A Grammar of Machine Learning Workflows

Data leakage has been identified in 648 published machine learning papers across 30 scientific fields. The knowledge to prevent it exists; the tools do not enforce it. This paper presents a grammar - eight typed primitives, a directed acyclic graph, and four hard constraints - that makes the most damaging leakage types structurally unrepresentable. The core mechanism is a terminal assessment gate: the first call-time-enforced evaluate/assess boundary in an ML framework, backed by a specification precise enough for independent reimplementation. A companion landscape study across 2,047 datasets grounds the constraints in measured effect sizes. Two reference implementations (Python, R) are available.