Guide users through Cosmos3 supervised fine-tuning (SFT) post-training: preparing the example dataset and Wan2.2 VAE, converting the base checkpoint to DCP, launching distributed training (paired…
$ npx skills add NVIDIA/cosmos-framework --skill cosmos3-post-training -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install NVIDIA/cosmos-framework cosmos3-post-training --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/NVIDIA/cosmos-framework.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/cosmos3-post-training .claude/skills/cosmos3-post-training && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "cosmos3-post-training" agent skill from https://github.com/NVIDIA/cosmos-framework/tree/main/.agents/skills/cosmos3-post-training into .claude/skills/cosmos3-post-training/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cosmos3-post-training", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/NVIDIA/cosmos-framework/tree/main/.agents/skills/cosmos3-post-trainingType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add NVIDIA/cosmos-framework --skill cosmos3-post-training -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install NVIDIA/cosmos-framework cosmos3-post-training --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NVIDIA/cosmos-framework.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/cosmos3-post-training .agents/skills/cosmos3-post-training && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "cosmos3-post-training" agent skill from https://github.com/NVIDIA/cosmos-framework/tree/main/.agents/skills/cosmos3-post-training into .agents/skills/cosmos3-post-training/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cosmos3-post-training", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add NVIDIA/cosmos-framework --skill cosmos3-post-training -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install NVIDIA/cosmos-framework cosmos3-post-training --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NVIDIA/cosmos-framework.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/cosmos3-post-training .cursor/skills/cosmos3-post-training && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "cosmos3-post-training" agent skill from https://github.com/NVIDIA/cosmos-framework/tree/main/.agents/skills/cosmos3-post-training into .cursor/skills/cosmos3-post-training/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cosmos3-post-training", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/NVIDIA/cosmos-framework.git --path .agents/skills/cosmos3-post-training--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add NVIDIA/cosmos-framework --skill cosmos3-post-training -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install NVIDIA/cosmos-framework cosmos3-post-training --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NVIDIA/cosmos-framework.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/cosmos3-post-training .gemini/skills/cosmos3-post-training && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "cosmos3-post-training" agent skill from https://github.com/NVIDIA/cosmos-framework/tree/main/.agents/skills/cosmos3-post-training into .gemini/skills/cosmos3-post-training/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cosmos3-post-training", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install NVIDIA/cosmos-framework cosmos3-post-trainingInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add NVIDIA/cosmos-framework --skill cosmos3-post-training -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/NVIDIA/cosmos-framework.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/cosmos3-post-training .github/skills/cosmos3-post-training && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "cosmos3-post-training" agent skill from https://github.com/NVIDIA/cosmos-framework/tree/main/.agents/skills/cosmos3-post-training into .github/skills/cosmos3-post-training/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cosmos3-post-training", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add NVIDIA/cosmos-framework --skill cosmos3-post-training -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install NVIDIA/cosmos-framework cosmos3-post-training --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/NVIDIA/cosmos-framework.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/cosmos3-post-training .opencode/skills/cosmos3-post-training && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "cosmos3-post-training" agent skill from https://github.com/NVIDIA/cosmos-framework/tree/main/.agents/skills/cosmos3-post-training into .opencode/skills/cosmos3-post-training/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cosmos3-post-training", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
cosmos3-post-trainingGuide users through Cosmos3 supervised fine-tuning (SFT) post-training: preparing the example dataset and Wan2.2 VAE, converting the base checkpoint to DCP, launching distributed training (paired…
Cosmos3 Post Training is an agent skill from NVIDIA/cosmos-framework, published by the product's own GitHub organization. Guide users through Cosmos3 supervised fine-tuning (SFT) post-training: preparing the example dataset and Wan2.2 VAE, converting the base checkpoint to DCP, launching distributed training (paired launch shell recommended, raw torchrun as an alternative), running T2V/I2V/V2V inference with the trained DCP checkpoint, and optionally exporting it to Hugging Face safetensors. Use when the user asks how to post-train Cosmos3, fine-tune on a custom video dataset, export a trained checkpoint, or invoke one of the recipe…
Its SKILL.md is about 2.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in AI & LLM Engineering, covering Fine-tuning, Model hubs and datasets and Deep learning. It works with Hugging Face and CUDA. The repository describes itself as: Our inference and training framework to run on the Cosmos Models.
8 steps, taken from the first numbered list in SKILL.md.
Read from SKILL.md and the folder at commit 8aca062. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Shell commands in SKILL.md call:
uvpythonuvxbashFrom the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md. Its commands use uv and uvx, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Names these keys or tokens, usually read from environment variables:
WANDB_API_KEYFrom names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Cosmos3 Post Training loads about 2.7k tokens when it runs. Until then it costs about 218 tokens; SKILL.md has 964 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 964 words (~2,742 tokens).
“All paths below are relative to the cosmos3 package root (../../../ from this skill file). All uv run / python / torchrun / bash commands should also be run from there.”
Just SKILL.md in .agents/skills/cosmos3-post-training of NVIDIA/cosmos-framework.
