NVIDIA

Cosmos Data Factory — Data skill for Claude Code

Data community

Synthetic Data Generation Workflows for Physical AI.

How to install Cosmos Data Factory

This entry records only its repository, not the path inside it, so there is no exact command to give. Open NVIDIA/physical-ai-data-factory and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Cosmos Data Factory does

Synthetic Data Generation Workflows for Physical AI.

Alternatives in Data

  • Bitbucket Data Center 71.9k ★
  • Financial Services — Reference agents, skills, and data connectors for the financial-services workflows we see most — investment ba 34.5k ★
  • Data Contract — This document defines which files belong to the system (auto-updatable) and which belong to the user (never to 32.5k ★

README

Cosmos Data Factory

NVIDIA Cosmos Data Factory (CDF) is a set of agent-driven blueprints and composable skills that turn real data into a large, labeled training dataset for physical AI. These blueprints and skills curate, enrich, label, augment and generate data on your own GPUs on premises or on the cloud of your choice.

Reference blueprints

Reference blueprint What it does Use cases
**Autonomous Vehicle Dataset Generation** Finds driving clips similar to an example video with Cosmos Curator and Cosmos Dataset Search, and optionally generates Cosmos3 variants guided by RGB or world scenario map controls. Improving autonomous driving with more examples of specific driving scenarios
**Defect Image Generation** Turns PCB, metal and glass component references into simulated SMT component images, then augments them and adds generated defects. Training inspection models without enough real defect examples
**Event Video Generation** Turns one seed image into annotated anomaly videos of events such as fires, falling, shoplifting and fighting. Getting data for rare or new events
**Image Attribute Augmentation** Generates variations of person object crops with controlled changes in clothing, colors and footwear. Improving person search and re-identification accuracy
[Video Data Augmentation](https://github.com/NVIDIA/cdf-orchestration/