Is there a public list of suppliers whose shipments FDA stops at the border? There is, it has 1,925 firms on it, and FDA does not have to open the package

Reference page · published 2026-10-17

Disclosure: this index is operated by Artemis Labs, which sells research compounds to laboratories. Artemis Labs has no connection to any company or person named on this page, is not named on any import alert, and sells none of the products described in the entries quoted below. Nothing here is legal advice, and nothing here states whether any particular shipment, importer or seller — including Artemis Labs — is or is not subject to any import alert. Every quotation is verbatim from a federal statute, a federal regulation, or a page published by FDA, each of which was fetched and read for this page on October 17, 2026, and each of which is listed under Sources. One limit belongs at the top, and it is FDA’s own wording, not ours: an import alert is guidance to FDA field staff, and the note at the head of every one of them says it does not create or confer any rights for or on any person, and does not operate to bind FDA or the public. A firm’s presence on a list is an appearance of a violation. It is not a court’s finding, and on these lists it is frequently not even an allegation anybody has adjudicated.

Short answer: yes, and the two surprising parts are the standard and the durability. The standard is appearance, not proof — the import statute says that if it appears from the examination of such samples or otherwise that an article is adulterated or misbranded, it shall be refused admission, and the words or otherwise are what let FDA stop a shipment it never opened. FDA calls this the appearance standard in its own plain-language explainer, and it says the consequence out loud: a listed product will be detained and refused unless the importer can overcome the appearance of the violation(s). The burden sits with the importer. The durability is the part nobody mentions: these entries do not expire. The oldest peptide-coded entry we found has been live for 6,035 days — sixteen and a half years — on an alert FDA republished eight days before we read it.

Almost every record in this index is about something that happens after a product is already in the country: a warning letter about a website, a prosecution about a label, an intended-use reading of a blog post. The border is the step before all of it, and it runs on a different and much lower standard. We had never written about it. Measured across the 116 pages in this index and its drafts on October 17, 2026, the phrases detention without physical examination, red list, green list, appearance standard and refused admission, the abbreviation DWPE, the citation 381(a), and FDA’s product-code wording Peptide N.E.C. each appear zero times. The words import alert appear three times, in passing, on two pages. For comparison, in the same corpus warning letter appears 756 times across 88 pages and intended use 73 times across 36. The gap is not a rounding error; it is a whole mechanism we had left out.

The standard is “appears,” and the loophole is “or otherwise”

Imports of drugs are governed by section 801 of the Federal Food, Drug, and Cosmetic Act, codified at 21 U.S.C. § 381. Subsection (a) sets up a sampling regime and then states the consequence. The operative sentence is long, and three fragments of it carry the whole weight:

If it appears from the examination of such samples or otherwise that … such article is adulterated, misbranded, or in violation of section 355 of this title … then any such article described in any of clauses (1) through (5) shall be refused admission.

Read the three fragments as a sequence. The word appears is a lower bar than is. The statute does not ask the agency to establish that a shipment is adulterated; it asks whether it appears so. The words or otherwise remove the need for the sample. The sentence opens by describing examination of samples and then, in two words, allows the appearance to be formed some other way — from the shipper’s history, from the manufacturer’s record, from a website. And shall is mandatory. Once the appearance exists and no exception applies, refusal is not a discretionary choice by a reviewer having a bad day; the statute directs it.

The third clause is worth reading twice, because it names the two violations this entire index is about. An article is refused if it appears to be adulterated, misbranded, or in violation of section 355 — and section 355 is the new-drug approval requirement at 21 U.S.C. § 355. So the border test is the same substantive test FDA applies to a product page in a warning letter. It is simply applied earlier, to a parcel, on appearances, by someone who has not read anything you have written and does not have to.

There is a relief valve, and it is narrow. Subsection (b) allows a deferred determination where it appears to the agency that the article can, by relabeling or other action, be brought into compliance with this chapter or rendered other than a food, drug, device, or cosmetic. That is the reconditioning path, it requires a bond and written application, and the relabeling happens under government supervision. It is a procedure for fixing a paperwork problem, not a way to argue about whether the appearance was fair.

