefore limited.

Convention9 / 15 · 3.0/5

The skill has frontmatter, an Apache-2.0 declaration, version 1.0.0, author information, linked phase documentation, help text, and invocation examples. It lacks dedicated Instructions, Examples, and Purpose sections; the benchmark also identifies missing prerequisite documentation. There is no changelog, and maintenance ownership and update path are only indirectly supported by repository-level material.

Effectiveness6 / 15 · 2.0/5

The workflow covers discovery, filtering, execution, failure analysis, iteration, and reporting, and the attached benchmark reports practical evaluation results. Static review cannot verify actual outputs, hardware compatibility, or repair success, and results still depend on the user's environment and choices, so the score remains within the static ceiling.

Verifiability4 / 10 · 2.0/5

The fixed revision, concrete commands, evaluation tasks, benchmark metrics, and phase procedures provide some auditability. However, there is no verifiable CI or committed key-path test suite, and the benchmark is an attached report rather than independently corroborated execution evidence; cross-source verification is limited.

Evidence confidence:Low Reviewed Jul 20, 2026 Reviewed revision 55f18499943e
Before you use it
  • Execution involves remote SSH, privileged Docker, host networking, /dev mounts, GPU access, and display permissions; manually verify the target host, image, devices, and command scope before running.
  • Resolved environment values may contain sensitive SSH options or path information and should not be displayed or copied into reports without filtering.
  • Container stopping and session teardown use broad name filters and remote rm -rf; confirm that targets are limited to resources created by this workflow and retain an external recovery path.
  • Core functionality depends on a configured NVIDIA/HSB devkit, physical hardware, Docker runtime, and local networking; Chinese-language interaction and mainland-China availability are not demonstrated.
Review evidence [1][2][3][4][5][6][7][8]
See the full review method →

What does this skill do, and when should you use it?

hsb-app is for a Holoscan Sensor Bridge development kit that has already been configured. It discovers example applications and filters them by platform, HSB software version, board type, and connected sensors. Selected applications run inside the demo Docker container, with timed execution, failure analysis, edit suggestions, and iterative reruns. The workflow requires explicit invocation and begins by verifying the remote kit, board, FPGA identity, and container readiness over SSH.

Reads and displays SSH_TARGET, REMOTE_ROOT, REMOTE_SUDO, REMOTE_SSH_OPTS, and HSB_PLATFORM; uses SSH to check cached session state, kit reachability, HSB board responsiveness, FPGA version, board identity, and demo-container readiness; reads the user guide and examples/ to filter applications for the user's platform, HSB version, board, and sensors; runs the selected application in the Docker demo container, using python3 for Python examples; monitors output, analyzes failures, suggests code or environment edits, and supports reruns; generates a session report and offers to save it.

  1. An HSB developer with a connected board needs to select an example compatible with the current platform and sensors.
  2. A hardware engineer wants to verify remote-kit connectivity, board identity, FPGA version, and demo-container readiness before running an application.
  3. An application developer wants log-based failure analysis, suggested edits, and a retry workflow after an HSB example fails.
  4. A test engineer needs to run an example for a defined number of seconds and receive a report of issues, fixes, and outcome.

What are this skill's strengths and limitations?

Pros
  • Covers remote verification, hardware identification, application selection, execution, debugging, cleanup, and reporting.
  • Filters examples by platform, HSB version, board type, and sensors before selection.
  • Supports watchdog timeouts, failure summaries, edit suggestions, and iterative reruns.
  • Its metadata names both Claude Code and Codex and uses broadly available shell, filesystem, and text-search tools.
Limitations
  • Requires a preconfigured devkit, connected HSB board, SSH access, Docker demo container, and host setup.
  • Depends on remote SSH, Docker, and specific hardware, so it is not a standalone local example runner.
  • It uses Claude Code-oriented invocation controls, slash-command arguments, and session-state paths; other clients need adaptation.
  • The supplied material provides no test-coverage or standalone benchmark evidence for this skill.

How do you install this skill?

Install only this skill through the Skills CLI documented in the repository README:

npx skills add nvidia/skills --skill hsb-app --yes

To target Codex explicitly:

npx skills add nvidia/skills --skill hsb-app --agent codex

The supplied material does not specify a manual-copy destination. The skill declares Apache-2.0. The repository README states that the repository is dual-licensed under Apache 2.0 and CC BY 4.0, while the supplied repository metadata says NOASSERTION.

How do you use this skill?

