复制安装命令
用 Codex 或 Claude 安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它先审查 Skill 页面再帮你安装。
复制前请先查看来源、License 和安全提示。
SkillAtlasSkill 详情
jetson-customize-uphy
Official, NVIDIA-verified Agent Skills for Claude Code, Codex, and other coding agents.
审核状态：已审核Quality 72Security 82
项目 README
来源文件：README.md
NVIDIA Agent Skills
Official, NVIDIA-verified Agent Skills for Claude Code, Codex, and other coding agents.
📖 Docs: docs.nvidia.com/skills · 📺 Livestream: From Vulnerable to Verified · 📝 Blog: NVIDIA Verified Agent Skills: Capability Governance for AI Agents
Skills are portable instruction sets that teach AI agents how to use NVIDIA software optimally: Physical AI and robotics workflows, simulation, CUDA-X libraries, RAG and AI Blueprints, and platform tools. This repository is a catalog: skills are maintained in their respective product repos, and mirrored here daily via an automated sync pipeline. Skills are being added continuously, so check back for updates. We are building this infrastructure in the open, and contributions are welcome. See the Roadmap for what is planned next.
Quickstart
Install NVIDIA skills with the default skills CLI flow:
npx skills add nvidia/skills
The CLI runs through npx and prompts you to choose a skill and install destination. You do not need to clone this repo or copy skill folders by hand.
Requires a current
skillsCLI (v1.5.16 or newer). Installing vianpx skills@latest add nvidia/skillsalways uses the latest. On older CLIs (v1.5.15 and earlier), skills may install but not appear in Claude Code — see Troubleshooting.
The skill is available the next time your agent loads skills and encounters a relevant task. For example, ask your agent to "solve a linear programming problem with cuOpt" and the skill guides it through the cuOpt Python API. In Claude Code, run /reload-skills to load newly installed skills in your current session.
Install One Skill Without Prompts
Use this when you already know the skill name and want to skip prompts.
npx skills add nvidia/skills --skill cuopt-numerical-optimization-api --yes
Replace cuopt-numerical-optimization-api with any skill name from the Skill Catalog.
Install for a Specific Agent
Use --agent to target a specific AI coding agent. Initially, we'll support common client targets, expanding the list over time. For the full list of clients supported by the spec, see the skills CLI Supported Agents table.
Claude Code
npx skills add nvidia/skills --skill cuopt-numerical-optimization-api --agent claude-code
Codex
npx skills add nvidia/skills --skill cuopt-numerical-optimization-api --agent codex
Snowflake CoCo
npx skills add nvidia/skills --skill cuopt-numerical-optimization-api --agent cortex
Cursor
npx skills add nvidia/skills --skill cuopt-numerical-optimization-api --agent cursor
Kiro
npx skills add nvidia/skills --skill cuopt-numerical-optimization-api --agent kiro-cli
Use --agent more than once to install the same skill into multiple agents.
npx skills add nvidia/skills \
--skill cuopt-numerical-optimization-api \
--agent claude-code \
--agent codex \
--agent cursor \
--agent kiro-cli
Keep Skills Up to Date
New skills land continuously, and existing ones are revised, renamed, or consolidated as the catalog evolves. Refresh what you have installed with:
npx skills update
Run it interactively and the CLI also flags skills that were removed or merged upstream (for example, when several skills are consolidated into one) and offers to remove the stale local copies. Use npx skills list to see what is installed and npx skills check to preview what is out of date first.
Browse the Catalog
Use this when you want to see available NVIDIA skills before installing anything.
npx skills add nvidia/skills --list
For non-interactive installs, global installs, agent-specific installs, updates, removals, and fallback manual copying, see Advanced installation.
Skill Catalog
Getting Help & Contributing
Where to file an issue depends on what's broken:
- Skill content issues (a specific skill has a bug, missing functionality, or incorrect content) — file in the source repo for that product, using the per-product table below.