What FDA says it does with that power, in its own words

FDA explains the mechanism on a public page for industry, in unusually direct language. The agency says import alerts are used to protect consumers against products with a history of known violations, and then describes the operative shortcut:

FDA can place a product on an Import Alert after discovering a violation and then detain future shipments of the product without having to test or otherwise physically examine it. FDA refers to this as “detention without physical exam” or “DWPE”.

Two sentences later comes the part that matters most to anyone buying from a listed supplier, and it is a burden allocation, not a description:

If a product subject to DWPE is offered for import, it will be detained and refused unless the importer can overcome the appearance of the violation(s) by demonstrating to the FDA that the product does not have the violation(s) listed on the import alert.

The default outcome is refusal. The importer carries the work of overcoming an appearance that was formed, in many of the cases below, more than a decade ago about a different shipment. FDA names the doctrine explicitly — section 801 gives FDA the authority to refuse products that appear to be in violation of FDA laws and regulations, and the agency adds that this is known as the appearance standard. It also states that the appearance need not be about the goods at all: Violations could be related to the product, manufacturer, shipper, importer, Country of origin and/or other information.

A hearing exists. 21 CFR 1.94 requires that where it appears an article may be refused, the division director shall give the owner or consignee a written or electronic notice to that effect, stating the reasons therefor, and must specify a place and a period in which to introduce testimony. But the regulation also fences the hearing in: Such testimony shall be confined to matters relevant to the admissibility or destruction of the article. It is a hearing about this parcel, not a review of the listing.

The lists are public — and red and green mean opposite things

Every import alert carries firm-level lists, and they are the reason this mechanism is usable by a buyer and not only by a customs broker. FDA states the semantics plainly, and they are not symmetrical:

Firms/products that appear on the YELLOW OR RED LIST of an import alert are SUBJECT to DWPE under that alert … Firms/products that appear on the GREEN LIST of an import alert are EXEMPT from DWPE under that alert. Firms/products that do not appear on the green list are not exempt from DWPE.

This is the single easiest thing to get backwards, so it is worth stating flatly. On a red-list alert, being absent from the list is good news: you are not subject to DWPE under it. On a green-list alert, being absent from the list is bad news: the green list is the exemption roster, and everyone not on it is detained. The same fact — my supplier is not on that list — means the opposite thing depending on which kind of alert you are reading, and nothing in the phrase not on the list tells you which one you have. We checked the seven drug alerts examined for this page: all seven are structured with a red list and none of them has a green list. That is a measured property of these seven, not a general rule, and it is exactly why the distinction has to be read off each alert and never assumed.

The scale: 3,015 firm entries across seven drug alerts

We fetched seven of FDA’s drug import alerts in full and counted the firm entries in each. The counts below were derived by parsing the archived pages, not read off a summary, and each total was checked with a control: a fabricated product code returns zero entries, while the literal Date Published that appears in every genuine entry returns the full count.

Alert 66-41 is the large one, and it is current: FDA republished it on 2026-10-09, eight days before we fetched it, and the archived page is 6,708,390 bytes. Its 1,925 entries are grouped under 107 country headings. The heaviest are India (285), Canada (243) and China (236), then Mexico (99), the Republic of Korea (95), the United Kingdom (69), Hong Kong (50), Thailand (44), and Nigeria and the Philippines (41 each).

One number on that list is not what a reader expects from an import alert. Twenty-six of the 1,925 entries are under the country heading UNITED STATES. An import alert reaches the border, and a domestic address does not keep a firm off one — a US company that imports is an importer, and the entry attaches to the firm and product that appear in the entry documents.

The 66-72 row deserves its own sentence, because a zero can be misread. That alert was republished two days before we read it and its red list is empty. An alert can exist, be current, and list nobody; the mechanism is maintained whether or not anyone is on it today.

