NVIDIA-NeMoOfficial·60 Agent Skills Included

nemo-platform

Build, evaluate, harden, and optimize production LLM agents

Builds, deploys, evaluates, and secures LLM agents on NVIDIA's NeMo Platform through one CLI, Python SDK, and web UI. Covers agent scaffolding, guardrails, red-teaming, PII anonymization, synthetic data, fine-tuning, telemetry analysis, and cost optimization. Eliminates manual YAML authoring, ad-hoc eval scripts, and guesswork by routing each task to a verified, step-by-step workflow.

58|16

↗

npx skills add NVIDIA-NeMo/nemo-platform --all -g -y

Available:

Directs the agent to route every NeMo Platform request to the repo's own SKILL.md files, follow their exact commands and verification steps, and never substitute generic skills or improvised code.

All Skills in This Repository (60)

Pure Emerald Level Indicators

📦 In Repo

NVIDIA-NeMo

langchain-framework

Guide building LLM-powered agents with LangChain, LangGraph, and Deep Agents.

nooa

Develop and optimize agents using the NVIDIA-labs OO Agents framework.

terminator

Analyze textual diagnostic reports to detect qualitative plateaus in optimization workflows.

guardrails-plugin

Implements input/output guardrailing for chat-completions traffic in NeMo Inference Gateway.

deploy-sandbox

Deploy NAT agents as isolated OpenShell sandboxes with Landlock filesystem protection.

example-hello

Validate plugin-provided skill registration and loading within the NeMo platform.

example-debug

Verify plugin skill directory installations and companion resource accessibility in NeMo.

refresh-data-designer-skill

Synchronize vendored Data Designer skill bundles and adapt CLI commands for NeMo Platform.

nemo-customizer

Orchestrate SFT and LoRA fine-tuning workflows for LLMs on the NeMo Platform.

anonymizer

Detect and replace sensitive personal information in CSV or Parquet datasets.

auditor

Manage audit configurations, targets, and jobs for LLM inference endpoint vulnerability scanning.

safe-synthesizer

Generate privacy-preserving synthetic tabular data using NeMo Platform workflows.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install nemo-platform?▼

Run `npx skills add NVIDIA-NeMo/nemo-platform --all -g -y` in your terminal to install all skills in this suite globally.

What can I do with NeMo Platform skills?▼

You can design, build, deploy, evaluate, secure, and optimize LLM agents, plus fine-tune models, generate synthetic data, redact PII, and analyze production traces.

Does nemo-platform work with Claude Code and Cursor?▼

Yes. The skills follow the standard SKILL.md format and install into Claude Code, Cursor, Codex, and OpenCode via the `nemo skills install` command.

How do I add guardrails to my agent?▼

The guardrails skills walk you through creating a content-safety config, validating it with test messages, and attaching it to a VirtualModel so unsafe input and output get blocked.

Can I use these skills without deep ML expertise?▼

Yes. Each skill gives your coding agent exact commands, templates, and verification steps, so you describe the goal in plain English and the agent executes the workflow.