A golden test runs a pipeline on a fixed input and compares its output with an expected output, row for row. If they differ, the test fails.
Each Emolysis template ships with golden tests built for its shape: three input files with one malformed for Ingest (batch), an attribute that changes across three runs for SCD Type 2, a fact that arrives two days late for Late-arriving, and so on.
The tests run on DuckDB on your machine. A red result sends the agent back to the template, and only a green run deploys. The post tests before production explains what golden tests prove and what they do not, and the tutorial preview walks through writing one.