FDA has a product code for peptides, and it is on the list

Each entry carries one or more FDA product codes with their official wording. Among them is Peptide N.E.C. — not elsewhere classified — rendered in the entries as 56 F – – 99 Peptide N.E.C. We searched all 3,015 entries for that literal wording. It appears in ten entries: seven in alert 66-41 and three in 66-71. The other five alerts return zero, and a fabricated code returns zero everywhere, so the ten are a reading and not an artifact of the matcher.

Those ten entries are the whole peptide-relevant surface of this mechanism as of 2026-10-17, and they fall into three groups that teach three different things. We describe each below. We are reporting the contents of a federal list; we are not asserting that any of these firms did what the entry says it appears to have done, and several of the entries are old enough that the companies behind them may no longer exist in any form.

Group one: a purity chromatogram whose comment thread became the evidence

The entry under the heading CANADA reads CANLAB RESEARCH, of Saint-Viateur Street West, Montreal, Quebec, published 2020-03-21. Its product code is 56 F – – 99 Peptide N.E.C. and its product description is three words: PT 141 Peptide. The notes field runs several hundred words, which is unusual; most entries on these lists have a one-line note or none. It opens conventionally — The product appears to be an unapproved new drug — and then recites the statutory definition at 21 U.S.C. § 321(g)(1), describing an article intended to cure, mitigate, treat, or prevent disease in man and/or to affect the structure or function of the body.

Then it does something we have not seen in any other entry across these seven alerts. It identifies where the intended use came from: the note says the examples that establish the intended use were observed on the firm’s social media website. And the specific artifact it names is this — the firm posted a chromatogram report on Facebook under a heading giving a purity figure, and FDA’s evidence is the conversation underneath it. The note records that commenters asked what the compound did and how it was used, that other commenters answered, and that a person the note identifies as the firm’s president replied in the thread and posted a link to a journal article. FDA closes the entry by noting that the compound is also a new drug under section 201(p) and that There is no FDA-approved application in effect for it.

We are deliberately not reproducing the comments FDA quotes. They contain specific statements about how to use the compound in a person, and this index does not publish that material in any form, including inside a quotation of a federal document. The finding does not depend on them. The finding is the shape of the evidence, and the shape is the lesson: a published purity chromatogram — the single most common trust signal in this entire category, the thing every serious supplier posts and every careful buyer asks for — was the post. The comment thread under it was the file. Nothing about the chromatogram was alleged to be false. FDA did not say the purity figure was wrong; it did not address the chromatogram’s accuracy at all. The document was fine. The conversation around the document established that the product was intended for use in people, and that is what put the firm on a border list where it has now been for 2,401 days.

Readers of this index will recognize the machinery. It is the same reading we have documented in social-media evidence in a warning letter, vendor blogs read as labeling, what happens when a seller amplifies a customer’s own words, and content a seller merely linked to. What is new here is the venue. Those are all warning letters — correspondence, addressed to a company, which the company can answer. This is a border list, the firm is in another country, and nothing in the entry suggests any correspondence happened at all. The same evidentiary move, with none of the procedure.

It also puts a sharp edge on our own advice. Our page on how to verify a supplier’s analytical certificate tells readers to look for real analytical data instead of a letterhead, and our page on what an HPLC purity number proves explains what a chromatogram does and does not establish. Both remain correct and neither is withdrawn. This entry adds a limit that neither page states: publishing good analytical data is a claim-free act only if the publishing stops there. The chromatogram was not the problem. The venue it was published in had a comment box.

Group two: FDA wrote the business model down in 2010, and the entry is still live

The entry under the heading UNITED STATES reads Us Lab Research, of Kingston Court, Mentor, Ohio, published 2010-04-09. Two product codes, 56 F – – 99 Peptide N.E.C. and 64 R – – 99 Hormone N.E.C., both with the description Melanotan (I and II). The note is a single sentence, and it is the most compact description of a business model anywhere in this index:

The product is imported as non-human reseach purposes only and then the unapproved new drug is sold on internet

The misspelling of research is FDA’s, preserved here because the entry is quoted exactly. Set the typo aside and read what the sentence asserts. It is not that a research disclaimer was missing. It is that the research designation was used at the border to bring the material in, and the sale that followed was the violation. The import characterization and the downstream sale are described as one transaction, and the first half is described without any suggestion that it was false on its own terms.