- Catalog issues (catalog README errors, sync workflow problems, distribution channels, signing/verification flow, docs in this repo) — file here using the catalog issue templates: Bug Report, Feature Request, or Documentation Request or Correction.
- Questions or general discussion — use Discussions. The issue tracker is reserved for bug reports, feature proposals with a design, and documentation issues.
- Security vulnerabilities — follow the disclosure process in SECURITY.md; do not open a public issue.
Per-product source repo links:
For issues with this catalog repo itself (README, structure, listing a new product): open an issue here.
Verifying Skills
Every published skill ships with a detached OMS signature (skill.oms.sig). The sync pipeline drops any skill missing the required artifacts before publishing, so every skill in the catalog carries:
SKILL.md— the skill instructions consumed by the agentskill-card.md— skill identity and governance cardskill.oms.sig— detached OMS signature (verifiable againstnv-agent-root-cert.pem)- A Tier-3 evaluation dataset — accepted at
evals/evals.json,evals/*.json,eval/*.json, orbenchmark/evals.json BENCHMARK.md— generated benchmark report capturing verifiable uplift data
Verify a skill against the NVIDIA trust anchor nv-agent-root-cert.pem:
pip install model-signing
model_signing verify certificate SKILL_DIR \
--signature SKILL_DIR/skill.oms.sig \
--certificate_chain nv-agent-root-cert.pem \
--ignore_unsigned_files
A successful verification confirms that the skill contents have not been modified since signing by NVIDIA.
See Verify Signed Agent Skills for signature layout, the trust pipeline, and policy options.
Roadmap
- ✅ Public skills catalog with NVIDIA-verified skills across multiple products
- ✅ Automated sync pipeline with skills mirrored from product repos daily
- ✅ Security scanning for all published skills covering instruction safety and supply-chain integrity
- ✅ Skills signing so every published skill carries a verifiable NVIDIA signature
- ✅ Skills universal evaluation criteria and task-specific criteria
- ✅ Skill Card with machine-readable metadata for identity, provenance, quality, and behavioral boundaries
- ✅ Sync-time compliance gates — signature drift detection and missing-artifact enforcement
- ✅ Syndication to external marketplaces — Skills.sh, Codex plugin, Claude Code plugin, ClawHub, Hermes Hub
- 🔲 Syndication to additional MCP hubs and partner channels
Repository Structure
NVIDIA/skills/
├── skills/ # NVIDIA-verified skills (count grows continuously),
│ │ synced from upstream product repos
│ ├── README.md # Browser-facing install guidance
│ ├── <product-prefix>-*/ # Flat layout — one dir per skill, product-prefixed
│ │ # e.g. aiq-*, cuopt-*, cupynumeric-*,
│ │ # dali-*, deepstream-*, dicom-*, digital-health-*,
│ │ # dynamo-*, earth2studio-*, holoscan-*, hsb-*,
│ │ # jetson-*, launch-nemo-rl, mcore-*,
│ │ # nemo-automodel-*, nemo-data-designer-plugin,
│ │ # nemo-evaluator-plugin, nemo-mbridge-* (20 skills),
│ │ # nemo-retriever, nemo-rl-* (4 skills),
│ │ # nemoclaw-user-guide, nemotron-*, nemotron-speech,
│ │ # nv-* (medical AI), physicsnemo-*, rag-*,
│ │ # skill-card-generator, tao-*, tilegym-*,
│ │ # vss-* (15 skills), accelerated-computing-cudf,
│ │ # cudaq-guide, portfolio-optimization
│ ├── omniverse-*/ # Physical AI — manually staged (see manual-components.yml)
