Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/robium-ai/robiumnpx agentmods add skills/robium-ai/robium/isaac-labWrote 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/robium-ai/robium/isaac-lab)<a href="https://agentmods.dev/skills/robium-ai/robium/isaac-lab"><img src="https://agentmods.dev/badge/skills/robium-ai/robium/isaac-lab.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 6 findings, 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 Prompt Injection · line 70 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Prompt Injection · line 87 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Prompt Injection · line 106 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Prompt Injection · line 117 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Prompt Injection · line 180 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- medium Excessive Agency · line 213 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00128 | $0.04297 |
| Opus 5 | $0.00064 | $0.02148 |
| Sonnet 5 | $0.00026 | $0.00859 |
| Haiku 4.5 | $0.00013 | $0.00430 |
Grade A, and why
isaac-lab 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 3d 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 — 286 lines — stays where its author put it; the contents beside it link to each section on GitHub.
isaac-lab
The GPU-parallel RL/IL training layer of robium's NVIDIA stack, built on top
of an already-running Isaac Sim: prebuilt environments and tasks
(isaaclab_tasks), training entry points for several RL libraries, an
imitation-learning path for generating and training on simulated
demonstrations, and exporting a trained policy. Isaac Lab
(isaac-sim/IsaacLab, current release v3.0.0-beta2.patch1, published
2026-07-02, verified via direct fetch of the GitHub releases API this
session) is NVIDIA's own framework layered on Isaac Sim, not a separate
product to install independently. Its main branch's own installation docs
state support for Isaac Sim 4.5/5.0/5.1 and recommend the latest 5.1.0
release specifically (verified via direct fetch of the installation docs
on 2026-07-10); that may trail the newest Isaac Sim release the isaac-sim
skill cites, so confirm the current supported-version pairing before
installing rather than assuming the two always track together.
When to use this skill
- Running a prebuilt Isaac Lab task, training a policy with an RL library on top of a working Isaac Sim install, generating/training on simulated demonstrations, or exporting a trained policy for deployment.
- The trigger phrases in the description: 'isaac lab', 'GPU RL for robots', 'train in isaac', sim-to-real policy training in the NVIDIA stack.
- Cross-references: go to the sibling skill instead when the question is:
- Isaac Sim itself is not installed/working yet (GPU floor, container,
USD scene, robot/sensor import, ROS 2 bridge) →
isaac-sim. This skill assumes Isaac Sim is already running; it only adds the training layer on top. - Imitation learning on datasets recorded from a real robot (the
LeRobotDataset format,
lerobot-train/lerobot-record) →lerobot. This skill's own imitation-learning path (see Usage patterns) starts from demonstrations recorded inside Isaac Sim, not real hardware; that distinction is the actual boundary, not "imitation learning" as a category. - Whether to use the NVIDIA stack (Isaac Sim/Lab) at all vs. LeRobot's
own sim/eval tooling →
architectdecides this, gated on the GPU floor (see Platform gotchas). - Which simulator to use in general, before Isaac Sim is chosen →
simulation. - Deciding data-sourcing strategy (how much sim-generated vs. real
data a project needs) → the
dataumbrella skill. This skill only covers the mechanics of Isaac Lab's own demonstration-generation and training tools, not the sourcing decision.
- Isaac Sim itself is not installed/working yet (GPU floor, container,
USD scene, robot/sensor import, ROS 2 bridge) →
What ships with it
2 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.
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.
- 3d ago First seen · 286 lines · 128 tokens per session scan A a887c8298532
isaac-lab is a skill published in the GitHub repository robium-ai/robium (9 stars, last pushed 9d ago), licensed MIT. It adds 128 tokens to every session and 4,297 once invoked, about $0.0006 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
spark-environment-setup
Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.
spark-memory-thermal-ops
Manage unified memory and thermals during long-running ML jobs on NVIDIA DGX Spark. Use when planning memory headroom for a training run on GB10, when a job OOMs on unified memory, or when monitoring temperature and power during multi-hour training.
spark-training-gotchas
Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.
solr-semantic-search
To build Solr phrase-tagging semantic search: concept tagging, taxonomy, graph paths.
coding-agents-prompt-authoring
To author, adapt, review, and validate prompts (skills, agents, workflows, rules, etc.) with brief, contracts, and a validation pack.
solr-query
To build and debug Solr queries: eDisMax, block join, JSON facets, kNN, explain.