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-mgbe
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-mgbe" 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-mgbe" 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-mgbe" 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-mgbe" 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-mgbe" 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-mgbe
description: >-
Enable Jetson Thor 25G/10G/1G MGBE QSFP via kernel-DT overlay.
Do NOT use for UPHY lane allocation or ODMDATA edits.
version: 0.0.1
license: "Apache-2.0"
metadata:
data-classification: public
author: "Jetson Team"
tags:
- bsp
- phase-2
- io
- mgbe
- ethernet
domain: meta
permissions:
file_read:
- "{workspace}/target-platform/"
- "{source.root_path}/"
- "{bsp_image.root_path}/Linux_for_Tegra/"
- "{documents.root_path}/"
file_write:
- "{workspace}/target-platform/"
- "{source.root_path}/bsp_sources/hardware/nvidia/"
shell:
- "dtc"
- "fdtoverlay"
- "cpp"
- "git"Customize MGBE / 25G QSFP
Overview
Thor T264 exposes mgbe0..mgbe3. On a custom carrier, the 25G QSFP cage
(or 10G / 1G fiber path) is wired to one of them through SerDes — with
or without an external MDIO PHY in front of the cage. This skill
renders the kernel-DT overlay that pairs the BPMP allocation with
kernel-side status="okay" + PHY plumbing on &mgbeN.
Out of scope:
- UPHY lane allocation — owned by
/jetson-customize-uphy. Refuse if the chosenuphy1-config-Ndoesn't allocate the target MGBE. - All ODMDATA tokens (
mgbeN-speed-*, sub-nodemgbeN_status=*) — owned by/jetson-customize-uphyin its single atomic ODMDATA commit. This skill MUST NOT touchODMDATA=.
Output is one commit to the composite custom overlay .dts in the
bsp_sources/ hardware repo. /jetson-build-source compiles the
composite to .dtbo and owns its Makefile + flash-conf registration.
When to invoke
- The user says "enable 25G", "configure QSFP", "set MGBE PHY mode", "wire MGBE to QSFP", or asks to bring up a 10G / 1G fiber path.
- Cold boot succeeds but
ip link show mgbe<N>reportsstate DOWNorNO-CARRIERon the configured controller, OR the controller never appears at all. jetson-customize-uphyran withuphy1-config-8(or another config allocating MGBE) and you now need to bring up the per-controller side.
Prerequisites:
- Active profile selected with
reference_devkit:(Thor) +custom_carrier:blocks. <source.root_path>/Linux_for_Tegra/.gitexists (/jetson-init-source)./jetson-derive-carrierhas run — carrier flash-conf fork is in the overlay tracker./jetson-customize-uphychose a UPHY config that allocates the target MGBE controller's lanes (uphy1-config-8on Thor for MGBE0..3 25G).- Source-of-truth docs registered or supplied at prompt: Adaptation Guide, Module Design Guide, SoC TRM.
- When
custom_carrier:is present, bothdocuments.custom_carrier_schematicANDdocuments.custom_carrier_pinmux_xlsare REQUIRED. Refuse the run if either is missing — MGBE routing on a custom carrier cannot be guessed. Reference-devkit-only profiles skip this check. dtcon PATH.
Procedure
See references/procedure.md for the full step-by-step procedure (Steps 1–8). Summary:
- Resolve active target + documents. Validate active profile, custom_carrier, overlay tracker; locate the relevant Adaptation Guide chapter and pinmap.
- Per-controller question loop. AskUserQuestion driven by
questions.json(controller, phy_mode, attach kind, I²C bus/addr, reset GPIO, compatible_list). - Derive max-speed from phy_mode. Decompile the BPMP DTB to pick sub-node vs top-level token grammar; cite the inspection in notes.
