Domainpack Gallery

Each domainpack ships a binding_spec.yaml that maps a real-world problem onto SCPN Kuramoto oscillators. The notebooks below demonstrate baseline vs orchestrated simulations.

What a domainpack is — and is not

A domainpack is a reusable scaffold: a working binding that parses, runs end to end, and gives you a starting point for that domain. It is not a validated detector for that domain. Exactly one domain — power grid — is externally validated (growth rate vs ANDES eigenvalues, ρ≤0.87); the rest are templates that demonstrate SPO is domain-agnostic. Before trusting any detector on real data, run it through the spo audit-detector auditor — generic early-warning detection is at chance outside the grid, and SPO reports that.

Broader SCPN ecosystem — experimental research domains

Two packs — metaphysics_demo and identity_coherence (marked 🔬 in the catalogue below) — are experimental research domains from the broader SCPN ecosystem, not part of the validated clinical/grid product surface. metaphysics_demo exercises the P/I/S channels on the framework itself; identity_coherence is a speculative self-model of the system that built SPO. They parse, run, and are tested like any scaffold, but they make no empirical claim about a real-world system — a buyer evaluating the grid or EEG product can ignore them.

This page is searchable — use the search box (top of the docs) to filter by domain, feature, or pack name. Packs marked ⭐ try first below are the recommended starting points.

⭐ Try these first

New to SPO? Start with one of these three. Each is a research-tier spec that runs end to end, and the first two run in a single command:

Pack	Why start here	One command
power_grid ⭐	The grid-forming (dVOC) reference pack — detect, flag, and damp a poorly-damped oscillation mode with auditable before/after evidence.	spo quickstart power
neuroscience_eeg ⭐	The neuroscience reference pack — six EEG bands, coherence control across the spectrum.	spo quickstart eeg
minimal_domain ⭐	The simplest possible binding spec — four oscillators — to learn the schema before scaling up.	spo run domainpacks/minimal_domain/binding_spec.yaml

The full catalogue follows, organised by domain.

Notebooks

#	Domainpack	Oscillators	Key Feature	Notebook
01	queuewaves	6 (micro/meso/macro)	Retry-storm recovery, supervisor 9x speedup	01_queuewaves_retry_storm.ipynb
02	minimal_domain ⭐	4 (lower/upper)	Simplest possible spec, coherence convergence	02_minimal_domain.ipynb
03	geometry_walk	8 (local/global)	Symbolic channel, graph-walk phases, zeta drive	03_geometry_walk.ipynb
04	bio_stub	16 (4 scales)	Multi-scale biology, imprint memory	04_bio_stub.ipynb
05	manufacturing_spc	9 (sensor/machine/line)	Bad-layer suppression, policy rules	05_manufacturing_spc.ipynb
06	stuart_landau	8	Phase-amplitude ODE, bifurcation tracking, PAC	06_stuart_landau_amplitude.ipynb
07	policy_petri_net	8	Policy DSL, regime FSM, Petri net sequencing	07_policy_petri_net.ipynb
08	audit_replay	8	SHA256-chained audit trail, deterministic replay	08_audit_replay.ipynb
09	binding_spec	varies	Binding spec schema walkthrough, validation	09_binding_spec.ipynb
10	adapters	varies	Bridge adapters (fusion, quantum, neurocore)	10_reporting_adapters.ipynb
11	identity_coherence	30 (6 layers)	SSGF identity model, chimera + plasticity	11_identity_coherence.ipynb
12	autotune	varies	Frequency ID, coupling estimation pipeline	12_autotune_pipeline.ipynb
13	ssgf	varies	SSGF free energy closure	13_ssgf_closure.ipynb
14	chimera	varies	Chimera state identification and visualization	14_chimera_detection.ipynb
15	spectral	varies	Spectral Alignment Function (SAF) optimization	15_spectral_analysis.ipynb
16	sleep_architecture	8 (4 layers)	EEG sleep stage classification from R	16_sleep_staging.ipynb
17	power_grid ⭐	12 (5 layers)	Inertial Kuramoto, generator trip transients	17_power_grid_stability.ipynb
18	financial_markets	8 (4 layers)	Hilbert phase, order parameter regime detection	18_market_regime_detection.ipynb
19	swarm_robotics	8 (3 layers)	Swarmalator spatial + phase coupling	19_swarmalator_dynamics.ipynb