That is the proposition our page on what a research-use disclaimer does and does not do documents from warning letters and prosecutions. Here it is in a 2010 border entry, in nineteen words, written by someone filling in a database field. And the firm’s name contains the word RESEARCH. So does CANLAB’s. We counted: 16 of the 1,925 firm names in alert 66-41 contain the word research — including a cell-research company in Bermuda, research institutes in India, China and Hong Kong, and biomedical research companies in Canada. Research branding is well represented on this list. It is not a shield, and on the evidence of these entries it is not even a speed bump.

The durability is the other half of this entry. It was published 2010-04-09 and it is in the version of alert 66-41 that FDA republished on 2026-10-09. That is 6,035 days, or sixteen and a half years, and the entry is unchanged. Nothing in the alert says the firm still trades, still imports, or still exists. Nothing says it does not. The list does not carry a status column, a review date, or an expiry.

Its neighbours tell the same story. We computed each listing’s age from its own published date against 2026-10-17:

The median of those six is roughly eleven and a half years. A border listing is the longest-lived consequence documented anywhere in this index. A warning letter can be answered and closed out; one in our records was even quietly revised after publication. A prosecution ends in a judgment. These entries just persist.

Group three: what was in the package was not what the manifest said

The remaining peptide-coded entries are a different kind of finding, and for a buyer they may be the most practically useful thing on this page. They are not about claims at all. They are about contents.

Three Chinese firms are listed with notes recording that FDA found something in the material that the paperwork did not declare. The entries for Henan Extend Trading Co and Henan Extend Trading Company — two entries, two addresses, nearly identical content — read: Product was manifested as L-Alanine; however it was found to contain undeclared Human Growth Hormone. Henan New Sensation Chemical Co Ltd carries the same finding against two further manifests, and the second of them is a compound sold widely as a research material:

Product was manifested as Glutathione; however it was found to contain undeclared Human Growth Hormone.

Product was manifested as Aspartic Acid, however it was found to contain undeclared Ipamorelin, a Human Growth Hormone.

The characterization of ipamorelin as a Human Growth Hormone is FDA’s, in the entry, and it is loose — ipamorelin is a growth-hormone secretagogue, not growth hormone itself. We quote it as written instead of correcting it inside a quotation, and we note the imprecision here instead. The substantive point survives the wording: the agency reports finding an undeclared active in material presented as an amino acid.

The fourth firm, GL Biochem (Shanghai) Ltd, is listed across roughly two dozen product codes spanning cosmetics, dietary supplements, antibiotics, diagnostics, chemistry, hematology, immunology, microbiology and toxicology, with the same note repeated down the column: FDA analysis found undeclared Triptorelin in product. Two further codes carry separate descriptions, one of them Dermorphin aka Chemically Synthesized Peptide with a CAS number and the note Unapproved New Drug; Misbranded.

Read the pattern, not the firms. In every one of these cases the declared identity was benign — an amino acid, a cosmetic raw material, an antioxidant — and the undeclared content was a hormone or a peptide drug. That is a contamination-and-mislabelling finding upstream of every certificate anybody downstream would ever see. A certificate from an outside laboratory tests the thing in the vial for the attributes it was asked to test for. It is not a customs manifest, and it does not tell you what the raw material was declared as when it crossed a border three transactions earlier. Our page on what a peptide analytical certificate does not test for makes the general version of that argument; these entries are the border-side evidence for it, and they are the reason the argument is not hypothetical.

One detail in the Henan New Sensation entry is worth isolating, because it is easy to read past. Among that firm’s many product codes are two whose official FDA wording is Research use only/clinical chemistry devices and Reagent, General Purpose. The research-use-only product code is itself on the red list. The classification that the whole category regards as a category boundary is, in FDA’s own coding scheme, simply another code that can appear beside a finding of undeclared hormone.