│ └── physical-ai-*/ # Physical AI — manually staged
├── components.d/ # Product registry — one file per component, teams onboard here
│ ├── README.md # Schema and onboarding instructions
│ └── <product>.yml # one file per registered product
├── plugins/ # Packaged plugin distributions
│ └── nvidia-skills/ # Curated NVIDIA skills bundle (Claude Code, Codex)
├── plugins.d/ # Plugin build registry — config for `build-plugins.py`
│ ├── README.md
│ ├── _defaults.yml
│ └── nvidia-skills.yml
├── .claude-plugin/ # Claude Code marketplace metadata
│ └── marketplace.json
├── .agents/plugins/ # Agent marketplace metadata (other clients)
│ └── marketplace.json
├── docs/ # Long-form documentation (published via Fern)
│ ├── README.md # How to build the docs locally
│ ├── index.mdx
│ ├── advanced-install.mdx
│ ├── agent-skill-trust-pipeline.mdx
│ ├── release-checklist.mdx
│ ├── scanning-agent-skills.mdx
│ ├── signing-agent-skills.mdx
│ └── skill-cards.mdx
├── fern/ # Fern docs site configuration
├── .github/
│ ├── workflows/ # Sync pipeline, plugin validation, DCO check, author verify
│ └── scripts/ # regenerate-readme.sh, build-plugins.py,
│ # manual-components.yml (temp Physical AI catalog
│ # exception, removed after Computex 2026),
│ # marketplace/metadata.json (skill metadata sidecar)
├── nv-agent-root-cert.pem # Trust anchor for OMS signature verification
├── skills.sh.json # Skills.sh marketplace grouping config
├── CHANGELOG.md
├── CONTRIBUTING.md # Contribution guidelines
├── SECURITY.md # Security reporting policy
├── CODE_OF_CONDUCT.md # Community code of conduct
├── LICENSE-APACHE # Apache 2.0 (source code)
└── LICENSE-CC-BY-4.0 # CC BY 4.0 (documentation/skills)
Skills are maintained in their respective product repos (see the Source column in the Skill Catalog) and synced to this repo daily. Products only appear under skills/ after the sync pipeline confirms each skill carries:
skill.oms.sig— detached OMS-format signature (verifiable againstnv-agent-root-cert.pem)skill-card.md— skill identity and governance card- A Tier-3 evaluation dataset — accepted at
evals/evals.json,evals/*.json,eval/*.json, orbenchmark/evals.json
When evaluation runs produce a BENCHMARK.md, it ships alongside the skill so consumers can see verifiable benchmark uplift data.
Standards & Compatibility
This repository adheres to the Agent Skills specification:
- Skills are portable directories with a
SKILL.mdfile at their root. - Metadata uses YAML frontmatter with required
nameanddescriptionfields. - Skills follow a progressive disclosure model — lightweight metadata loads at startup, full instructions load on activation.
- Validate your skill using the
skills-refreference library.
License
Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
This code is dual-licensed with documentation/skills under the CC-BY-4.0 AND source code under Apache-2.0 license terms. The full license texts can be found in LICENSE-APACHE and LICENSE-CC-BY-4.0 respectively.
其他
低风险
- 来源需自行核对维护者身份。
- 包含脚本或命令调用，安装前请复核。
- 可能需要外部 token、网络权限或第三方服务。
- 未检测到高风险命令。
- 扫描发现：0 条。
Codex — Git Clone 安装
- 安装前请先查看来源仓库和风险报告。
- 克隆仓库：git clone https://github.com/NVIDIA/skills.git
- 将 "skills/jetson-customize-uphy" 文件夹复制到 Codex 的 skills 目录中。
- 重启 Codex 让新的 skill 生效。
Codex — 手动复制安装
- 安装前请先查看来源仓库和风险报告。
- 从源仓库下载 SKILL.md 及相关文件。
- 在 Codex 的 skills 目录中创建新文件夹。