- Verify HSIO pins + auto-fix. Run
pin_verifier.pyfor MDC/MDIO/RESET/INT; surface mismatches and route to/jetson-customize-pinmux. - (no ODMDATA edits.) MGBE ODMDATA tokens are emitted by
/jetson-customize-uphy. Step 5 only records the BPMP DTB token-form inspection (sub-node vs top-level) innotes[]for audit. - Append composite-overlay fragments. Write one fragment per controller into the composite custom overlay
.dts; obey the/* custom-bsp: mgbe:mgbe... */marker contract; run the cpp/dtc/fdtoverlay pre-flight. - (Reserved.) Sibling-skill ordering / cross-cutting validation.
- Run-state sidecar + summary + next-step chain. Write
<profile-stem>.jetson-customize-mgbe.jsonand emit the one-line + table summary, then drive the downstream chain via sequentialAskUserQuestionprompts perreferences/procedure.mdStep 8. Never substitute a printed "Next step: …" line for the prompts.
Gotchas
- Stock Thor BPMP DTB has no
/mgbe/mgbe@Nsubtree — onlymgbe<N>-speedunder/uphy. Themgbe<N>_status=disabledsub-node token is silently rejected on these releases; the whole ODMDATA line is then dropped at flash time. Always decompile BPMP DTB (Step 3) before emitting; use the top-level dashed form (mgbe<N>-speed-delto remove,mgbe<N>-speed-25Gto set) when the sub-node isn't there. Same wrong-form failure surface asjetson-customize-uphy. mdiochild needs both#address-cells = <1>AND#size-cells = <0>whenphy_attach_kind=="phy". Missing either → kernel rejectsphy@<addr>reg property at probe; MGBE never comes up.- Overlay root
compatiblemust intersect live DT compatible. UEFI plugin-manager filters by compatible match. A mismatched overlay is silently skipped — flash succeeds, MGBE stays disabled, no error in dmesg. Always sanity-check against/proc/device-tree/compatibleon a booted reference DUT. OVERLAY_DTB_FILEordering isjetson-build-source's problem, not this skill's. This skill never touches the carrier flash conf. The composite custom overlay is registered (by/jetson-build-sourceStep 5.0a) AFTER the platform*-dynamic.dtbo, which is the correct ordering. If you find yourself appendingOVERLAY_DTB_FILE+=in this skill, you're duplicating ownership — stop, and let the build skill do it.- UPHY lane allocation is
jetson-customize-uphy's job. If the chosenuphy1-config-Ndoesn't allocate lanes for the target MGBE controller, BL31 SError (fmon_update_config: detected fault 0x80) on cold boot. Always run/jetson-customize-uphyfirst; cite the chosenuphy1-config-Nin this skill's run-summarynotes[]. - Don't disable a stock-okay controller via ODMDATA alone. Same
rule as
jetson-customize-uphy:mgbe<N>_status=disabledfor a controller that's already disabled in BPMP DTB is a no-op the parser may treat as ambiguous → drops the rest of the ODMDATA line. Disable via the kernel-DT overlay (status="disabled") only. - Don't touch the upstream BSP at
<bsp_image.root_path>. All edits land in the overlay tracker /bsp_sourcesmono-repo under the pristine + customization commit pattern. - JSON sidecar is structured state, not authoritative. Same
caveat as
jetson-customize-uphy: ODMDATA + overlay.dts+ two git commits are the device-facing outputs; the sidecar is for tooling and idempotency only.
References
questions.json— Q-1..Q-8 prompt schema consumed by Step 2.../../scripts/pin_verifier.py— shared HSIO pin verifier (Step 4).../../references/platform_template.yaml—documents:block consumed by Step 1.../../context/bsp-customization-workflow.md— overlay edit protocol (batched pristine + customization commit).../jetson-customize-uphy/SKILL.md— sibling skill that owns UPHY lane allocation. Must run before this skill to setuphy1-config-Nfor MGBE-allocated configurations.../jetson-customize-pinmux/SKILL.md— sibling skill invoked by Step 4 (with operator confirmation) to fix pin SFIO mismatches.../jetson-derive-carrier/SKILL.md— must run first; produces the carrier flash-conf fork edited in Step 5 and the carrier base overlay this skill orders after.../jetson-init-source/SKILL.md— produces the overlay tracker + bsp_sources repo this skill commits into.
评论 (0)
暂无评论，成为第一个评论者吧！