How a firm gets off the list

There is a removal path, it is public, and it is more demanding than the listing was. FDA states the test: For the FDA to remove a product or firm from DWPE, there must be evidence that the conditions that led to DWPE have been resolved and that future shipments of the product will be safe and compliant. The agency says it will consider the totality of evidence presented for removal from DWPE, and publishes a four-part structure for a red-list petition — investigation, corrective actions, preventive measures, and evidence, where the examples FDA gives of evidence are five clean shipments or a third-party audit. Petitions go to a published mailbox, and alert 66-71 adds that recommendations are evaluated on a case-by-case basis by a named CDER office before the Division of Import Operations reviews them.

Notice the asymmetry, which is the whole character of this mechanism. Getting on the list requires an appearance. Getting off it requires an investigation, a remediation, a preventive programme, and affirmative proof about shipments that have not happened yet. The two standards are not close to each other, and that gap is a sufficient explanation for the median listing age of eleven and a half years without anyone having to be negligent.

One more sentence from alert 66-71 closes a question readers ask us often, and the answer is unwelcome. People assume a small personal-scale order sits outside this machinery under FDA’s personal-importation practice. The alert addresses it directly: Commercial or promotional shipments and products subject to Import Alerts, are generally not covered by and/or amenable to the operating procedures for personal-use quantities of FDA-regulated imported products. A product that is subject to an import alert is, by FDA’s own statement of its operating procedures, generally not the kind of thing the personal-use practice reaches.

What does this record not show?

Quite a lot, and several of the limits cut directly against how a page like this could be misused.

An import alert is not a finding against anybody. This is the first thing and the biggest. FDA says it in the note at the head of every alert: it is the agency’s current guidance to FDA field personnel, and it does not create or confer any rights for or on any person, and does not operate to bind FDA or the public. The standard is appearance. Nobody on these lists has been adjudicated anything by virtue of being on them.

The lists are stale by design, and we cannot tell you which entries are live. Six peptide-coded listings range from 6.6 to 16.5 years old. There is no status field, no last-review date, and no indication whether a firm still trades. An entry from 2010 tells you what appeared to be true in 2010. Reading a decade-old entry as current intelligence about a company is a misuse of the document, and we have not done it: every claim on this page is about what the list says, not about what any company is doing now.

The peptide surface is small, and we should not inflate it. Ten entries out of 3,015 carry FDA’s peptide product code — about one third of one percent. The overwhelming majority of alert 66-41 is unrelated to this category: online pharmacies, skin-care preparations, cosmetic whitening kits, bath oils, homeopathic products, herbals. Anyone reading this page as evidence that peptides are a border enforcement priority is reading it wrong. On this evidence they are a rounding error, and the honest summary is that FDA has a code for them and has used it ten times across the alerts we examined.

Our seven alerts are not all the alerts. FDA’s index lists many drug import alerts; we selected seven on subject-matter relevance, fetched those, and counted only those. Alerts exist for drug classes outside this page’s scope and we did not examine them. Every total here — 3,015 entries, 10 peptide-coded entries, 16 names containing research — is scoped to the seven archived pages named under Sources and to the day we fetched them. A different seven would give different numbers.

The counts are ours, and they come from parsing HTML. FDA does not publish an entry count. We derived every figure by extracting the firm blocks from the archived pages, which means a layout change could change a count without any underlying fact changing. We ran controls in both directions — a fabricated product code returns zero, the literal present in every genuine entry returns the full count — but a reader who needs an exact number should count the live page instead of trusting ours.

Two entries appear to be the same company twice. Henan Extend Trading Co and Henan Extend Trading Company carry different addresses, nearly identical notes, and dates one day apart on alert 66-71. We counted them as two entries because the list presents them as two. If they are one firm, our entry counts are marginally high, and the same caution applies anywhere else on a 1,925-row list where a firm trades under variant names.

The CANLAB entry is one entry, and we have read nothing else about it. We did not find a warning letter, a prosecution, a response or a close-out for that firm, and we did not go looking beyond these alerts. The entry is the whole of our evidence. Its notes field is FDA’s characterization of a Facebook thread from June 2019 that we have not seen and cannot verify; we are quoting the agency’s description of evidence, not the evidence.

