Audit and create leakage-free features in reproducible Jupyter notebooks.
npx skills add https://github.com/tim-krausz/mlstack --skill feature-eng
Please help me install this Agent Skill. Skill: feature-eng Source: https://github.com/tim-krausz/mlstack/tree/main/feature-eng Command: npx skills add https://github.com/tim-krausz/mlstack --skill feature-eng
ML teams often struggle to produce interpretable, leakage-free features with strong documentation. This Skill provides a disciplined, notebook-driven workflow that enforces domain rationale, leakage checks, and reproducible feature pipelines from audit to deployment.
Provide your dataset and target; the system will perform a feature audit, generate defensible features with leakage checks, and output a documented notebook.
High-intent search queries and answers about installing and using this skill.
Data leakage detection in feature engineering is automated by auditing existing features for type, distribution, and target relationships to inform safe feature creation and prevent target information from bleeding into training inputs.
Documenting feature engineering decisions in a Jupyter notebook is handled through a disciplined workflow that records explicit rationales for each feature creation phase, producing an end-to-end reproducible notebook with robust documentation.
Building reproducible ML preprocessing pipelines is achieved by serializing preprocessing steps during end-to-end pipeline assembly, which records all feature engineering decisions and outputs a fully documented, reproducible notebook for deployment.
A Python environment with Jupyter notebooks is required to audit existing features for target leakage, as the workflow relies on notebook-driven execution to enforce domain rationale checks and generate reproducible feature pipelines.
Automating feature creation for structured tabular data in ML workflows is supported through phase-by-phase generation that includes explicit rationale recording, leakage auditing, and robust documentation directly within a Jupyter notebook environment.
anilcan-karaCreates a live Jupyter kernel for executing Python code with persistent variables across sessions.
Ricko12vPLAutomate end-to-end ML workflows from data preprocessing to production deployment.
haJ1tBuild, train, deploy, and monitor ML systems with Python and scikit-learn, TensorFlow, or PyTorch.
vasilyu1983Build production-ready ML workflows with EDA, feature engineering, and SQLMesh transformations.