All 36 Domainpacks

Pack	Domain	Layers	Oscillators	Channels
agent_coordination	Multi-agent AI coordination	3	12	I
autonomous_vehicles	Vehicle platoons	3	8	P/I
bio_stub	Multi-scale biology	4	16	P/I/S
brain_connectome	HCP-inspired structural connectivity	4	12	P/I/S
cardiac_rhythm	Cardiology	4	10	P/I
chemical_reactor	Process control	4	10	P/I
circadian_biology	Chronobiology	4	10	S
digital_twin_nchannel	Digital twins	3	6	P/I/S/Thermal/Quality/TwinResidual
edge_consensus_nchannel	Edge orchestration	3	6	P/I/S/Load/Trust/ConsensusHealth
epidemic_sir	Epidemiology	3	8	P/I
financial_markets	Stock sync and crash detection	4	8	P/I/S
firefly_swarm	Ecology	2	8	P/I
fusion_equilibrium	Fusion plasma	6	12	P/I
gene_oscillator	Repressilator + quorum sensing	3	6	P/I/S
geometry_walk	Graph systems	2	8	S
identity_coherence 🔬	Identity-coherence model (SSGF)	6	30	P/I/S
laser_array	Photonics	3	8	P/I
manufacturing_spc	Manufacturing	3	9	P/I/S
metaphysics_demo 🔬	P/I/S showcase	3	7	P/I/S
minimal_domain ⭐	Synthetic baseline	2	4	P
musical_acoustics	Consonance and groove via sync	3	9	P/I/S
network_security	Cybersecurity	3	8	I
neuroscience_eeg ⭐	Neuroscience	6	14	P/I
plasma_control	Tokamak plasma	8	16	P/I
pll_clock	Telecommunications	3	8	P/I
power_safety_nchannel	Power systems	3	6	P/I/S/Reserve/Weather/Risk
power_grid ⭐	Power systems	5	12	P/I
quantum_simulation	Quantum computing	3	8	P/I
queuewaves	Cloud/queues	3	6	P/I
robotic_cpg	Quadruped CPG locomotion	4	8	P/I/S
rotating_machinery	Vibration	4	10	P/I
satellite_constellation	Aerospace	3	8	P/I
sleep_architecture	AASM sleep staging from R values	4	8	P/I/S
swarm_robotics	Robotics	3	8	P/I
traffic_flow	Transportation	4	10	P/I
vortex_shedding	Wake dynamics (Stuart-Landau)	3	9	P/I/S

🔬 = broader SCPN ecosystem — experimental research domain (see the note above); exercises the pipeline but makes no empirical claim, and is not part of the validated clinical/grid product surface.

Benchmark Results (33-Domainpack Snapshot, Measured 2026-03-28)

The 2026-03-28 benchmark snapshot tested 33 domainpacks via spo demo --steps 20 on Kaggle (Linux, Python 3.12, NumPy fallback — no Rust kernel). Zero failures.

Domainpack	Oscillators	Layers	R (20 steps)	Regime
agent_coordination	12	3	0.402	degraded
autonomous_vehicles	8	3	0.394	degraded
bio_stub	16	4	0.274	degraded
brain_connectome	12	4	0.375	degraded
cardiac_rhythm	10	4	0.130	nominal
chemical_reactor	10	4	0.410	nominal
circadian_biology	10	4	0.357	degraded
epidemic_sir	8	3	0.590	nominal
financial_markets	8	4	1.000	nominal
firefly_swarm	8	2	0.566	degraded
fusion_equilibrium	12	6	0.196	nominal
gene_oscillator	6	3	1.000	nominal
geometry_walk	8	2	0.795	nominal
identity_coherence	35	6	0.352	degraded
laser_array	8	3	0.312	degraded
manufacturing_spc	9	3	0.530	nominal
metaphysics_demo	7	3	0.572	degraded
minimal_domain	4	2	0.524	degraded
musical_acoustics	9	3	0.998	nominal
network_security	8	3	0.575	nominal
neuroscience_eeg	14	6	0.114	nominal
plasma_control	16	8	0.321	nominal
pll_clock	8	3	0.209	degraded
power_grid	12	5	0.345	nominal
quantum_simulation	8	2	0.608	degraded
queuewaves	6	3	0.735	nominal
robotic_cpg	8	4	0.999	nominal
rotating_machinery	10	4	0.287	nominal
satellite_constellation	8	3	0.503	nominal
sleep_architecture	8	4	0.229	nominal
swarm_robotics	8	3	0.380	degraded
traffic_flow	10	4	0.492	nominal
vortex_shedding	9	3	0.999	nominal

Notes: R values at step 20 from random initial phases. "degraded" = R < 0.6 (not yet synchronised). Longer runs converge for all domainpacks. Financial markets and gene oscillator reach R=1.0 within 20 steps due to high coupling topology.

Running Locally

pip install -e ".[dev,plot]"
jupyter lab notebooks/

21 notebooks total. CI executes the shipped notebook suite on Python 3.12 with jupyter nbconvert --execute; see Notebook Execution Matrix for per-notebook extras, runtime class, and CI expectation.

Adding a New Domainpack

1. Create domainpacks/<name>/binding_spec.yaml following the binding spec schema
2. Add domainpacks/<name>/policy.yaml with supervisor rules
3. Create a notebook in notebooks/ following the pattern above
4. Add a row to this table