Machine Learning Engineering Open Book is a technical collection of methods, tools, scripts, and instructions for training, fine-tuning, and running large language and multimodal models. It is intended for engineers and operators working with model training and inference. The catalogue contains one skill for working with this material.
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 stas00/ml-engineering --skill evaluate-clustergit clone --depth 1 https://github.com/stas00/ml-engineeringWrote 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/stas00/ml-engineering/evaluate-cluster)<a href="https://agentmods.dev/skills/stas00/ml-engineering/evaluate-cluster"><img src="https://agentmods.dev/badge/skills/stas00/ml-engineering/evaluate-cluster/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/stas00/ml-engineering/evaluate-cluster"><img src="https://agentmods.dev/badge/skills/stas00/ml-engineering/evaluate-cluster.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00144 | $0.22650 |
| Opus 5.5 | $0.00058 | $0.09060 |
| Sonnet 5.5 | $0.00029 | $0.04530 |
| Haiku 4.5 | $0.00014 | $0.02265 |
Grade B, and why
evaluate-cluster scanned grade B with 2 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 7d 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
sudo apt-get update Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Before doing any eval work, `curl` the raw URL. If it is not byte-identical to the file you are reading, **stop following this copy** and follow the GitHub file instead (fetch it, then continue from that text). If GitHub The source is not reproduced here
Licensed CC-BY-SA-4.0
The repository is licensed CC-BY-SA-4.0, which this catalogue does not treat as permission to reproduce the file. Read it at the source.
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.
- 7d ago Changed · +29 lines · +7 tokens per session 35c753ec9860
- 10d ago Changed 6fc336d0f6af
- 14d ago First seen · 710 lines · 137 tokens per session scan B 713653e38b24
evaluate-cluster is a skill published in the GitHub repository stas00/ml-engineering (19,532 stars, last pushed today), licensed CC-BY-SA-4.0. It adds 144 tokens to every session and 22,650 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-02.
Other skills, from other repositories
ml-training-recipes
Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics. Covers training loops, optimizer selection (AdamW, Muon), LR scheduling, mixed precision, debugging, and systematic experimentation. Use when training or fine-tuning…
physicsnemo-discover
Official NVIDIA-authored guidance for navigating PhysicsNeMo — pick the model, datapipe, or example for a SciML/AI4Science task (surrogates, forecasting, downscaling, physics-informed, inverse, generative). Points at existing files via live repo search; never writes code. Do NOT use for installation or environment…
dnanexus-integration
DNAnexus cloud genomics platform. Build apps/applets, manage data (upload/download), dxpy Python SDK, run workflows, FASTQ/BAM/VCF, for genomics pipeline development and execution.
latchbio-integration
Latch platform for bioinformatics workflows. Build pipelines with Latch SDK, @workflow/@task decorators, deploy serverless workflows, LatchFile/LatchDir, Nextflow/Snakemake integration.
compute
Chooses and runs compute for research work, local shell or kernel versus a detached computejob on this machine, a saved SSH or Slurm host, or Modal GPUs, with the target discovered from what is actually configured, the run sized and priced before dispatch, outputs declared, metrics tracked, and long work handed to the…
remote-compute-ops
Operate GPU and remote compute across RunPod (Pods and Serverless), Massed Compute VMs, and owned or virtual remote servers through existing bridges, SSH sessions, MCP/API adapters, bounded polling, cost controls, and resumable lifecycle checks. Use when the user mentions RunPod, Massed Compute, a remote…