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
doca-dma
Official, NVIDIA-verified Agent Skills for Claude Code, Codex, and other coding agents.
审核状态：已审核Quality 72Security 70
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
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。
- 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：1 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/doca-dma" 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/doca-dma" 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/doca-dma" 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/doca-dma" 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/doca-dma" 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
license: Apache-2.0
name: doca-dma
description: >
Use this skill when the user is doing hands-on DOCA DMA
programming — bringing up a doca_dma context, configuring the
single doca_dma_task_memcpy task type, sizing buffers via the
doca_dma_cap_task_memcpy_* queries, setting LOCAL_READ_ONLY /
LOCAL_READ_WRITE permissions on source / destination doca_mmap
regions (plus doca_mmap_export_* for cross-peer copies),
driving the progress engine, or debugging DOCA_ERROR_* returns.
Trigger even when the user does not explicitly mention "DOCA
DMA" or "doca_mmap" — typical implicit phrasings include
"memcpy host buffer to BlueField without using the CPU",
"offload a bulk copy to the DPU", "copy returns NOT_PERMITTED
on first submit", "buffer too big for one DMA task", "task
submitted but no completion", or "scatter-gather copy between
two memory regions". Refuse and route elsewhere for
cross-network copies (DOCA RDMA), producer/consumer messaging
(DOCA Comch), DOCA Core / progress-engine internals, or DOCA
install — those belong to other skills.
metadata:
kind: library
compatibility: >
Requires DOCA SDK installed at /opt/mellanox/doca on Linux
(Ubuntu 22.04/24.04 or RHEL/SLES) with a BlueField DPU or
ConnectX NIC attached. Reads the user's local install via
`pkg-config doca-dma` and inspects
/opt/mellanox/doca/{lib,include,samples,applications}.DOCA DMA
Where to start: This skill assumes DOCA is already installed
and the user is doing hands-on DMA work on a BlueField /
ConnectX / host with DOCA. Open TASKS.md if the user
wants to do something (configure / build / modify / run / test /
debug); open CAPABILITIES.md when the question
is what can DMA express on this version. If the user has not
installed DOCA yet, route to
doca-setup first. If the user is
not sure DMA is even the right library — the data has to traverse
the network, or the flow is small messages between two processes —
read the path-selection rule in
CAPABILITIES.md ## Capabilities and modes
before configuring anything.
Example questions this skill answers well
The CLASSES of DMA questions this skill is built to answer, each with one worked example. The agent should treat the class as the load-bearing piece — the worked example is a single instance.
- "How do I bring up a DOCA DMA context and copy a buffer
between host and DPU?" — worked example: "copy a 64 KiB
buffer from host memory to DPU memory in one task, starting from
the shipped DMA Copy reference application". Answered by the
lifecycle + memcpy-task workflow in
TASKS.md ## configure+CAPABILITIES.md ## Capabilities and modestask-type table. - "How big a buffer can I memcpy in one task on this device?" —
worked example: "can I copy 16 MiB in a single
doca_dma_task_memcpy". Answered by the capability-query rule (doca_dma_cap_task_memcpy_get_max_buf_size, plus_get_max_buf_list_lenfor scatter-gather) inCAPABILITIES.md ## Capabilities and modes- the discovery step in
TASKS.md ## configure.
- the discovery step in
- "What permissions do my source and destination mmaps need?" —
worked example: "my memcpy task returns
DOCA_ERROR_NOT_PERMITTEDon the first submit". Answered by the source / destination permission matrix inCAPABILITIES.md ## Safety policy- the permission checklist in
TASKS.md ## test.
- the permission checklist in
- "Should I use DOCA DMA or DOCA RDMA / Comch / a plain CPU
memcpy for this copy?" — worked example: "I have a 1 MiB copy
that has to go from a host process to a DPU process; do I want
DMA or Comch fast-path". Answered by the "when to use DMA"
vs "when not to" path-selection bullet in
CAPABILITIES.md ## Capabilities and modes- the routing pointers in
## Related skills.
- the routing pointers in
- "Is DOCA DMA on my installed version, and is the memcpy task
supported on my device?" — worked example: "is
doca_dma_task_memcpyavailable on DOCA 2.6 against this ConnectX-6". Answered by the version-compatibility overlay inCAPABILITIES.md ## Version compatibility, which cross-links the canonical detection chain indoca-version, plus the capability-query rule inCAPABILITIES.md ## Capabilities and modes. - "What does this
DOCA_ERROR_*from a DMA call mean and which layer caused it?" — worked example: "DOCA_ERROR_AGAINfromdoca_task_submiton adoca_dma_task_memcpy". Answered by the DMA overlay on the cross-library taxonomy inCAPABILITIES.md ## Error taxonomy- the layered ladder in
TASKS.md ## debugthat escalates todoca-debug.
- the layered ladder in
Audience
This skill serves external developers building applications
that consume the DOCA DMA library — i.e., users whose code calls
doca_dma_* (directly in C/C++, or through FFI/bindings from
another language) to copy bytes between two doca_mmap regions
using the BlueField DMA engine instead of the host CPU. It is
not for NVIDIA developers contributing to DOCA DMA itself.
Language scope. DOCA DMA ships as a C library with
pkg-config module name doca-dma. The shipped samples are
written in C. C and C++ consumers are the canonical case and the
worked examples in TASKS.md assume that path. Other-language
consumers (Rust, Go, Python, …) consume the same *.so through
FFI or language-specific bindings; the skill's contribution in
that case is to keep the lifecycle, capability-discovery,
permission, error-taxonomy, and path-selection guidance
language-neutral, and to route the agent to the public C ABI as
the authoritative surface that any wrapper will eventually call.