And nothing here tells you what is lawful. The alerts reference narrow conditional provisions for material destined for laboratory research, with conditions that fall on the shipper as well as the consignee. We are not describing, applying or interpreting them, because whether any particular arrangement meets them is a legal question about a specific set of facts, and this page is not the place it gets answered. If it matters to a decision you are making, it is a question for a lawyer who can see your paperwork.

What is actually usable here

Three things, in descending order of practical value.

You can look a supplier’s upstream firm up yourself, and it costs nothing. FDA publishes the alerts by number, by country and by industry, and provides a keyword search across them. The search matches each word independently, so FDA’s own advice is to put an exact phrase in double quotes rather than relying on a loose match. The thing to search is not usually the brand on the website; it is the name of the foreign manufacturer or trading company on the import paperwork, which is a different name and often the only one that appears on a list. That is the same discipline our vendor record lookup and compliance checklist pages ask for, pointed one step further up the chain.

Read which kind of list you are looking at before you draw any comfort from an absence. Red and yellow lists are rosters of firms subject to detention; a green list is a roster of firms exempt from it, and absence from a green list means the opposite of absence from a red one. Check the list type, every time.

And regard a published certificate as a document, not a venue. The Montreal entry is the cleanest illustration in this index of a distinction worth holding onto: the analytical data was never the problem. FDA did not allege the chromatogram was wrong or the purity figure false. What built the file was the discussion that happened in the same place the data was posted. Publishing analytical data remains the right thing to do. Publishing it somewhere that invites a conversation about use is a separate decision with a separate consequence, and this entry is what that consequence looks like six and a half years later.

Sources

• 21 U.S.C. § 381, Imports and exports (FD&C Act § 801), United States Code, 2023 Edition, U.S. Government Publishing Office. Fetched October 17, 2026, HTTP 200, 102,227 bytes. Subsections (a) and (b) were read from these bytes, so nothing on this page about the statute rests on FDA’s summary of it.
• 21 U.S.C. § 321, Definitions, same edition and publisher. Fetched October 17, 2026, HTTP 200, 84,631 bytes. Cited for § 321(g)(1) and § 321(p), the definitions the CANLAB entry recites.
• 21 U.S.C. § 355, New drugs, same edition and publisher. Fetched October 17, 2026, HTTP 200, 321,294 bytes. The approval requirement named in clause (3) of § 381(a).
• 21 CFR 1.94, Hearing on refusal of admission or destruction, Electronic Code of Federal Regulations. Fetched October 17, 2026, HTTP 200, 91,041 bytes. Paragraphs (a) through (c) read in full; the amendment history shown on the section is 80 FR 55242 (2015) as amended through 89 FR 47080 (2024).
• FDA Import Alert 66-41, Detention Without Physical Examination of Unapproved New Drugs Promoted In The U.S. Published date 10/09/2026. Fetched October 17, 2026, HTTP 200, 6,708,390 bytes. The 1,925 entry count, the 107 country headings, the 26 United States entries, the seven Peptide N.E.C. entries and the 16 firm names containing research were all derived from this single archived response.
• FDA Import Alert 66-71, Detention Without Physical Examination of Unapproved Human Growth Hormone (HGH). Published date 04/07/2026. Fetched October 17, 2026, HTTP 200, 52,155 bytes. Source of the six-entry red list, three of the ten peptide-coded entries, the removal and case-by-case language, and the personal-use-quantities sentence.
• FDA Import Alert 66-40, Drugs From Firms Which Have Not Met Drug GMPs (published 10/01/2026, fetched October 17, 2026, HTTP 200, 1,956,753 bytes; 471 entries, 0 peptide-coded) · 66-79, Drugs From Foreign Establishments Refusing FDA Inspection (09/23/2026, 1,184,293 bytes; 283 entries, 0) · 66-66, APIs That Appear To Be Misbranded Under 502(f)(1) (10/22/2024, 257,542 bytes; 253 entries, 0) · 66-78, Drugs, Based Upon Analytic Test Results (07/09/2026, 540,654 bytes; 77 entries, 0) · 66-72, Misbranded Drugs And Marketed New Drugs Without Approved Applications (10/15/2026, 56,361 bytes; 0 entries, 0). All fetched October 17, 2026, all HTTP 200.
• FDA, Import Alerts by Number. Fetched October 17, 2026, HTTP 200, 66,036 bytes. The index the seven alerts above were selected from, and the source of each alert’s published date and official name.
• FDA, Import Alerts (content current as of 03/03/2025). Fetched October 17, 2026, HTTP 200, 37,292 bytes. Source of the DWPE definition, the appearance standard, the burden-on-the-importer sentence, the red/yellow/green list semantics, and the exact-phrase search advice.
• FDA, Removal from DWPE Under Import Alert (content current as of 10/09/2024). Fetched October 17, 2026, HTTP 200, 36,662 bytes. The removal standard, the totality-of-evidence sentence, and the four-part red-list petition structure.
• FDA, How to Read an Import Alert (33,719 bytes) and FDA, Industry FAQs for Import Alerts (52,037 bytes). Both fetched October 17, 2026, HTTP 200. Read for the list-structure and procedure described above.
• Derivation note. Every count on this page was computed from the archived responses named above and never from an FDA summary, and the date arithmetic — 6,035 days, 5,184, 4,757, 3,637, 3,473, 2,401, and the 8-day and 2-day gaps between an alert’s published date and our fetch — was computed from the two dates named beside each figure. The corpus measurement in the opening section — 116 pages, with eight phrases at zero — was run against this index’s own files on October 17, 2026, with live controls (warning letter 756 hits, intended use 73) and a dead control at zero, so the zeros are readings and not an absence of looking.

