The project is an open software catalogue for comparing fusion-device concepts. Its author, Miroslav Šotek, describes 21 device-family repositories, one per concept, plus a shared kernel library and a general tokamak physics toolkit. The author states plainly that no parameter set in the catalogue describes or validates a real machine. Each family repository is best read as a set of closed-form physics checks and geometry exports for a concept, not as a reactor design.
What the catalogue contains
The article, published September 28, 2026 on DEV Community, names the repositories below. Status entries reflect only what the article itself says; none of the repositories was independently opened for this piece.
As an Amazon Associate I earn from qualifying purchases.
| Repository or item | Role, as described by the author | Status, as stated in the article |
|---|---|---|
| scpn-reactor-kernels | Shared kernel library used across the family repositories | Status not stated |
| scpn-z-pinch-core | Example family repository for the Z-pinch concept | No release or DOI yet |
| scpn-tokamak-core | Example family repository for the tokamak concept | No release or DOI yet |
| scpn-fusion-core | Tokamak physics and control repository | Status not stated |
| 29 further families | Planned concepts covering fission, chemical reactors, and hybrids | Planned; the family repositories do not yet exist |
| Reactor Studio | Planned design environment that follows the families | Planned; not available |
The 21 repositories count is the author’s figure for the catalogue as described in the article. The 29 additional families are a roadmap count, not completed work.
Free tools Windows power users keep installed
One-click scans. No signup required.
What “level-0” means in this catalogue
The word “level-0” carries a narrow meaning, and most misreadings of the project come from ignoring it. The author defines it as “closed-form models from the device’s own literature, evaluated on the validated configuration, without solving any equation.” In practice, a level-0 model returns a value from a formula. It does not iterate a solver to find an equilibrium or a time evolution.
#1 Best Overall
The article is explicit about what that excludes. No equilibrium or stability equation is solved, and reactivity, yield, gain, and breakeven are out of scope. A reader who wants to estimate how much power a plant would produce, or whether it would reach gain above one, will not find that answer here.
Inside the Z-pinch example repository
The author uses scpn-z-pinch-core as the worked example. According to the article, it contains the following:
- A validated configuration model for the Z-pinch concept.
- Explicit diagnostic and timing semantics, so that the meaning of each reported quantity and its time base is written down.
- Concept-specific level-0 physics: the Bennett equilibrium, ideal-MHD and Kadomtsev stability limits, the Shumlak–Hartman model, and the Pease–Braginskii relation.
- Geometry exports in STL and glTF mesh formats, plus deterministic STEP CAD output.
- A test suite that the author reports at 100% statement and branch coverage. This is the author’s own software metric, and no independent check of it is reported.
The article also says that where two implementations of the same calculation exist, they must agree: “If the two implementations differ in the last bit, the test fails.” That is a useful standard for a software catalogue, but it tests agreement between implementations, not agreement with experiment.
What the catalogue does not claim
The limits are stated as clearly as the features, and they matter more for anyone deciding how to use the repositories.
Rank #3
- No real machine is described or validated. The author says “no parameter set describes or validates any real machine.”
- Not machine-ready, not safety-certified, not reactor-ready. The models are offered for inspection and comparison only.
- No solver, controller, experimental correlation, or dataset is available yet.
- The tokamak repository is not a specific tokamak. The author describes it as a cylindrical periodic equivalent, a simplified geometry standing in for tokamak physics in general, not a model of any named device.
What is planned rather than built
The author describes a further 29 families covering fission, chemical reactors, and hybrids, all following the same pattern, and a later Reactor Studio. These are roadmap items. Readers should not look for them among the current repositories, and the article does not give a date for them.
Because the Z-pinch and tokamak repositories have no release or DOI yet, a citation or reproducible reference should point to a specific commit rather than to a version number. The article does not state a release schedule.
Comparing the concepts as software
The catalogue is a set of software representations, not a comparison of operating machines. Performance, cost, gain, safety, and readiness cannot be read from it. If you compare two family repositories, the article supports four grounded axes:
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →- Which device-family physics are represented at level-0.
- Whether the family has configuration and diagnostic or timing semantics.
- Which geometry exports are provided.
- Whether the model has experimental correlation or machine-specific validation. The author says it does not for any family described.
Reviewing one family repository yourself
The author invites readers to check the models rather than take them on trust. A practical way to do that with scpn-z-pinch-core is:
- Open the repository and find the validated configuration model and its documentation.
- Locate the Bennett equilibrium and the ideal-MHD and Kadomtsev stability limits, and confirm that each is written as a closed-form expression rather than a solver step.
- Check that the diagnostic and timing semantics define what each output quantity means and on what time base it is reported.
- Look at the STL, glTF, and STEP exports and confirm that they are generated from the same configuration.
- Note what is absent: no solver, no controller, no experimental correlation, and no dataset to compare against.
The author’s own question for readers is which closed-form limit you would check first. Answering it against a specific model is the most useful contribution a reviewer can make at this stage.
How much weight the evidence can bear
The catalogue’s description comes from one source: the author’s own article, which was drafted with help from AI agents using the repository documentation and then reviewed and edited by the author. It is a detailed account of intent and design, but it is not independent corroboration. Claims about the repositories’ contents, test results, and physics correctness are the author’s claims until someone outside the project has inspected them. Readers should treat the 100% coverage figure, the list of level-0 physics, and the description of the tokamak repository as statements from the project, not as verified results.
The article names GitHub Sponsors as a funding route. It does not describe any commercial product or service.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →The project’s contribution is its structure: a common format for stating what a concept model includes, what it leaves out, and what it explicitly does not claim. That is worth examining on its own terms, provided the limits stated above are kept in view.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
