pers-02 reproduction
Public source retained with the related article. Generated results, dependency directories, runtime storage, secrets, and oversized binary artifacts are intentionally excluded.
README
# PERS-02 creation and reconstitution reproduction
This fixture tests one narrow claim: creating a new aggregate under current
policy is not the same operation as rebuilding meaningful state accepted in
the past.
The pinned PHP 8.4.12 and SQLite experiment retains current creation and
refusal, legacy loading, no-new-fact, stored-version, malformed-state,
contextual load-failure, no-silent-repair, and framework-source controls.
Laravel's `Model::newFromBuilder()` observation is pinned to framework commit
`1de10d979568ff823606d4980bafb409b8e7ddaa` and a hash of the inspected method.
It describes Eloquent record hydration only; it is not treated as domain
construction.
Run the complete retained experiment with:
```bash
./run.sh
```
Reproduce it from the exact committed evidence tree with:
```bash
./bin/run-clean-export.sh 6f84870
```
The order policy and stored history are synthetic. They are not Brian's
production rules or evidence about the frequency or cost of data migrations.