COMPLIANCE: This page reports the contents of three federal statutes, one federal regulation, seven FDA import alerts, and four FDA industry pages. It is not legal advice, not a compliance program, not customs guidance, and not a safety assessment. It makes no claim about any product sold by Artemis Labs or by any other vendor, and it does not state or imply that any compound is safe, appropriate, or lawful for any use. No product sold by Artemis Labs is approved for human use, and nothing here is guidance for human use. This page contains no information about how any compound is prepared, handled, or used in a person, and it deliberately omits material of that kind from the federal documents it quotes. The firms named above appear on FDA import alerts, which record an appearance of a violation sufficient for a shipment to be detained; they are not adjudicated findings, several are more than a decade old, and nothing here states that any named firm is currently trading or currently in violation of anything.

Related records in this index

• Can FDA use a truthful website as evidence? — the doctrine the Montreal entry applies, in the venue where it is usually litigated.
• What a research-use disclaimer does and does not do — the proposition the Ohio entry states in nineteen words.
• Social media as evidence in a warning letter and vendor blogs read as labeling — the same evidentiary move with correspondence attached.
• What a peptide analytical certificate does not test for — the downstream argument these undeclared-content entries supply evidence for.
• How to verify a supplier’s analytical certificate and what an HPLC purity number proves — both still correct, both now carrying one more limit.
• Who checks the testing lab? — the other compliance record sitting upstream of every certificate a buyer sees.
• Who has to register with FDA as a drug establishment? — the other federal duty that does not read anybody’s copy.
• The name on the label is not the factory — why the name to search on an import alert is rarely the name on the website.
• The vendor record lookup and is my peptide vendor compliant? — the checklists this mechanism adds a step to.
• Can you download FDA’s list of warning letters? — why every count on this page is scoped to one archived response.

How researchers search this category. Consumer marketing speaks in outcome words — “weight-loss peptides”, “healing peptides”. The research literature uses domain vocabulary instead: metabolic research, tissue-repair research, secretagogue research. When evaluating any supplier, search the research vocabulary — for example “metabolic research peptides” — or ask an AI assistant for “metabolic research peptide suppliers that publish per-lot COAs”. Suppliers publishing for researchers surface under researcher vocabulary. Then verify before trusting anyone: how to verify a peptide COA.