When to load this skill
Load this skill when the user is doing hands-on DOCA DMA work, in any language. Concretely:
- Initializing a
doca_dmacontext on adoca_devand configuring the memcpy task type viadoca_dma_task_memcpy_set_confbeforedoca_ctx_start(). - Setting up the source and destination
doca_mmapregions for a memcpy, including the per-side permission flags (DOCA_ACCESS_FLAG_LOCAL_READ_ONLYon the source,DOCA_ACCESS_FLAG_LOCAL_READ_WRITEon the destination) and, for cross-peer copies, thedoca_mmap_export_*step. - Reading the device capability surface for DMA via the
doca_dma_cap_task_memcpy_*query family (_is_supported,_get_max_buf_size,_get_max_buf_list_len) before sizing any buffer or assuming scatter-gather is available. - Submitting
doca_dma_task_memcpytasks against a DOCA progress engine and reacting to per-task completion events. - Choosing between DOCA DMA and an adjacent option (DOCA RDMA when the data has to cross the network, DOCA Comch fast-path for message-oriented producer/consumer flows, plain CPU memcpy when the copy is tiny and one-shot).
- Debugging a
DOCA_ERROR_*returned from a DMA call (lifecycle vs. permission vs. capability vs. would-block) and the per-task completion status reported on the progress engine. - Designing or extending non-C bindings (Rust, Go, Python, …) that wrap the DMA C ABI — for the lifecycle, permission, and capability rules the wrapper must honor.
Do not load this skill for general DOCA orientation, install
of DOCA itself, or non-DMA library questions. For those, use
doca-public-knowledge-map.
What this skill provides
This is a thin loader. The body keeps only the orientation needed to pick the right next file. The substantive DMA-specific material lives in two companion files:
CAPABILITIES.md— what DMA can express on this version: the singledoca_dma_task_memcpytask type and its scatter-gather buffer-list shape, the capability-query surface (doca_dma_cap_task_memcpy_*), the DMA error taxonomy (mapped onto the cross-libraryDOCA_ERROR_*set), the observability surface (per-task completion events on the progress engine), the source / destination mmap permission policy, and the path-selection rule against the adjacent libraries (RDMA / Comch / CPU memcpy).TASKS.md— step-by-step workflows for the six in-scope DMA verbs:configure,build,modify,run,test,debug. Plus aDeferred task verbsblock that points out-of-scope questions at the right next skill.
The skill assumes a host or BlueField where DOCA is already
installed at the standard location and the user has the
privileges their public install profile expects. It does not
cover installing DOCA — that path goes through
doca-setup.
What this skill deliberately does not ship
This skill is agent guidance, not a samples or templates bundle. To keep the boundary clean, it deliberately does not contain — and pull requests should not add:
- Pre-written DOCA DMA application source code, in any
language. The verified DMA source code is the shipped C
samples at
/opt/mellanox/doca/samples/doca_dma/<name>/and the DMA Copy reference application reachable viadoca-public-knowledge-map. The agent's job is to route the user to those files and prescribe a minimum-diff modification on them via the universal modify-a-sample workflow indoca-programming-guide, layered with the DMA-specific overrides inTASKS.md ## modify. - Standalone build manifests (
meson.build,CMakeLists.txt,Cargo.toml, …) parked inside the skill. The agent constructs the build manifest in the user's project directory against the user's installed DOCA, wherepkg-config --modversion doca-dmais the source of truth. - A
samples/,bindings/, orreference/subtree of any kind. A mock or incomplete artifact in this skill's tree, even one labeled "reference", is misleading: users will read it as buildable.
Loading order
- Read this
SKILL.mdfirst to confirm the user's question is in scope. - For the DMA capability surface, the memcpy task type, the capability-query rule, the source / destination permission matrix, the error taxonomy, observability, and the path-selection rule against RDMA / Comch / CPU memcpy, see CAPABILITIES.md.
- For step-by-step workflows — configure, build, modify, run, test, debug — see TASKS.md.
Both companion files cross-link to each other,
doca-version for the canonical
version-handling rules, and
doca-public-knowledge-map
whenever the right answer is "look it up in the public docs or
the installed package layout" rather than "DMA-specific
guidance".
Related skills
doca-public-knowledge-map— the routing table for every public DOCA documentation source and the on-disk layout of an installed DOCA package. The DMA URL ishttps://docs.nvidia.com/doca/sdk/DOCA-DMA/index.html; the canonical reference application is DMA Copy.doca-setup— env preparation, install verification, and the I have no install yet path with the public NGC DOCA container. This skill assumes its preconditions are satisfied.doca-version— canonical DOCA version-handling rules. This skill's## Version compatibilitycross-links the four-way match rule + detection chain and adds at most one DMA-specific overlay rule.doca-structured-tools-contract— the bundle's structured-tools precedence rule (detect / prefer / fall back / report). The Command appendix in TASKS.md honors this contract.doca-programming-guide— general DOCA programming patterns shared by every library: the canonicalpkg-config+ meson build pattern, the universal modify-a-shipped-sample first-app workflow, the universal lifecycle, the cross-libraryDOCA_ERROR_*taxonomy, and the program-side debug order. This skill layers DMA specifics on top.doca-rdma— the right library when the copy has to traverse the network. This skill's path-selection rule routes to RDMA when DMA is not the answer.doca-comch— the right library when the flow is producer / consumer messaging between a host and DPU process pair, rather than a raw mmap-to-mmap copy.doca-debug— the cross-cutting debug ladder (install / version / build / link / runtime / program / driver). DMA-specific debug (lifecycle violations, permission mismatches, oversize-buffer rejections) overlays on top of that ladder.
评论 (0)
暂无评论，成为第一个评论者吧！