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 nebius/nebius-physical-ai --skill cosmos3-inferencegit clone --depth 1 https://github.com/nebius/nebius-physical-aiWrote 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/nebius/nebius-physical-ai/cosmos3-inference)<a href="https://agentmods.dev/skills/nebius/nebius-physical-ai/cosmos3-inference"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/cosmos3-inference/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/nebius/nebius-physical-ai/cosmos3-inference"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/cosmos3-inference.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00040 | $0.01760 |
| Opus 5 | $0.00020 | $0.00880 |
| Sonnet 5 | $0.00008 | $0.00352 |
| Haiku 4.5 | $0.00004 | $0.00176 |
Grade A, and why
cosmos3-inference 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 4d 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 — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cosmos3 Inference
Source And Attribution
Adapted from NVIDIA cosmos-framework
skills/workflows/cosmos3-inference/SKILL.md.
Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. Used under OpenMDW-1.1.
See skills/LICENSE-NVIDIA-COSMOS3-OPENMDW-1.1 and
skills/NOTICE-NVIDIA-COSMOS3.
When To Use
Use this skill when the user wants to generate an image or video with Cosmos3,
change inference defaults, verify prompt handling, inspect guardrails behavior,
or connect NPA's Cosmos3 workflow to upstream Cosmos framework inference docs.
For environment errors, use
skills/atomic/cosmos3-env-troubleshoot/SKILL.md.
Real NPA Workflows
Choose the implementation matching the deployment: persistent Nano vLLM-Omni diffusion, containerized framework generation, or the text-to-image smoke.
Persistent Nano diffusion video
Read npa/deploy/cosmos3-nano-video/README.md for the CLI/SDK contracts, image,
frame mapping, measured results and reusable commands. nano-video-batch
generates a text-to-video segment followed by tail-conditioned continuations.
nano-video-augment transforms a complete source MP4 using official Canny edge
controls from every corresponding source interval. A continuation result alone
does not demonstrate visual augmentation or full source-motion conditioning.
The augmentation client accepts S3 input/output paths at 832×480 and 24 fps,
validates the complete media and publishes immutable artifacts with readback.
Use the supported sampling/control flags; there is no generic strength or
unchecked extra-parameter bag. Later windows use the preceding augmented
five-frame RGB tail for continuity and matching original-source edges for
structure. Preserve exact effective prompts, control provenance and all joins.
Keep actual source, augmented output and synchronized comparison clearly labeled.
nano-video-augment-recover retrieves the same request or retries artifact
publication without submitting generation. Preserve the original destination
and submission marker after interrupted generation or artifact retrieval; a
missing result is ambiguous. Never repeat GPU generation merely to retry an
upload. Validate visual change, identity, source motion, contact and temporal
joins separately from decode/hash checks, with a prior rubric and disclosed
agent/VLM sampling limits. The README's selected settings are measured examples,
not universal quality defaults.
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.
- 4d ago Changed · +19 lines · -2 tokens per session f8fddca3f992
- 9d ago First seen · 156 lines · 42 tokens per session scan A 1afcf2e47d48
cosmos3-inference is a skill published in the GitHub repository nebius/nebius-physical-ai (29 stars, last pushed today), licensed Apache-2.0. It adds 40 tokens to every session and 1,760 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 2026-09-03.
Other skills, from other repositories
llm-app-patterns
Production-ready patterns for building LLM applications. Covers RAG pipelines, agent architectures, prompt IDEs, and LLMOps monitoring. Use when designing AI applications, implementing RAG, building agents, or setting up LLM observability.
prompt-optimization
Improve a prompt on the evaluations workbench through a measured loop. Score the baseline first, then duplicate the target column, form a hypothesis from failing rows, edit the copy's prompt draft, run, compare pass rate and cost, and repeat until the numbers hold. Use when the user asks to optimize or improve a…
enhance-prompt
Transforms vague UI ideas into polished, Stitch-optimized prompts. Enhances specificity, adds UI/UX keywords, injects design system context, and structures output for better generation results.
prompt-engineer
Writes, refactors, and evaluates prompts for LLMs — generating optimized prompt templates, structured output schemas, evaluation rubrics, and test suites. Use when designing prompts for new LLM applications, refactoring existing prompts for better accuracy or token efficiency, implementing chain-of-thought or few-shot…
seedance-vocab-en
This skill should be used when an English Seedance 2.0 prompt needs clearer production wording, less generic prose, or precise vocabulary for camera, lighting, motion, VFX, audio, and constraints. Route blocked prompts through seedance-filter for context and boundary review.
ideogram4
Prompting patterns for Ideogram 4 text-to-image — best-in-class in-image text rendering and exact color/layout control via structured JSON captions. Use when generating images that need legible on-image text (title cards, thumbnails, logos, signage, CTAs), precise brand colors, or controlled spatial layout. Triggers…