skills
NVIDIA's AI agent skills help AI optimize the use of NVIDIA software.
- Type
- Agent Skill
- Open source
- Yes
- GitHub Stars
- ★ 1.8k
- Source
- skill-github
- Repository
- github.com/NVIDIA/skills
Overview
NVIDIA Agent Skills are a set of portable instruction sets that teach AI agents how to optimally use NVIDIA software, including CUDA-X libraries, AI Blueprints, and platform tools. These skills are installed via the `skills` CLI and automatically loaded when the AI agent encounters relevant tasks. For example, you can instruct the agent to solve linear programming problems using cuOpt. These skills are ideal for developers and researchers requiring high-performance computing and optimization.
Capabilities
- ▪Guidance on using CUDA-X libraries
- ▪AI Blueprint deployment
- ▪cuDF GPU DataFrame acceleration
- ▪cuFOLIO optimization
- ▪cuOpt optimization
- ▪cuPyNumeric parallel I/O
Use cases
Setup
npx skills add nvidia/skills --skill <skill_name> --agent <agent_name>
This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.
FAQ
How do I install a specific skill?
Use the command `npx skills add nvidia/skills --skill <skill_name> --agent <agent_name>`.
Which AI agents are supported?
Supports Claude Code, Codex, Cursor, Kiro, and others.
Related skills
gemini-cli
Gemini CLI is an open-source AI agent that brings Gemini's powerful capabilities directly into the terminal.
private-gpt
Provides a complete API layer for local models, supporting RAG, skills, tools, and more.
planning-with-files
File-based persistent planning skill, ideal for AI coding agents and long-running tasks.
scientific-agent-skills
Transform any AI agent into a scientific assistant with 147 ready-to-use research skills.
susi_alexa_skill
A skill that enables question-and-answer interactions between Alexa and Susi AI.
claude-context
Provides code search capabilities for Claude Code, turning the entire codebase into context.