Editorial policy

Peptides101 exists because most peptide writing is quoting other peptide writing. This page describes exactly how we verify what we publish, so you can check whether we did.

Funding

peptides101.com carries no advertising, accepts no sponsorship, sells no products, and earns no commission on anything we write about.

This is deliberate, and it is the reason we can publish the rest of this page. The only findings worth reporting here are frequently negative — that a substance has no human data, that FDA proposed not to approve it, that a legal theory sellers rely on has already failed twice. Sites funded by the people selling these compounds cannot report that. We have nothing to sell you, so we can.

How we verify a regulatory claim

We read the primary document. Not a summary of it, not a press release about it, and not another site’s description of it.

This matters more than it sounds. fda.gov returns a 404 to ordinary automated fetchers and requires a browser user-agent to retrieve a PDF. That single technical detail is a large part of why the peptide corpus is wrong so often: the documents are genuinely public, and almost nobody opens them. We download the PDF, extract the text locally, and read the relevant section.

Every regulatory claim on this site carries the document it came from, that document’s own revision date, and the date we last checked it. Those last two are different things and we never merge them.

How we assess evidence

We never assign an evidence rating from a count of trials or publications. Counting is how you end up confidently wrong.

A worked example. Search the literature for MOTS-c and you will find what looks like four human randomised controlled trials; for TB-500, one. Open them, and every one measures the peptide already present in the body as a marker of exercise or disease. None administers anything to anyone. The real number of trials in which these compounds were given to humans is zero. A pipeline that counts studies reports four. A person who opens them reports zero.

So for every study we cite, we confirm the compound was actually administered to human participants. Our data model will not accept an evidence rating unless the author explicitly affirms this check was performed. If it was not, the rating is Not yet verified, and we say so on the page.

Trial registrations are not evidence

Anyone can register a study on ClinicalTrials.gov. Registration is self-reported and nobody vets it before it appears. In this vertical specifically, we have found registrations cloned field-for-field from a pharmaceutical company’s real protocol and relabelled as recruiting; one with “(Mock Study)” in its own official title; and one that states in its own summary that it is fictional — which vendor sites are already citing as human trial data.

We check sponsor, status and results before citing any registration, and we treat an empty results section as what it is.

What we will not publish

Dosing, reconstitution, or administration protocols. Not for any compound, including approved ones. This is the highest-intent search traffic in the category and we are leaving it on the table on purpose. Nothing we cover is an approved treatment for what people are using it for, and a dosing page is the point at which an information site starts practising medicine badly.
A fifty-state legality map. We can verify the law in two states. We are not going to invent the other forty-eight, and a page that has to be wrong to exist should not exist.
“Research shows X can improve Y.”Not a style preference. FDA has quoted exactly this construction from sellers’ websites as the violation itself in warning letters. The literature-review voice is not a shield — it is the exhibit. We attribute findings to the study that produced them, never to the compound.
A reviewer byline we have not earned. If a page has not been reviewed by a clinician, it does not say it has, and our structured data omits the review fields entirely rather than filling them in.

Categories are not safety ratings

FDA’s 503A categories describe where a substance sits in an evaluation process. They are not a verdict on whether it is safe, and they are not a verdict on whether possessing or selling it is legal. Those are separate questions answered by separate parts of the law: federal misbranding provisions turn on how a product was labelled and where the material came from, not on which category a substance sits in. A substance under evaluation and a substance in no category at all can carry identical exposure. The category is not the line.

We will never present a category as an endorsement, and no status label on this site uses the words “safe” or “legal”.

Dates

We publish four dates, because they mean four different things:

Written — when the page was first published.
Medically reviewed — when a named clinician last read it. Linked to their bio, with their credentials and conflict-of-interest statement.
Fact-checked — when the citations were last re-opened and confirmed to say what we claim they say.
Updated — when the file last changed, including for a typo.

A single “last updated” stamp conflates a spelling fix with a clinical re-read, so we do not use one. We also do not stamp the whole library with the same review date on the same day: it would look diligent and it would be false, and unlike most false things on the internet it is trivially checkable.

Review cadence

Legal status: monthly, plus whenever something happens. FDA revised the 503A list twice between April and May 2026. It moves.
Compounds with established human evidence: every 12 months.
Everything else: every 6 months.

Corrections

We will get something wrong. When we do, we will fix it and say so, with a date, on our corrections page — not silently in place.

If you think something here is wrong, please tell us: corrections@peptides101.com. If you can point at the primary document, we will act on it quickly. Correcting an error costs us nothing. Leaving one up costs us the only thing we have.

Contact

Editorial: editorial@peptides101.com