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
i4h-workflow-create
Official, NVIDIA-verified Agent Skills for Claude Code, Codex, and other coding agents.
审核状态：已审核Quality 80Security 80
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.
Agent / MCP / Skill 创作
中风险
- 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。
- 未检测到明显外部权限要求。
- 未检测到高风险命令。
- 扫描发现：3 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/i4h-workflow-create" 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/i4h-workflow-create" 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/i4h-workflow-create" 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/i4h-workflow-create" 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/i4h-workflow-create" 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: i4h-workflow-create
version: "0.7.0"
description: Create a new agentic env by forking an existing env. Use for new env/task scaffolding, not scene edits or baking.
license: Apache-2.0
metadata:
author: "Isaac for Healthcare Team <isaac-for-healthcare-support@nvidia.com>"
tags:
- isaac-for-healthcare
- i4h
- agentic-workflow
- environment
- scaffoldingi4h Workflow - Create Env
Purpose
Create the first runnable version of a new workflows/agentic environment by forking the closest existing env. Keep this skill focused on env scaffolding: YAML, assets, task, env class, runtime, and validation. Do not use this skill to polish an existing scene, add optional props/cameras, or bake bridge edits; use [[i4h-workflow-scene-edit]] for that.
Base Code
Resolve and work from the i4h-workflows root:
ROOT="${I4H_WORKFLOWS:-$(git rev-parse --show-toplevel 2>/dev/null)}"
if [ ! -d "$ROOT/workflows/agentic" ]; then
ROOT="${I4H_WORKFLOWS:-$HOME/i4h-workflows}"
[ -d "$ROOT/workflows/agentic" ] || git clone https://github.com/isaac-for-healthcare/i4h-workflows "$ROOT"
fi
export I4H_WORKFLOWS="$ROOT"; cd "$ROOT"
What to Load
Before editing, load only the references that match the request:
- Always load
skills/i4h-workflow/references/repo-map.md. - Always load
skills/i4h-workflow-create/references/create-contract.md. - Load
skills/i4h-workflow-create/references/env-authoring-patterns.mdto choose the source env, robot owner, policy stack, and YAML pattern. - Load
skills/i4h-workflow-create/references/hybrid-layout-rules.mdonly for hybrid scene+robot envs, G1 footprint/table-height work, catalog USD scale, or support-surface layout. - Load a recipe file only when the prompt matches it exactly enough to avoid re-deciding components.
Recipe routing:
| Prompt shape | Reference |
|---|---|
Surgical tool sorting using G1 based on scissor_pick_and_place | skills/i4h-workflow-create/references/g1-surgical-tool-sort.md |
Create Contract
Create the env shell first. A normal create task should produce exactly these surfaces unless the chosen existing pattern requires an explicit exception:
| Surface | Path |
|---|---|
| Env YAML | workflows/agentic/config/environments/<env>.yaml |
| Assets | workflows/agentic/arena/arena/assets/<env>.py |
| Task | workflows/agentic/arena/arena/tasks/<env>.py |
| Env class | workflows/agentic/arena/arena/environments/<env>_environment.py |
| Runtime | workflows/agentic/arena/arena/runtimes/<env>.py |
Keep all paths repo-root relative and keep the workflows/agentic/ prefix. Do not create a new policy package, README, docs, or shared-module edits unless the user explicitly asks and the policy stack truly needs it.
Workflow
- Resolve components: env id, source env, scene/assets source, robot owner, policy stack, model/checkpoint, cameras, objects/destinations, success rule.
- Inspect the selected source env YAML, env class, assets, task, runtime, and policy stack files before writing code.
- Fork the closest working pattern. Preserve inline-scene vs registry-asset style; do not invent a new architecture.
- Keep the first version minimal and runnable. Optional visual polish, extra props, new cameras, and layout changes happen through scene-edit after the env exists.
- Run validation and fix source until the env builds and the rendered scene is visually sane.
For an exact recipe match, do not re-derive the architecture after loading the recipe. Read only the source files the recipe names, skip large source runtime files when the recipe says the runtime is a re-export, then write the five contract files immediately. Do not inspect third-party framework internals unless a validation error requires it.
Plan shape:
Env id:
Source env / recipe:
Scene/assets source:
Robot owner:
Policy stack + model/checkpoint:
Objects/destinations:
Success rule:
Files to create:
Validation:
Hard Rules
- Env YAML is the source of truth for robot, policy, cameras, task text, dataset mapping, and train defaults.
- Fork from the nearest existing implementation. For hybrid envs, robot integration comes from the robot owner and scene construction comes from the scene source.
- Sorting tasks need at least two object types, at least two destinations, and a success rule that fails swapped placements.
- Static/kinematic destination props may not expose
.data.root_pos_w; success checks must fall back toentity.get_world_poses(). - Functions referenced by
func=inside config classes must be defined above those classes. - G1 WBC envs need footprint clearance and matched ground/base-height values; load
hybrid-layout-rules.mdbefore choosing those numbers. - Additional fixed cameras are normally scene-edit/bake work. If a create prompt explicitly requires a policy/dataset camera, load the scene-edit camera and bake references and wire every surface in one pass.
Validation
Run these static checks:
python -m py_compile <changed-python-files>
workflows/agentic/policy/run.sh --list-envs
workflows/agentic/arena/run.sh --env <env> --dry-run
workflows/agentic/policy/run.sh --env <env> --dry-run
Then run the real build/visual gate from skills/i4h-workflow-create/references/create-validation.md. --dry-run is shallow: it does not prove the task/assets instantiate or the scene is visually usable. Do not report a new env as ready until the bridge reaches ready, key objects are valid, a viewport capture has been inspected, and the bridge is stopped with:
workflows/agentic/arena/stop.sh --env <env>
If Isaac Sim cannot launch on the host, report that as a blocker and include the static check results; do not present static-only as success.
Hand Off to Scene Edit
After the env shell passes create validation, use [[i4h-workflow-scene-edit]] for:
- Adding, moving, resizing, or replacing objects/assets.
- Adding fixed room/overhead/wrist cameras.
- Live bridge edits.
- Baking live changes into source.
- Running
local-agent/validate-bake.sh.
The scene-edit workflow owns object snippets, camera snippets, bridge endpoint details, and bake checklists so this create workflow stays small.
Prerequisites
- Workflow setup has completed via [[i4h-workflow-setup]];
.venv, third-party checkouts, and Isaac Sim launch support are present. - The source env and target env id are known, or the prompt provides enough information to choose them from existing patterns.
- Bridge validation runs on a GPU host that can launch Isaac Sim.
Limitations
- Create one env per invocation.
- Fork existing patterns; do not invent a new policy stack or shared framework.
- Use scene-edit for post-create polish, optional props, camera baking, and source persistence of live edits.
--dry-runis not a substitute for the bridge build and visual validation gate.
Troubleshooting
- Missing setup: if
.venv, third-party checkouts, orrun.shentrypoints are missing, run [[i4h-workflow-setup]] first. - Env not listed: check that the YAML exists at
workflows/agentic/config/environments/<env>.yamland all files use the fullworkflows/agentic/repo-root prefix. - Build fails after static checks pass: inspect the first Isaac/arena stack trace; common causes are bad cfg kwargs, a helper defined below a config class, or an observation pointing at a missing sensor.
- Scene looks wrong: fix source or use scene-edit to live-adjust and bake; do not report success from static checks alone.
Final Response
Report the env id, source choices, files created, validation commands and results, capture paths, and any blocker. Do not commit changes unless the user explicitly asks.
评论 (0)
暂无评论，成为第一个评论者吧！