Open the folder on GitHubat commit 8aca062
Cosmos3 Post Training next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Cosmos3 Post Training this skillNVIDIA/cosmos-framework | 558 | — | ~2.7k | Automated safety check: Pass | Custom licence | |
| Megakernel OptimizationRightNow-AI/AutoMegaKernel | 148 | — | ~1.8k | Automated safety check: Pass | MIT | |
| Kermt FinetuneNVIDIA/skills | 3.5k | 1 repos | ~4.1k | Automated safety check: Pass | Apache-2.0 | |
| Discover MLrand/cc-polymath | 181 | 1 repos | ~574 | Automated safety check: Pass | MIT | |
| Mamba State-Space ModelsOrchestra-Research/AI-Research-SKILLs | 13k | 3 repos | ~1.8k | Automated safety check: Pass | MIT | |
| Hugging Face LLM Trainerhuggingface/skills | 11k | 3 repos | ~7.2k | Automated safety check: Pass | Apache-2.0 |
RightNow-AI/AutoMegaKernel
A skill your agent uses when optimizing or generating a CUDA megakernel for a HuggingFace Llama-family model with AutoMegaKernel (AMK), drives the correctness-gated propose - eval - keep/revert loop…
NVIDIA/skills
Finetune a pretrained KERMT encoder on a labeled CSV. An agent skill from NVIDIA/skills.
rand/cc-polymath
Automatically discover machine learning and AI skills when working with machine learning, PyTorch, training, inference, RAG, embeddings, fine-tuning, LLM, DSPy, HuggingFace, or diffusion models.
Orchestra-Research/AI-Research-SKILLs
Guide to using Mamba selective state-space models for linear-time sequence modeling, from the Mamba block and pretrained checkpoints to Mamba-2 and speed comparisons.
huggingface/skills
Trains or fine-tunes language and vision models with TRL or Unsloth on Hugging Face Jobs cloud GPUs, then converts the results to GGUF.
awslabs/agent-plugins
Generates code that transforms datasets between ML schemas for model training or evaluation.
NVIDIA/cosmos-framework
Navigate the Cosmos3 package codebase to find where parameters, configs, defaults, scripts, and documentation live.
NVIDIA/cosmos-framework
Diagnose and fix Cosmos3 environment, installation, and runtime errors.
NVIDIA/cosmos-framework
Guide users through running Cosmos3 inference — offline batch generation, online serving with Ray and Gradio, parallelism options, input formats, sampling parameters, and prompt upsampling.
NVIDIA/cosmos-framework
Guide users through Cosmos3 installation, environment setup, checkpoint downloading, and verification.
Works with
Categories
Guide users through Cosmos3 supervised fine-tuning (SFT) post-training: preparing the example dataset and Wan2.2 VAE, converting the base checkpoint to DCP, launching distributed training (paired…. Cosmos3 Post Training is an agent skill from NVIDIA/cosmos-framework, published by the product's own GitHub organization.2 VAE, converting the base checkpoint to DCP, launching distributed training (paired launch shell recommended, raw torchrun as an alternative), running T2V/I2V/V2V inference with the trained DCP checkpoint, and optionally exporting it to Hugging Face safetensors.
Cosmos3 Post Training fits situations like: the user asks how to post-train Cosmos3; fine-tune on a custom video dataset; export a trained checkpoint; invoke one of the recipe launch shells (launchsftvisionnano.sh.
Run `npx skills add NVIDIA/cosmos-framework --skill cosmos3-post-training -a claude-code`. Or copy the skill folder (.agents/skills/cosmos3-post-training in NVIDIA/cosmos-framework) into .claude/skills/cosmos3-post-training in your project. Claude Code loads it when a task matches its description.
Run `npx skills add NVIDIA/cosmos-framework --skill cosmos3-post-training -a codex`. Or copy the skill folder (.agents/skills/cosmos3-post-training in NVIDIA/cosmos-framework) into .agents/skills/cosmos3-post-training in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add NVIDIA/cosmos-framework --skill cosmos3-post-training -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/cosmos3-post-training, .gemini/skills/cosmos3-post-training, .github/skills/cosmos3-post-training and .opencode/skills/cosmos3-post-training in your project.
Going by SKILL.md and its folder, Cosmos3 Post Training needs the command-line tools its instructions call (uv, python, uvx and bash) and credentials named WANDB_API_KEY. Our summary lists: Python 3; A credential in WANDB_API_KEY.
SKILL.md contains no URLs. Its commands use uv and uvx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Cosmos3 Post Training has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.
About 2.7k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Cosmos3 Post Training: Megakernel Optimization (RightNow-AI/AutoMegaKernel, 148 stars), Kermt Finetune (NVIDIA/skills, 3.5k stars), Discover ML (rand/cc-polymath, 181 stars) and Mamba State-Space Models (Orchestra-Research/AI-Research-SKILLs, 13k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
NVIDIA (a GitHub organization, an official publisher) maintains it in NVIDIA/cosmos-framework, which has 558 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on October 8, 2026.
Source: NVIDIA/cosmos-framework on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.
