AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Home/
Dev Tooling/
skills
skills logo

skills

Active·★ 4.0k·Apache-2.0·Updated 2026-11-18

NVIDIA Agent Skills is the official, open-source repository maintained by NVIDIA Corporation (Apache 2.0 / CC-BY-4.0) delivering verified, enterprise-grade skills for AI coding agents adhering to the open agentskills.io standard, with primary support for Claude Code and Codex.

The skills serve as specialized instruction sets and tool wrappers that teach coding agents how to interact with NVIDIA's deep software ecosystem. It spans Physical AI and Isaac robotics simulations, CUDA-X accelerated computing libraries, TensorRT-LLM model optimization, NeMo microservices, and enterprise NVIDIA AI Blueprints, enabling agents to configure complex GPU pipelines without stumbling over obscure driver flags or syntax errors.

Daily synchronization pipelines mirror skills directly from internal product repositories, ensuring that agents always reference validated, up-to-date APIs. However, developers adopting these skills must have access to NVIDIA GPU hardware to execute CUDA and robotics runtimes, and troubleshooting deep kernel optimizations still demands sound GPU computing fundamentals.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool. Data as of 2026-11-12.

License
Apache-2.0
Latest release
No GitHub releases
Last commit
2026-11-11
Contributors
98
Forks
511
Open issues and PRs
16
First published
2026-02-25

Source: GitHub, npm and PyPI public APIs.

#agent-skills#agentic-ai#ai-agents#claude-code#codex#coding-agents#cuda#dev-tooling
$ Install
$ npx skills add nvidia/skills
↗ Visit site★ GitHub
01

Features

01Covers Physical AI, robotics, simulation, CUDA-X libraries and AI Blueprints
02Installs with `npx skills add` for all or one skill
03Targets a specific agent with `--agent`, such as Claude Code or Codex
04Skills are mirrored daily from product repos by an automated sync
02

Why choose it

+Official NVIDIA-verified instructions ensuring coding agents use correct, up-to-date GPU and robotics APIs.
+Adheres strictly to the open agentskills.io standard for seamless compatibility with Claude Code and Codex.
+Broad coverage spanning CUDA-X computing, Isaac robotics, TensorRT-LLM, and enterprise AI Blueprints.
+Automated daily sync pipeline guarantees alignment with upstream NVIDIA enterprise SDK releases.
03

Trade-offs

!Execution of underlying code requires compatible NVIDIA GPU hardware with appropriate CUDA drivers installed.
!Complex robotics simulation workflows require substantial local VRAM and high-performance computing setups.
!Instruction sets guide code generation but do not replace deep human domain expertise in GPU kernel performance tuning.
!Certain enterprise blueprints assume connectivity to proprietary NVIDIA AI Enterprise cloud microservices.
04

Maintenance check

ActiveChecked: 2026-11-12
Last commit
2026-11-11
Commits, last 90 days
100+
Releases, last 90 days
0
Latest release
—
Open issues
16
Recent issues with at least one comment
1/2
Commits per week, last 12 weeks
Commits per week, last 12 weeks: 18, 57, 18, 33, 26, 45, 15, 33, 52, 38, 56, 19

Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.

Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.

Most discussed open issues

  • #298 Feature Request: Bundle-level shared assets: let a plugin/component carry assets shared across many skills5 comments · 2026-06-26
  • #562 Group catalog skills by product to support product-scoped installation1 comments · 2026-09-14
  • #216 Skills dropped from sync — 59 sig drift, 1 sig mismatch, 3 sig missing1 comments · 2026-06-01
  • #751 Skills dropped from sync — 53 sig drift, 5 sig mismatch, 2 sig missing0 comments · 2026-10-22
  • #563 Sync failure tracker — 8 component(s) currently failing (2026-11-12)0 comments · 2026-09-15

Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.

05

Compatibility

Claude Code
Supported
Verified via docs
Codex
Supported
Verified via docs
Cursor
Supported
Verified via docs
06

Quick start

1
$ npx skills add nvidia/skills
07

Use cases

↳Asking an agent to solve a linear programming problem with cuOpt
↳Giving coding agents guidance for NVIDIA robotics and simulation stacks
↳Building RAG with NVIDIA blueprints through an agent
08

Alternatives

⚡Top VS:
VStaste-skill★117.7k→VSdarwin-skill★7.0k→VSpaperthin★1.5k→VSskills★1.5k→
taste-skill logo
taste-skill★ 117.7k
Agent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
VSCompare→
darwin-skill logo
darwin-skill★ 7.0k
达尔文.skill —— 一个让你的Skill无限进化的系统:评估→改进→测试→保留或回滚 | Autoresearch-inspired autonomous skill optimization for Claude Code. Evaluate, improve, test, keep or revert.
VSCompare→
paperthin logo
paperthin★ 1.5k
The skillbook for recursive self-improvement (RSI). NVIDIA's latest proof, unpacked with timeless engineering. Distilled from the OpenAI harness engineering hackathon winner's zero-touch method.
VSCompare→
skills logo
skills★ 1.5k
Reusable AI agent skills for software factories: explore ideas, write specs, implement, review, and run autonomous research. Works with Claude Code, Codex, and other skill-compatible agents.
VSCompare→
diagram-design logo
diagram-design★ 67.9k
Editorial diagram design for Claude Code, Codex, GitHub Copilot, Cursor, Factory Droid, Qoder, and Pi. 45 diagram types. Self-contained HTML + SVG. No shadows. No Mermaid slop.
VSCompare→
skills logo
skills★ 57.8k
Skills for Designers and Engineers.
VSCompare→
pstack logo
pstack★ 4.4k
Use pstack with claude, codex, pi - Skills and principles for rigorous AI-assisted engineering
VSCompare→
SkillEvaluator logo
SkillEvaluator★ 3.3k
Multi-tier framework for evaluating AI agent skills with quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation that measures how skills affect agent behavior.
VSCompare→
See all alternatives →

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs04Maintenance check05Compatibility06Quick start07Use cases08Alternatives
NVIDIA /
skills
↗

Official NVIDIA-verified Agent Skills for Claude Code, Codex, and coding agents covering CUDA, robotics, and AI Blueprints

4.0k511Apache-2.0
Last commitToday
StatusActive
CategoryDev Tooling
Contributors98
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.