Prepare the devkit with SSH, the demo container, host configuration, and a connected HSB board. Set SSH_TARGET and REMOTE_ROOT; if setup is incomplete, run /hsb-setup first. Then invoke the skill explicitly:

/hsb-app

Supported options include /hsb-app --timeout 60, /hsb-app --verbose, /hsb-app --full, and /hsb-app --help. The first run requires confirmation of Phase 0, and the normal workflow asks for confirmation between phases. If SSH_TARGET or REMOTE_ROOT is missing, the skill asks the user to provide it.

FAQ

Can it run without a connected HSB board?
Not for its intended workflow. It is designed for a connected devkit and verifies board responsiveness, FPGA version, board identity, and demo-container readiness.
What access and setup are required?
You need SSH access, a remote working directory, and an appropriate sudo setting where required. The devkit, demo container, host, and board should already be configured.
What happens when an application fails?
The skill summarizes the failed command, likely cause, recommended safe action, and blocking status, then can suggest code or environment edits and rerun the application.
Does it run automatically when installed?
No. The SKILL.md requires explicit invocation, and the initial workflow also requires confirmation of the phase plan and subsequent phases.

More skills from this repository

All from NVIDIA/skills

Automation & Ops ✓ NVIDIA · Official

Holoscan Sensor Bridge Demo Bring-Up

Build and verify a Holoscan Sensor Bridge demo environment on NVIDIA devkits.

Dev & Engineering ✓ NVIDIA · Official

HSB QA Test Runner

Filter, execute, and report automatable QA tests on Holoscan Sensor Bridge development hardware.

Automation & Ops ✓ NVIDIA · Official

HSB FPGA Flash Assistant

Safely flash FPGA firmware on supported HSB boards and VB1940 cameras.

Automation & Ops ✓ NVIDIA · Official

Holoscan NGC Container Installer

Install, validate, and launch NVIDIA Holoscan SDK through an NGC Docker container.

Dev & Engineering ✓ NVIDIA · Official

Holoscan SDK Setup Guide

Inspects a Linux host and selects the most suitable Holoscan SDK installation path.

Automation & Ops ✓ NVIDIA · Official

TAO NVIDIA GPU Host Setup

Checks and standardizes NVIDIA drivers, CUDA, and container runtime prerequisites for TAO GPU hosts.

Data & Analysis ✓ NVIDIA · Official

VSS Video Analytics Query

Read video-analytics incidents, metrics, alerts, and sensor data through VA-MCP.

Automation & Ops ✓ NVIDIA · Official

VIOS Video I/O and Storage Manager

Manage VIOS sensors, streams, recordings, snapshots, clips, and storage through its REST API.

Automation & Ops ✓ NVIDIA · Official

VSS Alert Management

Operate NVIDIA VSS real-time alerts, subscriptions, incident queries, Slack notifications, and camera onboarding.

Dev & Engineering ✓ NVIDIA · Official

Holoscan SDK Source Builder

Build Holoscan SDK from source in a Docker container and produce a local CMake-consumable installation tree.

Automation & Ops ✓ NVIDIA · Official

TAO Inference Microservice Deployment

Start, query, and stop TAO inference microservices across supported compute platforms.

Automation & Ops ✓ NVIDIA · Official

NVIDIA GPU Docker Runtime Guide

Reusable Docker conventions for authenticating with NGC, exposing NVIDIA GPUs, mounting data, inspecting containers, and diagnosing GPU workload failures.

Dev & Engineering ✓ NVIDIA · Official

NeMo MBridge GPU Memory Tuning

Diagnose GPU OOMs in Megatron Bridge training and reduce fragmentation, activation, and PEFT memory usage.

Dev & Engineering ✓ NVIDIA · Official

DOCA Hardware Telemetry Counter Reader

Guides developers in reading per-domain hardware counters from BlueField and ConnectX devices.

Automation & Ops ✓ NVIDIA · Official

TAO Workflow Launch Gate

Preflight TAO jobs across platforms, credentials, data, containers, and monitoring.

Dev & Engineering ✓ NVIDIA · Official

Megatron Bridge Activation Recompute

Trade extra compute for lower GPU activation memory in Megatron Bridge training.

Dev & Engineering ✓ NVIDIA · Official

DOCA Compress Hardware Offload

Guides hands-on DEFLATE and LZ4 decompression development and debugging on BlueField and ConnectX platforms.

Dev & Engineering ✓ NVIDIA · Official

DOCA Debug

A cross-library diagnostic guide for NVIDIA DOCA build, link, runtime, and program failures.

Dev & Engineering ✓ NVIDIA · Official

NeMo MBridge Parallelism Strategy Guide

Choose, combine, and validate Megatron Bridge parallelism configurations.

Automation & Ops ✓ NVIDIA · Official

NVIDIA AI-Q Deployment

Deploy, verify, and operate NVIDIA AI-Q Blueprint infrastructure.

Related skills