3 kits · read in dependency order · save $48

Three readings of your database.Only one order makes them mean something.

These are usually sold as three separate opinions about your database. They're not — they sit on top of each other. The Sanity Bench reads the shape of your tables, the RLS Read-Off reads the rules written on them, and What You Now Hold inventories what you're storing about people. Bought together and read in dependency order, $48 less than buying them separately.

Get the Bundle — $149$197save $48
Three kits · one rollup
The Data Model Sanity Bench$69
The Supabase RLS Read-Off$69
What You Now Hold$59
Bundle vs. separately$197
Bundle price$149
01.Why The Data Layer

Green, green, green — and none of it means what you think.

Each kit is right about its own half. The rules kit reads the policies on your tables and reports them accurately. The bench reads the shape of the tables and reports that accurately too. Neither is wrong. Neither can see the other.

But one of them depends on the other existing. A rule about who may read a row has to compare that row to somebody — if no column names an owner, there's nothing to compare against. That's not a weak rule; it's a rule with no operand, and "policies present, owner-scoped" describes a shape that can't hold what it claims. Nothing on a single kit's page tells you that — it's not a fact about either kit, it's a fact about the order they sit in.

02.What's Inside

Three readings, stacked in dependency order.

Each tool is the same full product sold on its own page — nothing trimmed for the bundle. Click any card to see the individual product page.

$69

The Data Model Sanity Bench

Reads the shape of your tables — whether the model can hold a rule at all.

View product
$69

The Supabase RLS Read-Off

Reads the rules written on those tables — who is actually allowed to read a row.

View product
$59

What You Now Hold

Inventories what you're storing about people, and whether you could delete it.

View product

Bought separately: $197 · Bundle: $149 — What You Now Hold is effectively free.

03.Why They Belong Together

They sit on top of each other, not beside each other.

Model, read first

The Sanity Bench reads the shape of your tables — whether a column even exists to name an owner. It's the prerequisite everything else is written against, so the rollup reads it first.

Rules, read in that light

The RLS Read-Off reports its policies accurately on its own terms. But a rule about who may read a row has to compare that row to somebody — with no owner column, the rollup rewrites the reading rather than averaging it in.

Holding, read independently

What You Now Hold inventories every field you store about a person and whether you could delete it. Invalidation never runs upward into this layer — it's an honest count, not conditioned on the other two.

When the model reads NOTHING TO BUILD ON, the rollup rewrites the rules reading to LOCKED AGAINST NOTHING however clean it looked alone — and prints the original beside it. That's a stronger claim than worst-of, and the invalidation only ever runs downward.

04.The Standard

Three rules the rollup will not bend.

Verdicts land as DATA LAYER HOLDS, GAPS IN THE LAYER, or NOTHING UNDERNEATH IT — ties break to the earliest layer, and anything unsupplied reads NOT RUN and blocks the top verdict.

A reading gets rewritten, not outvoted

This isn't worst-of. Worst-of would print TABLES LOCKED and NOTHING TO BUILD ON side by side and take the lower. This rewrites the rules reading to LOCKED AGAINST NOTHING and prints what it was before — the claim is stronger: that reading was never measuring what you thought it was.

Invalidation only runs downward

A broken rules layer doesn't rewrite the model reading, and nothing rewrites your holding inventory. The model is the prerequisite; the others aren't prerequisites for it. Pushing both ways would launder one verdict through another.

A check you didn't run isn't a check that passed

Any layer you don't supply reads NOT RUN, and the rollup will not clear DATA LAYER HOLDS while a layer is missing — two clean of three isn't clean. Ties break to the earliest layer. What You Now Hold names no law and makes no legal determination.

05.Who It's For

For the person whose database was written for them.

Builders who turned on row-level rules

You flipped RLS on and want to know whether it's actually doing anything — or just present and unenforced against a table with nothing to enforce it on.

Teams whose schema grew feature by feature

The tables were generated one migration at a time and nobody has sat down with all of them at once. The Sanity Bench is the read you've been putting off.

Anyone who stores data about people

You couldn't produce the list of what you hold about a customer without opening the database. What You Now Hold is that list, honestly counted — no law named, no determination made.

06.The Bundle

One purchase. Three kits. Save $48.

  • The Data Model Sanity Bench$69
  • The Supabase RLS Read-Off$69
  • What You Now Hold$59
  • Bought separately$197
$149
Save $48 · 24% off
Get the Bundle
Lifetime access · 30-day refund

Already own one of these? Each tool also stands alone — buy the bundle only if two or more fit how you work. See all RedHub bundles for the rest of the Past the Prototype line.

07.Common Questions

Answers before you ask.

Because one of them is a prerequisite for another. A rule about who may read a row has to compare that row to somebody — if no column names an owner, there is nothing to compare against. So the rules verdict is not merely worse in that situation, it is measuring something that cannot hold what it claims.

3 kits · save $48 · $149

Find out whether your rules
are locked against anything.

Three kits, read in dependency order, offline and connected to nothing. $48 less than buying them separately.

Get the Bundle — $149See all bundles

Sold by RedHub AI LLC · Secured by Stripe · redhub.ai