TestCoupon turns a written description into a spec you correct, then compiles that spec into a real KiCad project, then checks it and tells you what it could not check. The generating is the easy half. Everything here exists to make the generating trustworthy.
That is a real result on a real board, and it is clean. The board still had nine defects and would not have powered on. This is why the pipeline is shaped the way it is.
Every line below is a part fitted the wrong way round, wired to a pin that does not exist, or ordered as a description rather than a part you can actually buy. None of it is exotic. All of it came from someone trusting what they remembered about a component instead of opening its datasheet.
Every one of those came from a part interface being assumed instead of retrieved. None of them were findable by ERC, DRC, schematic parity, or the house rules, because all of those check the capture against itself. Nothing in them knows what pin 3 of a real part does.
TestCoupon is what got built after that board. It is in private alpha behind an invite gate.
Four steps, in order, because each one refuses to run until the one before it holds up. That refusal is the product.
“A 940nm illuminator for night vision, battery powered, has to reach 300m.” What comes back is a draft spec: every value tagged with where it came from, and everything the model could not support marked as a guess rather than quietly presented as an answer. It also lists what it should have asked you and did not.
“You never picked switching versus continuous drive, the battery will die fast.” It revises and tries again, for as many rounds as you want. Every round is kept, so the wrong first draft survives next to the corrected one and the reasoning is recoverable later.
A typed block description becomes a real KiCad schematic and layout. The build is refused outright while any safety-relevant field in the spec is still a guess, so design effort is never spent underneath an assumption nobody checked. A refusal is not a failure; it is the gate doing the one thing it is for.
KiCad’s own electrical and design rule checks, house rules for what those two do not look at, and every part compared against the pin table in its manufacturer’s datasheet. Alongside the findings: what the run did not examine, so a short list can never be mistaken for a clean board.
Generating a plausible circuit is not hard, and that is the problem. Across 115 failed machine-synthesized designs, error rate rose with circuit size at r = 0.863: the bigger the thing you ask for in one go, the more of it is wrong. A generator that answers confidently at any size is not a feature.
So nothing here is synthesized a whole board at a time, nothing is built from a spec that has not cleared its gate, and no part reaches a netlist without its pin numbers having been read out of the manufacturer’s own datasheet and recorded. The board at the top of this page is what the alternative looks like.
Findings are graded, and the grades mean specific things. An error is something that will not work. A warning is something worth an argument. The checker also reports whether it finished at all, separately from what it found, so a run that broke halfway through says so instead of quietly reporting nothing wrong.
Shares of 115 failed LLM‑synthesized designs, by category. This is the order the review checklist is worked in, most common first, rather than top‑to‑bottom through the schematic. It is also an honest map of what the rule pack does and does not reach.
A pin map is only worth what the document behind it is worth. Every part carries the document number, revision, and hash of the file the pins were read from, fetched from the manufacturer's own domain. Aggregator and reseller mirrors are refused, because a mirror cannot tell you authoritatively which revision you are reading.
A gate that passes everything manufactures false confidence, and so does a page that only lists what a gate catches. These are the current limits, and they are kept in the repository as a file that has to be updated in the same change that adds a skip or a waiver.
A test coupon is a real thing. Board fabricators carry a small extra strip along the edge of every panel, built in the same run as the boards themselves. It gets cut off, sliced through, and measured, so the plating and the drilling can be checked without destroying anything anyone paid for.
It is the part of the panel that exists to be tested instead of trusted. That is the mark above, and it is the whole idea.
Creating an account joins the waitlist. There is no trial to start and nothing to configure yet; access is granted by invite while the rule pack is still being built out.