ML teams need repeatable pipelines that connect data preparation, training, validation, deployment, and monitoring. This skill guides Claude, Codex, and Claude Code through practical MLOps workflow design.
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Agent request
Review the Skillstore skill "ml-pipeline-workflow" from https://skillstore.io/skills/wshobson-ml-pipeline-workflow.md and its manifest at https://skillstore.io/api/skills/wshobson-ml-pipeline-workflow/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.
Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
All nine static findings are false positives caused by Markdown inline code or fenced examples in SKILL.md. I found no prompt injection, data exfiltration intent, or unsafe external command behavior in the reviewed skill text.
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Convert an experimental notebook into a staged training and validation workflow.
Design production model operations
Plan orchestration, deployment, rollback, and monitoring patterns for production models.
Compare MLOps architecture options
Evaluate pipeline tools and rollout practices before selecting an implementation approach.
Try These Prompts
Map a Basic Pipeline
Create a simple machine learning pipeline for [use case]. Include data ingestion, training, validation, deployment, and monitoring stages.
Choose Orchestration Tools
Compare Airflow, Dagster, Kubeflow, and Prefect for my ML workflow. My constraints are [team size, cloud, latency, governance].
Design Validation Gates
Design validation gates for a model release. Include data quality checks, baseline comparison, approval criteria, rollback triggers, and reports.
Create a Production MLOps Plan
Build a production MLOps workflow for [model type] serving [traffic pattern]. Include DAG structure, experiment tracking, CI/CD, canary rollout, monitoring, and incident response.
Best Practices
Start with a simple linear pipeline, then add validation, monitoring, and rollout controls.
Version data, code, models, metrics, and deployment configurations together.
Define rollback criteria before production traffic reaches a new model.
Avoid
Skipping data validation before training or deployment.
Deploying models without baseline comparison, drift checks, or rollback triggers.
Choosing orchestration tools before defining pipeline stages and ownership.
Frequently Asked Questions
Does this skill execute ML pipelines?
No. It provides workflow guidance and planning support. Users run tools and infrastructure in their own environment.
Which platforms does it cover?
It discusses Airflow, Dagster, Kubeflow, Prefect, MLflow, Weights & Biases, SageMaker, Vertex AI, Azure ML, and Kubernetes serving.
Can it help with data preparation?
Yes. It covers validation, feature engineering, splitting strategies, dataset versioning, and lineage planning.
Can it design deployment workflows?
Yes. It explains shadow deployment, canary release, blue-green deployment, rollback, monitoring, and production validation.
Is it suitable for regulated ML systems?
It can support planning, but teams must add required privacy, security, audit, and governance controls.
Does it include complete templates?
The scanned skill file references templates and examples, but it mainly provides guidance rather than a full project scaffold.