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
proteinmpnn-nim
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. 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：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/bionemo-proteinmpnn-nim" 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/bionemo-proteinmpnn-nim" 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/bionemo-proteinmpnn-nim" 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/bionemo-proteinmpnn-nim" 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/bionemo-proteinmpnn-nim" 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: proteinmpnn-nim
description: >
Run ProteinMPNN inverse folding via NVIDIA NIM to design protein sequences for a target backbone. Sends user-provided PDB files and design parameters to NVIDIA's hosted API, authenticated with an environment API key, or to a user-selected local NIM. Use for sequence design, backbone redesign, fixed chains and residues, omit_AAs, sampling temperature, soluble model, local Docker, and multi-FASTA output.
license: Apache-2.0 AND CC-BY-4.0
compatibility: "Python >=3.10; requests>=2.28"
allowed-tools: Bash, Read, Write, AskUserQuestion
permissions:
- network
- envProteinMPNN NIM
Design protein sequences for a supplied backbone PDB. Use this guide for first-pass hosted/local usage; load supplemental files only when needed:
references/api.md: exact endpoints, schemas, Docker flags, response fields.references/science.md: inverse-folding uses, limits, and validation.references/parameters.md: design controls, fixed positions, sampling.references/validation.md: FASTA, score, and structure checks.references/examples.md: compact hosted/local request patterns.
Choose Mode
Honor the user's explicit mode; otherwise use the configured runtime. NIM_API_MODE=local selects
the local service at PROTEINMPNN_NIM_URL; the URL defaults to
http://localhost:8000 for a NIM running in the same host or container. Ask only
when neither the environment nor the user's request makes the mode clear:
Hosted NVIDIA API or local Docker NIM?
- Hosted:
https://health.api.nvidia.com/v1/biology/ipd/proteinmpnn/predict - Local: append
/biology/ipd/proteinmpnn/predicttoPROTEINMPNN_NIM_URL(default base URL:http://localhost:8000).
Local inference paths do not include /v1/. Hosted requests use Authorization: Bearer $NGC_API_KEY. Supported local Docker
startup uses NGC_API_KEY (or NVIDIA_API_KEY via the preflight) for
registry login, entitlement checks, and first-run model downloads; pass it
into the container with -e NGC_API_KEY. Local inference requests use no
auth header after readiness. Warm-cache key-free startup varies by
image/version and should not be assumed.
Data Transfer and Authorization
Before a hosted request, tell the user that the entire PDB file and design parameters will be uploaded to NVIDIA's hosted API at the endpoint above. Proceed if the user has explicitly requested hosted processing of that PDB or already approved the transfer; otherwise ask for confirmation before submitting. For confidential structures, recommend a local NIM in the user's approved environment. A configured local URL may point to another machine; use only the configured or user-selected destination. Do not switch from local to hosted processing without the user's authorization.
The client reads NIM_API_MODE, PROTEINMPNN_NIM_URL, and, for hosted mode,
NGC_API_KEY from the environment. It sends the key only in the HTTPS
Authorization header to the hosted endpoint; local inference sends no key.
Keep credentials out of logs and saved artifacts. The output directory contains
the full input PDB in request.json and the returned sequences and scores, so
use a location appropriate for the input's sensitivity. See
references/api.md for endpoint and data-handling details.
Local Docker
For local setup, run the full sequence — env preflight, docker login,
docker run, readiness loop, then the no-auth localhost request; do not answer
with only a localhost Python request. For the exact preflight (.env sourcing,
NGC_API_KEY/NVIDIA_API_KEY handling, and the docker run for
nvcr.io/nim/ipd/proteinmpnn:latest), copy the command block in
references/api.md under Docker Reference verbatim.
This NIM's cache mount is /home/nvs/.cache/nim, not /opt/nim/.cache.
When PROTEINMPNN_NIM_URL is supplied, the service is already managed elsewhere;
use that URL and do not start another Docker container.
Readiness:
proteinmpnn_nim_url="${PROTEINMPNN_NIM_URL:-http://localhost:8000}"
until curl -sf "${proteinmpnn_nim_url%/}/v1/health/ready"; do sleep 5; done
Instructions
For a request to execute a design, run scripts/design.py
and inspect its results. Writing a request script alone does not complete an
execution request. If the user asks only for code or setup instructions, provide
those without submitting an inference request.
- Use the user's PDB path and requested sequence count. The client reads the
entire PDB into
input_pdb; do not replace or truncate the supplied backbone. - Select
--mode hostedor--mode localand follow Data Transfer and Authorization above before submitting. Hosted mode uploads the PDB to the documented NVIDIA endpoint and requiresNGC_API_KEYin the environment. Check only whether the key is set; do not print it, dump the environment, or save authentication headers. Local inference sends no authorization header. - Choose a new
--output-dirfor each request. The client reserves it before submitting, preserves the raw response for diagnostics, and validates the designed sequence count and score alignment before reporting completion. - Read
summary.jsonand report the actual results described below. If the request or validation fails, report the failure and diagnostic path; do not substitute example sequences or repeatedly resubmit the same request.
Examples
Run from this skill's directory, or use an absolute path to scripts/design.py.
Substitute the user's input path and a new output directory:
python scripts/design.py --mode hosted \
--pdb /path/to/backbone.pdb --num-sequences 10 \
--temperature 0.1 --output-dir /path/to/new-design-run
For a running local NIM, use --mode local; the client honors
PROTEINMPNN_NIM_URL. To design only chain A, exclude cysteine, or request the
soluble model, add --chains A, --omit-aas C, or --soluble respectively.
--seed sets random_seed; --ca-only selects the CA-only model. The helper
uses one temperature per request; run separate output directories for a
temperature sweep. For advanced JSONL controls or a custom batch request,
use references/api.md and the post-response example in
references/examples.md.
Save And Report Output
The client writes request.json, response.raw, response.json,
designed_sequences.fa, and summary.json into the requested output directory.
The FASTA preserves the complete returned mfasta, including a native/WT entry
when present. The summary contains only designed sequences, each paired with
its actual score, and records whether scores came from the JSON array or FASTA
headers. It is also printed after the artifacts are saved and checked.
In the final response, report:
- The number of designed sequences, excluding the native/WT reference.
- Each design's identifier and actual returned score, plus its sequence (for long sequences, give a clearly labelled preview and link to the full FASTA).
- The saved FASTA and summary paths, and the raw response path for provenance.
- That these are inverse-folding candidates, with no fold-back validation performed unless it was actually requested and run.
Do not treat a score as proof that a sequence folds or binds. Further validation
with Boltz2 or OpenFold3 is an optional next step. For FASTA/score sanity checks,
read references/validation.md.
Limits And Troubleshooting
- Minimum GPU VRAM: about 3 GB.
sampling_tempmust be a list, even for one value.- Empty
mfasta: check non-emptyinput_pdbandnum_seq_per_target >= 1. - PDB parse errors: use valid PDB ATOM records.
- Local URL 404 usually means an accidental
/v1/prefix. - Cache mount error: use
/home/nvs/.cache/niminside the container.
评论 (0)
暂无评论，成为第一个评论者吧！