agentmods.dev

cosmos-framework: Skill for Claude Code

.agents/skills/cosmos3-qwen3-vl-robovqa-sft/SKILL.md

cosmos3-qwen3-vl-robovqa-sft is a skill for Claude Code, Codex from NVIDIA/cosmos-framework. It costs 192 tokens per session (4,364 once invoked), scanned B, original, no licence file.

A procedure for fine-tuning Qwen3-VL-8B-Instruct, a model that understands both images and text, on robot-video questions using the Cosmos Framework.

In plain words
What is it for?
Use it to train the model on RoboVQA, a dataset of questions about robot videos, on one GPU and compare its scores before and after training.
Why use it?
It provides a defined training and scoring path instead of requiring you to assemble the fine-tuning steps yourself.

Skill for Claude CodeCodex ✓ vendor

Which agent this was written for is unclear — body not stored (licence); the path alone says nothing. Also seen: installed under .agents/ (shared by several agents).

This is NVIDIA/cosmos-framework's own configuration. It tells Claude Code and Codex how to work on cosmos-framework itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything cosmos-framework configures →

not rated 618repo +29 today B scan 192 tokens custom licence
View source ↗ NVIDIA/cosmos-framework nvidia.github.io
Reuse

Borrowing it

Nothing to install: this file belongs to NVIDIA/cosmos-framework. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/NVIDIA/cosmos-framework/main/.agents/skills/cosmos3-qwen3-vl-robovqa-sft/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/NVIDIA/cosmos-framework

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.

agentmods badge for cosmos3-qwen3-vl-robovqa-sft

README.md
[![agentmods](https://agentmods.dev/badge/skills/nvidia/cosmos-framework/cosmos3-qwen3-vl-robovqa-sft/github.svg)](https://agentmods.dev/skills/nvidia/cosmos-framework/cosmos3-qwen3-vl-robovqa-sft)
Your own site
<a href="https://agentmods.dev/skills/nvidia/cosmos-framework/cosmos3-qwen3-vl-robovqa-sft"><img src="https://agentmods.dev/badge/skills/nvidia/cosmos-framework/cosmos3-qwen3-vl-robovqa-sft/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.

agentmods 80×15 button for cosmos3-qwen3-vl-robovqa-sft

Your own site · 80×15
<a href="https://agentmods.dev/skills/nvidia/cosmos-framework/cosmos3-qwen3-vl-robovqa-sft"><img src="https://agentmods.dev/badge/skills/nvidia/cosmos-framework/cosmos3-qwen3-vl-robovqa-sft.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 192 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,364 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin unknown 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00192 $0.04364
Opus 5.5 $0.00077 $0.01746
Sonnet 5.5 $0.00038 $0.00873
Haiku 4.5 $0.00019 $0.00436

Measured 6d ago against content hash 94203a43fe64, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

cosmos3-qwen3-vl-robovqa-sft scanned grade B with 1 finding 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 6d ago.

The scan reads SKILL.md. This mod also ships 3 executable files (scripts/eval_robovqa.py, scripts/prepare_robovqa.py, scripts/run.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

Four stages: the host check (CUDA 13 driver, `git`, `ffmpeg`, `g++`, uv; run as root it installs the missing tools
.agents/skills/cosmos3-qwen3-vl-robovqa-sft/SKILL.md · 248 lines

The source is not reproduced here

A licence we could not identify

The repository carries a LICENSE file, but it is custom or dual enough that GitHub cannot name it and neither can this catalogue. Unknown terms are not permission, so the body is not copied here. Read the licence at the source and decide for yourself.

Read it on GitHub

Files

What ships with it

13 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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.

  1. 6d ago First seen · 248 lines · 192 tokens per session scan B 94203a43fe64

Subscribe to this mod's changes

cosmos3-qwen3-vl-robovqa-sft is a skill published in the GitHub repository NVIDIA/cosmos-framework (618 stars, last pushed today), with no licence file. It adds 192 tokens to every session and 4,364 once invoked, about $0.0008 per session on Opus 5.5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-31.

Data and AI
Related

cosmos3-qwen3-vl-robovqa-sft compared with skills from other repositories

cosmos3-qwen3-vl-robovqa-sft next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
cosmos3-qwen3-vl-robovqa-sftNVIDIA/cosmos-frameworkSkill618todaynoneB192Claude Code, Codex
agent-platform-model-registrygoogle/skillsSkill22kyesterdayApache-2.0A60Claude Code, Codex
foundry-config-setupmicrosoft/agent-frameworkSkill15kyesterdayMITA65Claude Code, Codex
google-cloud-solution-agentic-analytics-spark-knowledge-cataloggoogle/skillsSkill22kyesterdayApache-2.0A132Claude Code, Codex
training-checkwanshuiyin/Auto-claude-code-research-in-sleepSkill19k3d agoMITA35Claude Code
nemo-automodel-launcher-configNVIDIA/skillsSkill3.9kyesterdayApache-2.0A30Claude Code
finetuningmicrosoft/skillsSkill3.2kyesterdayMITA132Claude Code