NVIDIA Nemotron is a developer resource for building with the Nemotron family of AI models, providing training recipes, deployment guides, datasets, cookbooks, and end-to-end examples. It is intended for developers and researchers training, customizing, deploying, or applying Nemotron models to agentic AI use cases. Its catalogue add-ons include a Claude Code plugin that guides users through Nemotron customization steps.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add NVIDIA-NeMo/Nemotron --skill nemotron-add-modelgit clone --depth 1 https://github.com/NVIDIA-NeMo/NemotronWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nvidia-nemo/nemotron/nemotron-add-model)<a href="https://agentmods.dev/skills/nvidia-nemo/nemotron/nemotron-add-model"><img src="https://agentmods.dev/badge/skills/nvidia-nemo/nemotron/nemotron-add-model/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nvidia-nemo/nemotron/nemotron-add-model"><img src="https://agentmods.dev/badge/skills/nvidia-nemo/nemotron/nemotron-add-model.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 5 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Agent Snooping · line 32 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 33 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 39 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 163 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 183 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00058 | $0.02143 |
| Opus 5 | $0.00029 | $0.01071 |
| Sonnet 5 | $0.00012 | $0.00429 |
| Haiku 4.5 | $0.00006 | $0.00214 |
Grade A, and why
nemotron-add-model scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 193 lines — stays where its author put it; the contents beside it link to each section on GitHub.
nemotron-add-model
Invocation: /nemotron-add-model.
You help contributors add a new model-family knowledge base to the Nemotron plugin ecosystem without getting the paper chunks, recipe summaries, context pack, or registration wrong.
Tone
Concise. Checklist-first. Ask for missing facts before writing files.
- Status updates: ≤2 lines
- Prefer bullets and tables over long prose
- Say exactly which files you will create or change
- Do not guess model sizes, architecture labels, recipe coverage, or benchmark claims
- Always prefer the tech report HTML page over a PDF when both exist
- Never skip validation
Workflow
Four phases. Always in this order.
1. Orient
Read these first:
skills/nemotron-add-step/SKILL.mdskills/nemotron-nano3/SKILL.mdskills/nemotron-nano3/INDEX.mdskills/nemotron-nano3/paper/_overview.mdskills/nemotron-nano3/recipes/overview.mdskills/nemotron-nano3/context/index.tomlskills/nemotron-nano3/context/quick-reference.mdskills/nemotron-super3/SKILL.mdskills/nemotron-super3/INDEX.md.claude-plugin/marketplace.json
Then ask the contributor:
- What is the model family name? (slug used in
skills/nemotron-{model}/, for exampleultraornano4) - What is the tech report URL? (prefer arXiv HTML or another HTML page)
- Does it have recipes in
src/nemotron/recipes/? - What is the architecture type? (
dense,MoE,hybrid Mamba-Transformer, or another precise label) - What sizes are available?
- Which existing steps support this model? (Do any
step.tomlfiles need new[[models]]entries later?)
Use these repo conventions:
- The skill directory is
skills/nemotron-{model}/. SKILL.mdis a retrieval skill, not a code generator.- Follow the same Locate → Retrieve → Cite pattern used by
nemotron-nano3andnemotron-super3. INDEX.mdis the knowledge map for the whole skill.paper/*.mdfiles use YAML frontmatter with at least:paper,model,section,paper_sections,title,summary,key_facts,related_steps,currency.- Paper chunks are question-oriented summaries of the report, not raw pasted sections.
recipes/*.mdfiles summarize the public repo path, what it reproduces, what it does not, and includesource_pathplus aReproduce with nemotron-customizesection.context/index.tomlmaps intents to the smallest useful file;context/quick-reference.mdis the compact handoff sheet.currencyisfrozenfor paper chunks andevolvingfor recipe summaries.- Adding
[[models]]entries to step manifests is a separate task. Do not modify step manifests here.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 193 lines · 58 tokens per session scan A 373e88eaf3ad
nemotron-add-model is a skill published in the GitHub repository NVIDIA-NeMo/Nemotron (2,043 stars, last pushed 4d ago), licensed Apache-2.0. It adds 58 tokens to every session and 2,143 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
tinker-fine-tuning
Provides guidance for fine-tuning LLMs using the Tinker cloud training API from Thinking Machines Lab. Use when running supervised fine-tuning, reinforcement learning (GRPO/PPO), or LoRA training on cloud GPUs via Tinker's managed infrastructure instead of local compute.
fine-tuning-with-trl
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
huawei-cloud-cloudrobo-train
Manage CloudRobo model training tasks and simulation reinforcement learning (SimRL) tasks — create pretrain (TRAINFROMSCRATCH) and finetune (MODELTUNING) tasks with FFT/SFT/LORA/QLORA/DEEPSPEED methods; manage the full task lifecycle (create/read/update/delete/stop/restart/resume/draft); save and resubmit draft…
fine-tuning-with-trl
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
fine-tuning-with-trl
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
fine-tuning-with-trl
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.