Board Pack
Provenance and versioning
Every pack records the engine, scanner, severity scope, regions and confidence that produced it, so it can be reconstructed after the evidence expires.
What is recorded with every pack
| Recorded | Why |
|---|---|
| Engine version | The scoring implementation that produced the number |
| Scanner version | Which checks existed, and therefore what the denominator was |
| Severity scope | Which severities were collected at all |
| Scan regions | What was in scope |
| Evidence confidence | Whether the rating was supportable |
| Generation time | When the environment was observed |
| Assessment reference | Which run produced it |
These are stored alongside the pack, independently of the PDF and independently of the workflow platform.
Why independence matters
Once the assessment artefacts expire, the evidence set is gone. If the only record of "what produced this document" lived in the evidence, the question would become unanswerable within days.
An earlier version stored none of this: the answer to "what exact engine and evidence produced the document this customer received" could only be obtained by opening the PDF, and only until retention lapsed.
The scoring fingerprint
Beyond the version numbers, a fingerprint is derived from the scoring constants in force. The next assessment compares it, and withholds score comparison if the constants changed.
A version number tells you the code changed. The fingerprint tells you whether the change affected the calculation this pack performed, which is the question a comparison actually turns on.
Reading it
The Scope and Reproducibility page of the pack prints what was covered and what it would take to reproduce the result. The Methodology page prints the constants.