Energy substrate of the HEXA family. n=14 verb lattice + 7-group integration govern the substrate's identity. v1.0.0 ships as SPEC_FIRST (own 5): 14/14 markdown specs extracted from
canon/domains/energy/@c0f1f570(2026-05-06), 0/14 working.hexa/.pykernels.Sister CLIs called directly when their axes are needed (out of scope here):
hexa-fusion·hexa-antimatter·hexa-rtsc·hexa-earth.
Energy is the load-bearing substrate beneath every other HEXA axis (compute,
mobility, manufacturing, life-support). The canon monorepo had
energy specs scattered across 14+ subdomains; hexa-energy consolidates
them into a single MIT-licensed public spec-first substrate that downstream
consumers can install via hx install hexa-energy.
The n=14 verb lattice is the falsification budget:
verb_count ≡ 14(battery_arch, battery_energy, nuclear, smr_dc, dc_reactor, grid, pv_microgrid, solar, pemfc, hvac, thermal, mineshaft, arch, efficiency)group_count ≡ 7(battery / nuclear / grid / fuel-cell / thermal / mining / meta)- per-group:
2 / 3 / 3 / 1 / 2 / 1 / 2— sum 14 - Sum-of-divisors of 14: σ(14) = 1+2+7+14 = 24 (verb pairs across the partition graph; not used as a hard invariant at v1.0.0)
- Number-of-divisors: τ(14) = 4 — matches the 4-axis exclusion set (fusion / antimatter / RT-SC / climate-cousin earth) handled by sibling CLIs
Every group answers to the lattice. Where a verb roster change is proposed,
the lattice equality must be restated across all five surfaces (domain.md,
spec.md, hexa.toml, cli, verify).
| Axis | Other energy substrates | hexa-energy v1.0.0 |
|---|---|---|
| Scope | Single energy subtype (battery only / grid only) | 14-verb / 7-group integration substrate |
| Format | Often binary or proprietary | Spec-first markdown, MIT |
| Provenance | Vendor-confidential | n6-arch SHA-pinned (c0f1f570) |
| Wire status | Vendor-validated | SPEC_FIRST (0/14 wired, honest) |
| Frontier axes | Bundled or excluded | Federated to siblings (out of scope) |
| Falsifiability | Limited | n=14 lattice equality (verb_count == 14) |
Each of the 14 verbs inherits its own precursor stack from upstream
canon/domains/energy/<upstream-name>/. The integration
substrate adds two cross-cutting requirements:
| Precursor | Role |
|---|---|
| n6-arch domains/energy/ @ c0f1f570 | upstream SSOT for all 14 verb specs |
| HEXA family raw 270/271/272/273 + arch.001 | structural pattern (core+module+ai-native triplet) |
For per-verb precursors see core/energy/domain.md and
module/<verb>/<verb>.md.
| Group | Verbs | Count | Source (canon/domains/energy/) |
|---|---|---|---|
| battery | battery_arch, battery_energy |
2 | battery-architecture/, battery-energy/ |
| nuclear | nuclear, smr_dc, dc_reactor |
3 | nuclear-reactor/, smr-datacenter/, datacenter-reactor/ |
| grid | grid, pv_microgrid, solar |
3 | power-grid/, rooftop-pv-2nd-life-microgrid/, solar-architecture/ |
| fuel-cell | pemfc |
1 | pemfc/ |
| thermal | hvac, thermal |
2 | hvac-system/, thermal-management/ |
| mining | mineshaft |
1 | amd-ree-mineshaft-phes/ |
| meta | arch, efficiency |
2 | energy-architecture/, energy-efficiency/ |
| total | 14 |
T0 repo root (README/.own/hexa.toml/cli/verify/tests)
│
▼
T1 core/energy/ — single SSOT
spec.md (this file)
domain.md (numerical SSOT)
doc/ (cross-references)
│
▼
T2 module/<verb>/ — per-verb fan-out (14 modules)
README.md / README.ai.md
<verb>.md (verbatim from upstream)
[verify_<verb>.{hexa,py}] (when wired)
T2 modules MUST NOT import sibling T2 modules. Cross-verb references go
through T1 (core/energy/domain.md).
| Mk | Date target | Wired count | Note |
|---|---|---|---|
| v1.0.0 (mk1) | 2026-05-06 | 0/14 | Initial extraction. SPEC_FIRST. |
| v1.x.0 (mk1.x) | TBD | 1/14 | First wired verb (likely battery_energy via verify_battery-energy.hexa). |
| v2.0.0 (mk2) | TBD | ≥ 7/14 | Half-wired threshold; one verb per group has a working kernel. |
| v3.0.0 (mk3) | TBD | 14/14 | Fully wired; SPEC_FIRST verdict promotes to WIRED. |
SPEC_FIRST → WIRED promotion criteria are in .own own 5 + this §6.
Mk-ladder monotonicity: never insert a Mk regression (lower wired count
than predecessor).
verify/energy_verify.py (stdlib Python) checks:
- Verb sentinel sweep: 14
module/<verb>/directories present. - Group sentinel sweep: 7 group names referenced in
hexa.toml [verbs]. - n=14 lattice equality: verb_count == 14 across
domain.md+spec.md+hexa.toml+cli VERB_DIRS.
verify/cross_doc_audit.py checks:
- Per-group verb counts match (2/3/3/1/2/1/2).
- Verdict honesty —
SPEC_FIRSTand0/14 wiredconsistent across README + hexa.toml + cli output. - Out-of-scope axes mention is consistent (fusion / antimatter / rtsc / earth labelled "call sibling CLI directly", not "federated" or "passthrough").
Both verifiers must exit 0 to land any change touching the verb roster or group taxonomy. PASS does NOT imply any empirical claim has been validated (own 5).
| Item | Cost |
|---|---|
| Mac local default (spec read + CLI dispatcher placeholder) | $0 |
selftest (verb count check) |
$0 |
verify/energy_verify.py |
$0 (Python stdlib) |
verify/cross_doc_audit.py |
$0 (Python stdlib) |
module/<verb>/verify_<verb>.hexa (when wired) |
TBD per verb |
v1.0.0 substrate is doc-first; no CPU/GPU/cloud cost.
v1.0.0 verdict: SPEC_FIRST (own 5).
- 14/14 verbs ship as markdown specs (
module/<verb>/<verb>.md) - 0/14 verbs ship a working
.hexa/.pykernel cli/hexa-energy.hexais a placeholder dispatcher (group routing + selftest + status)selftest PASS= 14module/<verb>/present (sentinel sweep only)- Out-of-scope axes (fusion · antimatter · RT-SC · climate-cousin) call their
sibling CLIs directly — no proxy through
hexa-energy
| Sibling | Coupling |
|---|---|
hexa-fusion |
Out of scope. Call hexa-fusion … CLI directly when needed. |
hexa-antimatter |
Out of scope. Call hexa-antimatter … CLI directly. |
hexa-rtsc |
Out of scope. Call hexa-rtsc … CLI directly. |
hexa-earth |
Cousin (climate sink). Call hexa-earth … CLI directly. |
hexa-bio |
Sister extraction (life substrate). Pattern reference. |
hexa-sscb |
Sister extraction (compute substrate). Structural pattern reference. |
For full provenance see doc/lineage/origin.md.