- 将所有 skill 文件复制到新文件夹中。
- 重启 Codex 让新的 skill 生效。
Claude Code — Git Clone 安装
- 安装前请先查看来源仓库和风险报告。
- 克隆仓库：git clone https://github.com/NVIDIA/skills.git
- 将 "skills/jetson-customize-uphy" 文件夹复制到 Claude Code 的 skills 目录中。
- 重启 Claude Code 让新的 skill 生效。
Claude Code — 手动复制安装
- 安装前请先查看来源仓库和风险报告。
- 从源仓库下载 SKILL.md 及相关文件。
- 在 Claude Code 的 skills 目录中创建新文件夹。
- 将所有 skill 文件复制到新文件夹中。
- 重启 Claude Code 让新的 skill 生效。
Cursor — Git Clone 安装
- 安装前请先查看来源仓库和风险报告。
- 克隆仓库：git clone https://github.com/NVIDIA/skills.git
- 将 "skills/jetson-customize-uphy" 文件夹复制到 Cursor 的 skills 目录中。
- 重启 Cursor 让新的 skill 生效。
Cursor — 手动复制安装
- 安装前请先查看来源仓库和风险报告。
- 从源仓库下载 SKILL.md 及相关文件。
- 在 Cursor 的 skills 目录中创建新文件夹。
- 将所有 skill 文件复制到新文件夹中。
- 重启 Cursor 让新的 skill 生效。
GitHub Copilot — Git Clone 安装
- 安装前请先查看来源仓库和风险报告。
- 克隆仓库：git clone https://github.com/NVIDIA/skills.git
- 将 "skills/jetson-customize-uphy" 文件夹复制到 GitHub Copilot 的 skills 目录中。
- 重启 GitHub Copilot 让新的 skill 生效。
GitHub Copilot — 手动复制安装
- 安装前请先查看来源仓库和风险报告。
- 从源仓库下载 SKILL.md 及相关文件。
- 在 GitHub Copilot 的 skills 目录中创建新文件夹。
- 将所有 skill 文件复制到新文件夹中。
- 重启 GitHub Copilot 让新的 skill 生效。
Windsurf — Git Clone 安装
- 安装前请先查看来源仓库和风险报告。
- 克隆仓库：git clone https://github.com/NVIDIA/skills.git
- 将 "skills/jetson-customize-uphy" 文件夹复制到 Windsurf 的 skills 目录中。
- 重启 Windsurf 让新的 skill 生效。
Windsurf — 手动复制安装
- 安装前请先查看来源仓库和风险报告。
- 从源仓库下载 SKILL.md 及相关文件。
- 在 Windsurf 的 skills 目录中创建新文件夹。
- 将所有 skill 文件复制到新文件夹中。
- 重启 Windsurf 让新的 skill 生效。
查看 SKILL.md 原文
name: jetson-customize-uphy
description: Configure Jetson UPHY lane allocation (uphy0/uphy1-config) on Orin/Thor custom carriers. Do NOT use for pinmux or PCIe-only edits.
version: 0.0.2
license: "Apache-2.0"
metadata:
data-classification: public
author: "Jetson Team"
tags:
- bsp
- phase-2
- io
- uphy
domain: metaCustomize UPHY lane allocation
Purpose
Select a UPHY lane allocation on a Jetson custom carrier and edit the
carrier flash-conf fork's ODMDATA="..." to apply the chosen
uphyX-config-N token(s) (plus the UPHY_CONFIG="" clear required for
uphy0-config-6). Kernel-DT alignment per controller is not done
here — after the ODMDATA commit lands, this skill dispatches to the
per-controller skills (/jetson-customize-pcie,
/jetson-customize-mgbe, /jetson-customize-usb), each of which must
compare the chosen allocation against the reference kernel DTB
node-by-node and emit an overlay fragment only when the K-stock value
disagrees with the chosen allocation. Discovery is agentic: options,
lanes, and controllers come from the Adaptation Guide, carrier
schematic, and Module / SoC TRM at run time — never hard-coded. Every
user-visible step renders its data as a markdown table, and the final
summary includes a changes-summary table.
Prerequisites
- Active target-platform profile with
reference_devkit:andcustom_carrier:. <source.root_path>/Linux_for_Tegra/.gitinitialized (/jetson-init-source).- Forked carrier conf present (
/jetson-derive-carrier). - Reachable Adaptation Guide via
documents.adaptation_guide->documents.bsp_developer_guide-> web fetch -> Step-1 prompt fallback. - When
custom_carrier:is present, bothdocuments.custom_carrier_schematicANDdocuments.custom_carrier_pinmux_xlsare REQUIRED. The skill refuses to run if either is missing — routing decisions for the custom carrier cannot be guessed. Reference-devkit-only profiles (nocustom_carrier:block) do not require these.
