What is the billing replay study?

A proposed $2,500 study of one recurring-subscription price change. Nobulex independently calculates the expected amount, compares it with a dry run, and leaves a regression test. It needs no production access or customer payment data. No customer has purchased or run it.

A separate prototype. Before an automated financial action runs, it checks the evidence that action depends on and returns permit, block, or escalate. It is not deployed and has no customer result.

Billing replay study

The match was the mistake.

Q: Is Nobulex in production?

No. The HTTP decision API and Observe Mode wrapper are implemented and tested locally. They are not deployed. Nobody is blocking live orders through Nobulex.

Billed $50. Expected $150.

The issue reporter says the dry run repeated a wrong $50 invoice. Our released-tag H2 test reproduced it.

See the study The reproduced case

Proposed $2,500 study. One scenario, one release test. No customer deployment yet.

Flat recurring

Mismatch

August 1 – September 1, 2026

Issued: $50
Owed: $150

Unseen$100

The preview and the invoice agreed. The approved price did not.

What happened

Nothing in the path disagreed.

01
Approved price

$150 for the month.
02
What the system said

Dry run and invoice: $50.
03
What was missing

An independent check of the approved price, before issuance.

The study

One price change, checked before release.

We calculate the expected amount ourselves, compare it with a dry run, and leave a regression test. No production access. No customer payment data.

Read the scope

Price$2,500, proposed
ScopeOne scenario
StatusNo buyer yet

Where this stands

Said plainly.

Built

Open-source checks for completeness, window, and truncation. Runnable today, with a published method.

Built locally

The HTTP decision API and Observe Mode wrapper are implemented and tested locally. They are not deployed or running in anyone's production path.

Not yet

No customer is blocking live orders through Nobulex. A separate gateway prototype returns permit, block, or escalate. It has no customer result.

Questions

Short answers.

What are you actually selling?

A proposed study of one billing change. Not a deployed billing product, and not a customer result.

Is anyone running this in production?

No. The checks are open source. The gateway is local. This page will keep saying so until that changes.

Does a signature prove the amount was right?

No. A signature shows who signed a decision. It cannot turn a wrong source into a right one.

Where is this going?

Billing accuracy for usage-based products. A fixed-fee audit now, a read-only scanner for self-contradictions in design, mutation checks in CI after that, and each layer gets built only once the one before it pays. The approach, with its limits stated.

Check one price before it ships.

Tell us how you check invoice amounts before release.

nobulex.dev@gmail.com