agentsclimarketplace
Sign in

Ml feature engineering

Skill planifest/planifest-framework/planifest-framework/external-skills/ml-feature-engineering

ML feature engineering workflow for feature definition, lineage, and online-offline parity. Use when model performance depends on explicit feature design and parity controls; do not use for generic API-layer or infrastructure-only changes.From its SKILL.md

Install
npx -y skills add planifest/planifest-framework --skill ml-feature-engineering

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

1.9 KB, 315 tokens by cl100k_base, as published. Nobody here has run it

Ml Feature Engineering

Overview

Use this skill to design features that are useful, explainable, and consistent across training and serving.

Scope Boundaries

  • Use this skill when the task matches the trigger condition described in description.
  • Do not use this skill when the primary task falls outside this skill's domain.

Shared References

  • Online/offline parity rules:
    • references/online-offline-parity-rules.md

Templates And Assets

  • Feature specification template:
    • assets/feature-spec-template.csv

Inputs To Gather

  • Candidate feature hypotheses and business rationale.
  • Data sources and freshness constraints.
  • Serving path capabilities and latency budget.
  • Leakage/fairness/compliance constraints.

Deliverables

  • Feature catalog with lineage and ownership.
  • Parity validation plan for train vs serve paths.
  • Feature risk and maintenance notes.

Workflow

  1. Define feature specs in assets/feature-spec-template.csv.
  2. Validate parity assumptions with references/online-offline-parity-rules.md.
  3. Prioritize features by incremental value vs complexity.
  4. Verify leakage and freshness assumptions.
  5. Publish feature rollout and deprecation plan.

Quality Standard

  • Feature definitions are versioned and reproducible.
  • Online/offline behavior is consistent for decision-critical features.
  • Feature ownership and monitoring are explicit.

Failure Conditions

  • Stop when feature logic diverges between training and serving.
  • Stop when feature value cannot justify operational complexity.
  • Escalate when parity gaps remain unresolved.

What ships with it: 1 file

11.3 KB alongside SKILL.md

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.