Overview
UPHY (unified PHY) is the shared high-speed PHY pool on Tegra264 (Thor)
and Tegra234 (Orin). Lane allocation is selected by ODMDATA tokens
(uphy0-config-N, Thor also uphy1-config-N) parsed at flash time by
tegraflash_impl_t264.py::tegraflash_update_bpmp_dtb() and written
into /uphy/uphy{0,1}-config of the BPMP DTB.
Output is a single atomic ODMDATA commit in
<source.root_path>/Linux_for_Tegra/ carrying the chosen
uphyX-config-N token(s), the UPHY_CONFIG="" clear (for
uphy0-config-6), AND every per-controller ODMDATA token derived
from the chosen allocation (pcie@N_status=*, mgbeN-speed-*, USB SS
per-port tokens). Sub-skills (/jetson-customize-pcie,
/jetson-customize-mgbe, /jetson-customize-usb) own only the
kernel-DT overlay fragments — they MUST NOT touch ODMDATA. All
commits follow the batched pristine + customization pattern in
../../context/bsp-customization-workflow.md. Upstream BSP at
<bsp_image.root_path>/ is never edited.
When to invoke
- User says "configure UPHY", "uphy lane allocation", "set uphy0-config-N", "change MGBE speed", or asks to remap PCIe / MGBE / USB3 / UFS on a custom carrier.
- A UPHY-fed controller doesn't enumerate after flash, OR cold boot
dies in BL31 SError /
BPMP firmware is not ready. - A downstream skill reports FMON fault or BPMP-DTB lane mismatch.
Procedure (summary)
Eight steps; full detail in references/procedure.md.
-
Resolve target + docs. Refuse without active profile, custom carrier, source-tree git, or forked carrier conf. Resolve Adaptation Guide / schematic / Module Design Guide / SoC TRM.
-
Locate "Configure the UPHY Lane" in the Adaptation Guide (PDF / HTML mirror /
WebFetch). Cross-check Module Design Guide + SoC TRM. -
Cross-reference the carrier schematic. Cite UPHY net names (
MGBE2_TX_P/N,PEX5_LN0+-, etc.). Zero matching nets = unrouted. -
Enumerate matching UPHY options. Surface every documented
uphy0-config-N(and Thoruphy1-config-N) index. -
Ask the user which config (HARD GATE). Print tables first, then
AskUserQuestion— one per UPHY surface, plus carrier-routing confirmation if any allocated lane is unrouted. Persist answers to the JSON sidecar (references/run-state-sidecar.md). -
Edit carrier flash-conf fork (atomic ODMDATA commit). This skill owns every ODMDATA token for the run. One
ODMDATA="..."line, one commit, all tokens. Sub-skills MUST NOT touch ODMDATA.Decompile the BPMP DTB at
<bsp_image.root_path>/Linux_for_Tegra/bootloader/generic/<BPFDTB_FILE>(BPFDTB_FILEfrom the carrier conf) to snapshot stock state, then emit tokens in this order:a. UPHY surface tokens — every chosen
uphyX-config-N(Thor: both surfaces, even if one equals the guide default). Orderuphy0thenuphy1; separator,. b. Per-controller tokens — one per row whose plan-state differs from BPMP-stock. Match-rows get no token (redundant tokens can drop the whole line).- PCIe:
pcie@N_status=okay|disabled. - MGBE:
mgbeN-speed-<rate>on allocate,mgbeN-speed-delon disable. FMON arms on the controller's own clocks regardless of UPHY allocation — missingdel⇒ BL31 SError reboot loop. Single most common post-flash failure on Thor. - USB SS: per-port tokens when the SoC grammar exposes them.
c.
