Audit Log Schema

The audit log is the metadata surface that explains how a generated artifact was produced.

It should be read as provenance and execution context, not as a substitute for the raw input dataset or for researcher interpretation.

Access model

Audit-oriented metadata is exposed through authenticated product surfaces.

Example route shape:

GET /api/artifacts/{artifactId}/generation-meta

Authentication is required, and access should be read as scoped to artifacts the authenticated user is allowed to inspect.

What the audit log is for

The audit log helps answer questions such as:

• what analysis or generation path produced this artifact?
• what version or snapshot context was involved?
• what decision-relevant metadata was recorded at generation time?
• what warnings, notes, or supporting context were attached to the result?

Typical contents

The exact payload may evolve, but readers should expect audit metadata to carry information such as:

• generation context
• route or engine provenance
• version or snapshot identifiers
• export-relevant metadata
• guard, warning, or note context where applicable

What the audit log does not guarantee

The audit log should not be over-read.

It does not automatically guarantee:

• that every possible study-design problem was detectable
• that the artifact is publication-ready without review
• that every external reporting requirement has been satisfied
• that a recorded warning was fully resolved rather than merely documented

For interpretive limits, continue to Known Limitations and Interpreting Guards and Heuristics.