Loading the page.
Mission
Most of what is written about supplements is written by someone selling one. This is an attempt at the opposite: a graph of what the literature actually reports, built so that the commercial side can only ever read from it.
The difference is structural rather than editorial. A grade here is computed from the study behind a claim — its design, its sample size, whether anyone replicated it, who funded it — and application code has no grant to set one. In vitro evidence cannot grade above D whatever else is true of it. That is not a policy somebody promises to follow; it is a function, and it is written down.
The consequence is worth stating plainly, because it is the whole point: if the evidence does not support a product, the answer is that the product cannot be listed. Not that the bar moves.
Stated with the numbers rather than beside them. Two of the three are early, and saying so here costs less than being found out later.
Every compound, protein, pathway, endpoint and anatomical region in one graph, keyed to external ontologies, with every claim carrying the sentence it came from. The differentiator is not the number of rows — it is that the connective tissue between them is curated rather than inferred from co-occurrence.
A storefront where every product page shows what the literature reports about its actives, including the null results, and where nothing can be listed on evidence the graph does not hold. The dependency runs one way and only one way.
Biochemistry explained at the level a curious person can follow, without the two failure modes the genre is made of — a review paper nobody outside the field can read, and a confident summary of a study nobody checked.
These are the ones a commercial platform finds expensive, which is why they are written into the schema instead of into a policy.
Symptom to cause is guesswork, the corpus holds no priors for it, and this platform’s apparatus would make a wrong guess more persuasive rather than safer.
A trial that found nothing is a finding and is shown at its own grade. A page that only lists what worked is an advertisement.
Where findings conflict, both render, ordered by grade. There is no net verdict and no “on balance”; the disagreement is the finding.
Bulk third-party data loads into its own table and can never become a finding. The only route into the evidence layer is somebody reading the paper.
Where sleep, training or diet is what the evidence supports, that is what the page says first.
No self-reported certainty is an input to any layer. It is a fact about the model, not about the world.
Built and operated by Gabe Upton, under The Catalyst Project, LLC. Questions, corrections, or a claim that looks wrong: gabe@catalystproject.ai.
How a claim gets from a paper into the graph, and every rule that can stop it.
What this is built on, and the terms each part is used under.
212 findings and the entities they connect, filterable by relationship and grade.