SPM-Kit implementation map¶

SPM-Kit architecture from instrument files through readers, domain models, pure analysis and typed results to Python, CLI, Fathom and exports — Core owns computation. Python, the CLI and Fathom orchestrate the same public numerical paths.

Concept-to-code map¶

Concept	Current implementation	Fathom	CLI/public route	Strongest retained evidence	Known limitation
file inspection/routing	`core.io.load_any`, `core.plugins`, built-in readers	automatic open route	`spmkit info`	format-specific Level 1/2	support is variant-specific
leveling	`core.analysis.leveling`	Imagen (`image`)	`roughness --level`, `analyze --level`	Level 1 + synthetic cases	changes the reference surface
arc-revolution background	`core.analysis.background`	not exposed	Python API	Level 1 with test-local 1D oracle and synthetic physical tests	finite geometric Z only; no masks, CLI, Fathom or external-equivalence claim
sphere-revolution background	`core.analysis.background`	not exposed	Python API	Level 1 with independent 2D nearest/reflect oracles and synthetic physical tests	true 2D spherical cap; no masks, CLI, Fathom, external-equivalence or performance claim
Sa/Sq/Sz/Ssk/Sku	`core.analysis.roughness.statistics`	Imagen (`image`)	`spmkit roughness`, `analyze`	scoped Level 3 for Sa/Sq/Sz	external campaigns do not cover Ssk/Sku or every preprocessing route
line profile	`core.analysis.profiles.line`	Imagen (`image`)	Python API	Level 1	interpolation and coordinate choice matter
grain segmentation	`core.analysis.grains.detect`	Granos (`grains`)	`spmkit grains`	Level 1 + synthetic tests	threshold/overlap/tip effects
radial PSD/fractal fit	`core.analysis.spectral`	Espectral (`spectral`)	`spmkit psd`	Level 1/path-specific numerical tests	bandwidth and fit-band dependence
KPFM statistics/work function	`core.analysis.kpfm.statistics`	Imagen (`image`)	`analyze --tip-wf`, Python	Level 1	sign convention and tip calibration
baseline/contact/elastic fit	`core.analysis.forcecurve`, `mechanics`	Curva de fuerza (`force`)	`forcecurve`, `nanomech`	model-specific Level 2	contact, geometry and material assumptions
JKR	`core.analysis.experimental`	Curva de fuerza (`force`) where exposed	`spmkit jkr`	experimental synthetic recovery	adhesive-contact assumptions; CLI marks experimental
SMFS WLC/FJC/events	`core.analysis.chain`	SMFS (`smfs`)	Python API	Level 2 synthetic recovery	polymer model and event-window dependence
force-volume maps	`core.analysis.forcevolume`, `forcevolume_fast`	Mapa (`map`)	`spmkit forcemap`	Level 2 synthetic recovery	invalid curves/calibration propagate spatially
batch analysis	`core.batch`, `core.forcebatch`	Batch (`batch`)	`batch`, `fbatch`	Level 1	homogeneous channel/parameter assumptions
SHO and thermal spectrum	`core.analysis.resonance`	Sintonía térmica (`resonance`)	Python API	Level 1 software tests and controlled numerical cases	no frozen calibrated physical-reference campaign
frequency-shift mass model	`core.analysis.resonance`	Evaporación (`evaporation`)	`spmkit evaporation`	Level 1 software tests and controlled numerical cases	point-mass/effective-mode assumptions
publication figure	`core.viz`	Figura (`figure`)	`spmkit figure`	Level 1	figure quality is not scientific validation
3D rendering	GUI panel over image domain	Vista 3D (`view3d`)	none	GUI/software tested	display transform is not analysis
educational simulation	`core.analysis.simulation`	Simulador (`simulator`)	Python API	software-tested analytical construction	not a calibrated instrument simulator or physical evidence

Public boundary¶

core/  ← numerical truth and domain objects
  ↑
  ├── cli/  arguments, files and terminal output
  └── gui/  Fathom routing, parameters, views and reports

Validation invokes the installed public package from another process. Phantoms does not import Core. Those two separations are necessary for useful evidence, but reference independence is still assessed per campaign.

How maturity advances¶

Define units, assumptions and failure behavior in Core.
Add deterministic software tests.
Recover a known numerical case, preferably from independent Phantoms logic.
Freeze an external comparison/reference and tolerance in Validation.
Add calibrated physical and interlaboratory evidence when it exists.
Expose the stable result in Fathom without changing the equation.

Scientific status · Architecture · Ecosystem workflow E