UPHY_CONFIG=""clear whenuphy0-config-6is selected (BCT pinmux clear per Adaptation Guide).
- PCIe:
-
Build the per-controller allocation table and dispatch. Derive one row per UPHY-fed controller (PCIe / MGBE / USB SS / UFS) with
{class, instance, allocated?, BPMP-stock, K-stock, routed?, Desired K state}. This table drives both (a) Step 6's ODMDATA tokens and (b) the sub-skills' overlay fragments — build it before Step 6 commits.Then invoke
/jetson-customize-pcie,/jetson-customize-mgbe, and/jetson-customize-usbfor kernel-DT overlay fragments only (no ODMDATA edits — Step 6 owns the line). Each sub-skill re-reads K-stock from<bsp_image.root_path>/Linux_for_Tegra/kernel/dtb/tegra<soc>-*-nv.dtband skips emission when K-stock matches Desired K state. UFS handling stays inline here (no UFS sub-skill).Invoke all three whenever their controller class is present on this SoC (e.g. skip MGBE on Orin). Ask the operator first; on
yes, run the sub-skill inline. -
Summary + next-step chain. Headline, breakdown, choices table (UPHY surface | chosen config | lane summary | UPHY_CONFIG-clear), changes-summary table (file | repo | commit SHA | one-line summary covering this skill's commit + every dispatched sub-skill's commit), then drive the downstream chain (more I/O? build & promote? flash? validate?) via sequential
AskUserQuestionprompts perreferences/procedure.mdStep 8. Never substitute a printed "Next step: …" line for the prompts.
Limitations
- Only supports Tegra234 (Orin) and Tegra264 (Thor) UPHY surfaces.
- Does not edit pinmux, PCIe-only DT properties absent from BPMP DTB
(
num-lanes,pcie-mode), or upstream BSP files. - Does not flash, build, or promote — chain into
/jetson-build-sourceand downstream skills. - Hard-coded option tables are forbidden; if no Adaptation Guide source resolves the skill refuses rather than guessing.
- Not table-driven across releases: every run re-reads the Guide for the active BSP version.
Troubleshooting
- Cold boot reboot loop / BL31
plat_setup.c:726/BPMP firmware is not readyafteruphy0-config-6: a later^UPHY_CONFIG=line in the carrier conf re-overrode the clear. Comment it (seereferences/procedure.mdStep 6). wait-for-device failedat flash, BPMP DTB unchanged: an ODMDATA token had wrong shape (e.g.mgbe0-speed-0). One bad token drops the wholeODMDATA="..."line. Inspect grammar inreferences/procedure.md.- BL31 SError reboot loop after disabling an MGBE: missing
mgbeN-speed-del. FMON arms on the controller's own clocks regardless of UPHY allocation. - Newly-routed controller doesn't enumerate: stock kernel DTB had
status="disabled". Overlay must emitstatus="okay"(matrix row 4 inreferences/procedure.md). - Pinmap delta=0 but board still misbehaves: UPHY differential
pairs are absent from pinmux
.xlsm. Drive decisions off schematic net names, not the pinmap. - Duplicate
pcie@<addr>fragments: another skill (jetson-customize-pcie) already owns that node. Scope this skill to MGBE / UFS / USB3 SS / PCIe-status-only and cite the other overlay.
References
references/procedure.md— full eight-step procedure.references/gotchas.md— cross-cutting gotchas.references/run-state-sidecar.md— JSON sidecar schema + idempotency.../../references/platform_template.yaml—documents:schema.../../context/bsp-customization-workflow.md— overlay edit protocol.../../references/bsp-customization-kernel-dtb.md— composite-overlay filename / append protocol.../jetson-derive-carrier/SKILL.md— produces the conf this skill edits.../jetson-init-source/SKILL.md— produces the two git repos this skill commits into.../jetson-generate-kb/SKILL.md— KB consulted for chip family + file locations.
评论 (0)
暂无评论，成为第一个评论者吧！