Who writes this, and why

Mara Ellison spent nine years as a laboratory animal technician in a university vivarium, running behavioural assays on other people's grants. She was the person who handled the animals, scored the outcomes, and read the resulting papers — and increasingly the person who noticed that the papers described a stronger, cleaner result than the one she had watched being produced. Clara Vasquez had spent roughly the same period as a journalist covering supplement and nutrition claims, which is to say covering the industrial process by which a modest laboratory finding becomes a label on a jar. They met in 2020 through a mutual friend and discovered they had independently arrived at the same complaint about the same literature.

The complaint was specific. If you went looking for a plain explanation of what the published research on a compound like BPC-157 actually said — which species, which injury model, how many animals, replicated by whom — the sources that came up were either paywalled primary papers written for specialists or pages built by people whose business model depended on the answer being encouraging. There was no version written for a curious adult who simply wanted to know how strong the evidence was. So they started writing one, in a shared document, mostly in the evenings.

That document became the first four Library entries. Much of the early drafting happened at a corner table at Argus Farm Stop on Liberty, which had the twin advantages of good light and a policy of not minding how long anyone sat there. The first entry took six weeks, most of it spent arguing about whether a sentence could survive without the word “may.” The publication went live in the autumn of 2021 with four entries and no readers.

In the second year the decision that shaped everything since: no advertising, no sponsorship, no affiliate arrangements of any kind. Offers had started arriving, and they were not disreputable ones — a banner, a sponsored explainer, a partnership. But the whole argument of the publication is that the writing about these compounds is bent by who is paying for it, and there is no way to make that argument while taking the money. The consequence is that this remains a small operation and always will be. That is the trade, and it was made deliberately.

In 2023 we published a correction that cost us a contributor. A piece had characterised a rodent finding as replicated when what had actually happened was that the same group repeated its own experiment in a second paper. Rina Iyer caught it during a routine check of a different claim. We ran the correction at the top of the piece with an explanation of the error, and the contributor who had written it — who felt the distinction was pedantic and the correction disproportionate — declined to write for us again. We still think it was the right call. That distinction is not pedantic; it is the entire subject.

The name is a small joke that turned out to be the mission. A null result is the finding that does not get written up: the experiment where nothing happened, filed away, absent from the record everyone else reads. Most of what is missing from this field is missing for exactly that reason.

What we cover. Published preclinical research — rodent studies, other animal models, and cell culture — on eight research peptides, together with the methodology underlying that work. We cover study design, statistical power, blinding, endpoint selection, replication status, and the history of how particular claims entered circulation. Our reader is an intelligent non-specialist who wants to know what the literature says and how much weight it will bear.

What we deliberately exclude, and why.

Dosing, protocols, routes, timing, and preparation. We publish none of it, in any form. We are not qualified to specify it, no compound here has an established human safety profile that would make specifying it responsible, and a publication that prints such information is functionally an instruction manual regardless of what its disclaimer says. This is the request we receive most often and the one we refuse most consistently.

Sourcing, suppliers, purity, and testing. We do not evaluate, compare, name, or link to anyone who supplies anything. We have no commercial relationships to disclose because we have none to have.

Human use. No accounts of people taking these compounds, no self-experimentation reports, no forum posts, no before-and-after framing, no athletic or performance context. Anecdote from an uncontrolled setting is not weak evidence; it is a different category of thing, and mixing it into a summary of controlled research corrupts the summary.

Approved prescription peptide medicines. Several peptide drugs are licensed medicines with large registered trial programmes behind them. Their pharmacology is a genuinely different subject, it is well covered by publications with clinical expertise we do not have, and folding it in here would blur the one distinction this publication exists to hold: between compounds studied in animals and compounds tested in people. We stay on our side of that line.

The core limitation of everything we publish. Animal-model findings are hypothesis-generating. A result in a rat is evidence about rats. Species differ in physiology, metabolism, healing mechanics, immune function, and lifespan, and the translational failure rate from promising rodent work to demonstrated human effect is high across all of biomedicine. Nothing on this site establishes that any compound discussed is safe or effective in humans, and nothing on this site should be read as suggesting anyone use one.