skills.rest
NVIDIA
NVIDIAOFFICIAL
Explore all skills by this author →
WebsiteGitHub

jetson-video-pipeline

Execute and verify Jetson Video Codec SDK and PyNvVideoCodec encode, decode, and transcode pipelines with hashed artifact handoffs.

3.2k|370|Updated Feb 25, 2026
Search Tags:#artifact-verification#jetson#transcode#video-codec#NVENC#nvdec#pynvvideocodec
One-click install
npx skills add https://github.com/NVIDIA/skills --skill jetson-video-pipeline
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: jetson-video-pipeline
Source: https://github.com/NVIDIA/skills/tree/main/skills/jetson-video-pipeline
Command: npx skills add https://github.com/NVIDIA/skills --skill jetson-video-pipeline

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Running video codec operations on NVIDIA Jetson devices often fails silently: outputs are stale, stages consume the wrong files, or completion markers are ambiguous. This Skill executes official Video Codec SDK and PyNvVideoCodec sample stages and cryptographically proves that every consumer used the exact artifact produced by the preceding stage.

Core Features & Use Cases

  • Verified Codec Pipelines: Run encode-then-decode verification, native H.264-to-HEVC transcode, PyNvVideoCodec segmentation, container decode triage, and AV1 operation verification through authenticated NVIDIA sample executables.
  • Exact Artifact Handoffs: Every stage output is hashed (SHA-256) and re-verified before the next stage consumes it, with strict frame-count and completion-marker validation.
  • Acceptance Packages: Produce a compact customer acceptance package with sealed evidence files and a checksum manifest.
  • Use Case: Ask your agent to verify that a Jetson device can encode raw NV12 video to HEVC and decode it back; the Skill authenticates the installed SDK samples, runs the pipeline in an isolated workspace, and returns a structured JSON result proving each handoff.

Quick Start

Ask your agent to run an encode-then-decode verification on your Jetson using the jetson-video-pipeline skill with a local raw video file.

Frequently Asked Questions about jetson-video-pipeline

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify video encode and decode on a Jetson device?▼

Use the encode controller to run one recipe-bound encode followed by an independent decode through authenticated NVIDIA samples. The Skill requires exact frame-count markers and rehashes the encoded bitstream before the decoder consumes it, proving the handoff.

What is the difference between the native and PyNvVideoCodec surfaces?▼

The native surface uses compiled Video Codec SDK samples like AppEncCuda, AppDec, and AppTrans from the installed dpkg package. The PyNvVideoCodec surface runs Python samples such as encode.py and decode.py from a validated PyNvVideoCodec 2.1.0 wheel environment.

Does this skill support AV1 encoding on Jetson?▼

Yes, it includes an AV1 verification route that validates IVF container structure, AV01 fourcc, dimensions, and exact frame counts. AV1 output must be a well-formed IVF stream matching the recipe's width, height, and frame count.

Can this skill measure PSNR or SSIM video quality?▼

No, the Skill explicitly does not provide objective quality metrics such as PSNR or SSIM. It covers codec execution and artifact verification only; quality measurement requires a separately authorized workflow.

Why does the pipeline return input_required before running?▼

The media gate requires one exact user-supplied local path or HTTP(S) URL before any probing or execution. The Skill never substitutes catalog or synthetic media, so it stops and asks for the missing input first.

What dependencies does the Jetson video pipeline need?▼

It requires the nvidia-video-codec-sdk dpkg package for native routes or a validated PyNvVideoCodec 2.1.0 environment for Python routes, plus a CUDA 13.x toolkit. Recipe-bearing routes also need the sibling jetson-video-recipe skill installed.

Related Skills

Explore More Software Engineering Skills→
GeneralReasoningGeneralReasoning
4

FFmpeg Format Conversion

Convert media files between containers and codecs using FFmpeg.

Official
Intermediate
drpedapatidrpedapati
88

benchmark-logging

Log and compare benchmark runs with metrics and artifact references.

Community
Intermediate
ckorhonenckorhonen
12

video-editor

Guide FFmpeg-based video editing, encoding, and processing workflows.

Community
Advanced
wgpsecwgpsec
45

Multi-Cloud Deployment

Coordinate multi-cloud infrastructure deployments across AWS, Azure, GCP, Alibaba Cloud, Tencent Cloud, Huawei Cloud, and Volcengine.

Official
Advanced