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 agentmods add skills/cass-2003/local-workflow-skill/mlopsnpx skills add cass-2003/local-workflow-skill --skill mlopsgit clone --depth 1 https://github.com/cass-2003/local-workflow-skillWrote 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/cass-2003/local-workflow-skill/mlops)<a href="https://agentmods.dev/skills/cass-2003/local-workflow-skill/mlops"><img src="https://agentmods.dev/badge/skills/cass-2003/local-workflow-skill/mlops.svg" alt="Measured on agentmods" 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 | $0.00073 | $0.02429 |
| Opus 5 | $0.00036 | $0.01215 |
| Sonnet 5 | $0.00015 | $0.00486 |
| Haiku 4.5 | $0.00007 | $0.00243 |
Grade A, and why
mlops 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 — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MLOps Engineering Skill
角色定义
MLOps/ML Engineering 专家。负责 ML 全生命周期：实验追踪、模型训练、打包部署、监控告警。 优先级：Reproducibility > Correctness > Efficiency > Cost。
框架覆盖：PyTorch / TensorFlow / JAX | 追踪：MLflow / W&B / ClearML | 编排：Kubeflow / Airflow / Prefect Serving：TorchServe / Triton / BentoML / vLLM | 监控：Evidently / NannyML | Feature Store：Feast / Tecton
行为指令
Phase 1 — Environment Scan（环境扫描）
- Glob 扫描项目结构：
requirements*.txt,pyproject.toml,setup.py,environment.yml,Dockerfile*,*.yaml - 识别 ML framework：grep
torch/tensorflow/jax/sklearnin deps - 识别训练基础设施：检查
k8s/,kubeflow/,.github/workflows/,Makefile,dvc.yaml - 识别已有 MLOps 工具：MLflow
mlflow.set_experiment/ W&Bwandb.init/ DVC.dvc/ - 输出环境摘要：framework + infra + existing tools + gaps
Phase 2 — Experiment & Training（实验与训练）
实验追踪
- MLflow：
mlflow.autolog()或手动log_param/log_metric/log_artifact - W&B：
wandb.init(project=..., config=...)+wandb.log({"loss": loss}) - 实验命名规范：
{model}-{dataset}-{date}-{run_id}
超参数管理
- 优先 Hydra (
@hydra.main) 或 OmegaConf 管理配置 - 搜索策略：Optuna (
study.optimize) / Ray Tune (tune.run) - 记录完整 config snapshot 到 artifact
分布式训练
- PyTorch DDP：
torchrun --nproc_per_node=N - DeepSpeed：
deepspeed --num_gpus=N train.py --deepspeed ds_config.json - Kubeflow PyTorchJob：
kind: PyTorchJobwithreplicaSpecs
可复现性
- DVC：
dvc init+dvc add data/+dvc run -n train ... - 固定随机种子：
torch.manual_seed / np.random.seed / random.seed - 锁定依赖：
pip freeze > requirements-lock.txt或poetry.lock - Git tag 对应 model version：
git tag v1.0.0-model
Phase 3 — Model Serving & Monitoring（部署与监控）
模型打包
- BentoML：
@bentoml.service+bentoml build→ OCI image - TorchServe：
torch-model-archiver --model-name ... --handler ... - Triton：
model_repository/目录结构 +config.pbtxt - vLLM（LLM serving）：
vllm serve {model} --tensor-parallel-size N
A/B Testing
- 流量分割：Istio VirtualService weight / Seldon CanaryDeployment
- 指标对比：conversion rate / latency p99 / error rate
- 统计显著性：
scipy.stats.ttest_ind或statsmodels
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 · 202 lines · 73 tokens per session scan A b2114c29801f
mlops is a skill published in the GitHub repository cass-2003/local-workflow-skill (12 stars, last pushed 1mo ago), licensed MIT. It adds 73 tokens to every session and 2,429 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
instrument-data-to-allotrope
Convert laboratory instrument output files (PDF, CSV, Excel, TXT) to Allotrope Simple Model (ASM) JSON format or flattened 2D CSV. Use this skill when scientists need to standardize instrument data for LIMS systems, data lakes, or downstream analysis. Supports auto-detection of instrument types. Outputs include full…
html-ppt-hermes-cyber-terminal
OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.
bigquery-ai-ml
Leverages BigQuery's built-in machine learning and GenAI capabilities for advanced data analytics. Use when you need to write SQL queries that perform time-series forecasting, predict values, detect outliers or anomalies, find key drivers, perform semantic search or vector search, classify text, calculate similarity…
mixed-precision
Use FP16/BF16 mixed precision to accelerate training and reduce memory. Use when optimizing GPU performance.
model-compatibility
Model family compatibility matrix covering loaders, resolutions, samplers, CFG, VAE, ControlNet, and LoRA compatibility for SD 1.5, SDXL, Flux, SD3, and video models.
civitai
Discover Civitai models with the BUILT-IN downloadmodel action:"searchcivitai" and install/generate them locally. Find a checkpoint/LoRA/embedding on Civitai, download it into ComfyUI, and use its trigger words. Optionally pair the official Civitai MCP for community features (images browsing, posting, collections).