# D3D-001-LAB1 — learner workspace

Build `Production Mesh Modeling and Topology Evidence Pack v1` from visible synthetic evidence. The starter does not contain a prefilled final decision.

Set `fixture_ref` to one of the visible IDs listed in `CHECK_RULES.json` and set `evidence_ref` to `fixture://<fixture_ref>`. For a mutation, use the mutation ID itself; the checker deterministically applies that mutation to the shipped baseline before comparing submitted evidence. It never supplies the final state.

Use `baseline_fixture.json`, diagnose each mutation independently, then recompute the transfer fixture from its own target. A chosen state needs at least one applicable learner-authored future trigger named `review`, `migration`, `escalation`, `stop`, or `pivot`; all five categories are not required.

Run: `python run_checks.py evidence/your_pack.json`

`CONSISTENT` means only that the submitted JSON binds to the declared shipped fixture and satisfies the published rules in `CHECK_RULES.json`. It is not human approval, rights proof, interoperability certification, or a hidden answer key.

### Positive-state fixture binding
For `READY` or `TARGET_SPECIFIC`, every P8-critical/positive-required field and every explicit target requirement present in the resolved shipped fixture must be supplied and match it. `material_slot_count` is included as non-critical inventory evidence: you may record it and the checker binds it when supplied, but omitting it alone does not block a positive state.