# Requirement traceability

The standalone v3 manuscript retains the 20 base sections and adds 6A, 13A, 14A and targeted compiler/scaling/audit extensions, followed by references and the required final assessment. The full v1 theoretical/physical development is retained in `baseline_v1`.

| Requested item | Main v2 coverage | Implementation / boundary |
|---|---|---|
| Title, abstract, central contribution | Cover, abstract, §1 | Authorship matches requested string |
| Mathematical framework | §§2–6 | `contracts.py`, `recovery.py` |
| Theorems A–F | §§3–10 | Restricted proofs; no chemical universality claimed |
| Developmental complexity, seed, position | §§7–8 | `contract_genome.py`; v1 geometry compiler |
| Error correction and scalable yield | §§3–6, 9 | R1–R9; coded chemical references unimplemented |
| Growth genome and compiler | §8 | Bounded family implemented; general compiler proposal |
| Physical finite alphabet | §§8, 12 | Ingredients/roles distinguished from molecular species |
| Multi-material and soft-to-hard conversion | §§11–12 | Selected prior material evidence; new actuator unbuilt |
| Multiscale precision | §§7, 10, 16 | Explicit atomic-control fraction and interface-complexity warning |
| Transport and concurrency | §10, §16 | Capacity-aware service model; retained v1 diffusion model |
| Thermodynamics, entropy, heat | §10 | Calculations clearly conditional; no fabricated energy measurements |
| Growth versus distributed supply | §10 | Causal, flux, diffusion bounds |
| Topology and occlusion | §4, §10 | Local service-tree closure wired into simulation |
| Functional Levels 0–7 | §15 | Quantitative criteria by function |
| Leverage, autonomy, compilation ratio | §7 | Charges target environment, platform and finite precision |
| Fabrication channel capacity | §7.5 | Capacity region; no unsupported Shannon-achievability claim |
| Simulation, errors, repair, comparison | §13 | 512 + 32 runs; reproducible logs |
| Ablations and negative controls | §13, §17 | Conventional tie, bias, early closure, seal stress |
| Experimental roadmap | §§14–15, 19 | Prioritized by information gain, cost, difficulty |
| Minimal decisive experiment | §14 | Four-module post-conversion repair/access factorial |
| Failure/falsification | §§14, 17–18 | Claim-specific observations and alternatives |
| Self-repair | §12.1 | Mission-time references, stock, access, and maintenance limits |
| Relation to biology | §12.1 | Component evidence distinguished from engineering solution |
| Prior art and novelty discipline | §§1, 17, references | `SOURCE_AUDIT.md`, source documents/hashes |
| 100 nm through 1 cm scaling | §16 | Unit counts, diffusion, flux, time, measurement, strain |
| 1 / 3 / 5 / 10 / 20 year program | §19 | Conditional milestones versus long-term speculation |
| Final adversarial audit | §17 | Explicit remaining gaps |
| Required assessment | Final assessment | Moderate subjective maturity, no 100% claim |


## v3 additions

| New question | Coverage | Boundary |
|---|---|---|
| Does absolute scale have to be accurate? | G1, 6A.2 | Only a declared scale-invariant function can omit it |
| Can overshoot be repaired with additions? | G2, 6A.3 | Exact independent scalar reserve interval |
| Can finite noisy increments reach the target? | G3, 6A.4 | Sufficient guard, reserve, horizon and access conditions |
| Is there a sharp manufacturability boundary? | G4, 6A.5 | Exact ideal bounded-disorder reachability threshold |
| Can matter use an imperfect reference? | G5-G6, 6A.6-6A.8 | Unknown absolute scale allowed; differential bias remains |
| Is local computation actually implementable? | 6A.7 and source local_estimate | Neighbor updates verified separately from ensemble inference |
| When can diagnostic paths be removed? | 6A.9 | Full backbone retained until discharge; early optimization open |
| Is the new function nontrivial? | 13A | Four external terminals plus an internal voltage map |
| What new experiment decides the claim? | 14A | Four-module converted-material bridge with explicit negative controls |
| What remains expensive? | 13A, 16.1, 17.4 | Comparator hardware, millions of samples, local inference latency |