What is cosmos3-setup?

A setup workflow for Cosmos3, an NVIDIA framework and model-checkpoint system, using NPA tools. It checks access to the source and Hugging Face, verifies credentials and settings, and stages files for later use.

Use it to check access, fetch the public framework and checkpoints, prepare a temporary runtime cache, and choose the appropriate NPA workflow.

Cosmos3 setup involves several possible sources, credentials, caches, and hardware or installation choices. This workflow helps determine the right preparation path before running inference, which means using the model to produce results.

cosmos3-setup — Skill for Claude Code and Codex by nebius

Install

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.

Any agent

npx skills add nebius/nebius-physical-ai --skill cosmos3-setup

Clone the repo

git clone --depth 1 https://github.com/nebius/nebius-physical-ai

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/cosmos3-setup/github.svg)](https://agentmods.dev/skills/nebius/nebius-physical-ai/cosmos3-setup)

Your own site

<a href="https://agentmods.dev/skills/nebius/nebius-physical-ai/cosmos3-setup"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/cosmos3-setup/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/nebius/nebius-physical-ai/cosmos3-setup"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/cosmos3-setup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 47 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,002 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector warn 7 Sept 2026

SkillSpector: 1 finding, up to high

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 →

high Tool Misuse · line 64
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

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	$0.00047	$0.01002
Opus 5	$0.00023	$0.00501
Sonnet 5	$0.00009	$0.00200
Haiku 4.5	$0.00005	$0.00100

Measured 2d ago against content hash 17ec8d33b54c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

cosmos3-setup 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 2d 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.

skills/atomic/cosmos3-setup/SKILL.md · 106 lines

How it starts

The opening of the file, as written. The whole thing — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Cosmos3 Setup

Source And Attribution

Adapted from NVIDIA cosmos-framework skills/atomic/cosmos3-setup/SKILL.md.

When To Use

Use this skill when the task is about Cosmos3 first-run setup, source checkout, checkpoint access, credential verification, CUDA or uv install choices, or selecting the right NPA Cosmos3 workflow. If the user is already failing with an ImportError, CUDA error, Docker runtime error, or checkpoint download failure, switch to skills/atomic/cosmos3-env-troubleshoot/SKILL.md.

NPA Boundary

Cosmos3 skills are agent skills. Do not add or invoke Cosmos skill-display subcommands; those are intentionally not part of the NPA human CLI surface.

Real NPA setup surfaces are:

npa workbench cosmos check for redacted source, Hugging Face, optional NGC, and parser-setting validation.
npa workbench cosmos fetch for source plus checkpoint staging in a temporary runtime cache.
workflows/testing/cosmos-fetch.yaml for a workflow fetch workflow that runs those real commands.
workflows/testing/cosmos3-text-to-image.yaml for the H100 text-to-image smoke workflow.

Setup Flow

Inspect the NPA integration before changing behavior:

rg -n "Cosmos3|cosmos3|NPA_COSMOS3" npa/src/npa npa/workflows npa/tests

Confirm access locally without downloading weights:
```
npa/.venv/bin/npa workbench cosmos check --output json
```
This command must redact model repo IDs, private source URLs, and token values in its output.
Stage source and checkpoint only when credentials are available:
```
npa/.venv/bin/npa workbench cosmos fetch --output json
```
Use --skip-checkpoint for source-only checks. Use --cache-dir for an ephemeral path, not a repository path.
For GPU inference, use the text-to-image SkyPilot workflow. The setup skill should point to that workflow; it should not create a skill launcher command.

Read the full file on GitHub · 106 lines

Changes

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.

2d ago Changed 17ec8d33b54c
10d ago First seen · 106 lines · 47 tokens per session scan A 75a58a5e19ce

Subscribe to this mod's changes

cosmos3-setup is a skill published in the GitHub repository nebius/nebius-physical-ai (27 stars, last pushed today), licensed Apache-2.0. It adds 47 tokens to every session and 1,002 once invoked, about $0.0002 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.

Data and AI

agent-platform-rag-engine-management

Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…

google/skills · 85 tokens

agent-platform-model-registry

Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.

google/skills · 60 tokens

foundry-config-setup

Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.

microsoft/agent-framework · 65 tokens

google-cloud-solution-agentic-analytics-spark-knowledge-catalog

Discovers requirements and generates guidance to design and deploy a governed, secure agentic-analytics solution for data that's distributed across Google Cloud, other cloud providers, or on-premises. Data that's outside Google Cloud (such as data from Databricks, Snowflake, Salesforce, SAP, or Oracle systems) is…

google/skills · 138 tokens

training-check

Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.

wanshuiyin/Auto-claude-code-research-in-sleep · 35 tokens

nemo-automodel-launcher-config

Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.

NVIDIA/skills · 30 tokens