leakage-adversary

name

leakage-adversary

description

Actively search ML datasets and pipelines for target, temporal, group, split, join, and preprocessing leakage. Use when an agent needs a judgment-heavy data science workflow for attack ml pipelines for leakage, including evidence review, local artifact inspection, risk classification, stakeholder-ready decisions, reproducibility, governance, or agent-to-agent handoff. Trigger for Codex, Claude, Gemini, Copilot, Cursor, Windsurf, Gravity, LangGraph, CrewAI, AutoGen, or local agents when this exact workflow is needed.

Leakage Adversary

Mission

Actively search ML datasets and pipelines for target, temporal, group, split, join, and preprocessing leakage.

Current pain point: Models look excellent offline and fail in production because future or target information leaked into training.

Why this skill exists: Leakage must be attacked like a failure mode; ordinary metric review usually detects it too late.

Operating Rules

• Start by restating the decision this skill is supporting.
• Inspect local artifacts first before asking for context.
• Treat scripts as evidence collectors, not as substitutes for judgment.
• Preserve raw data, notebooks, configs, and model artifacts unless the user explicitly asks for mutation.
• Mark missing context as unknown, not provided, or owner decision; do not invent it.
• Classify each blocker as stop, fix-first, monitor, accepted risk, or owner decision.
• Use the output contract exactly unless the user asks for a different format.
• Keep final recommendations auditable: every important claim needs evidence, assumption, or caveat.

What This Skill Must Do

• Transform a vague request into an explicit Leakage Adversary decision with named owner, evidence, assumptions, and action threshold.