How to verify a Janoshik test at the source

About seven in ten of the lab reports in our data come from Janoshik, which makes it the lab most readers of this site will actually need to check. This post covers one narrow, useful skill: taking a Janoshik report you were shown and confirming, at the source, that it's real. Not how to read the chemistry on it — that's covered in how to read a peptide COA — just how to confirm the document itself hasn't been altered, reused, or misapplied.

Everything below is based on what Janoshik's own site actually does, checked directly. Where we couldn't confirm a detail, we've left it out rather than guessed.

What you're actually checking

Every Janoshik report carries two identifiers: a task number and a unique key. Janoshik's own homepage states that "every report in our database is verifiable via Verify tab in the menu or QR code on the report" — so the verification path is built into the document itself, not something you have to go hunting for separately. The tool that checks those identifiers lives at janoshik.com/verification.

Why two separate identifiers rather than one code? Our reading is that a task number alone is a small, often short reference — easy to guess or brute-force if it were the only thing standing between a forger and a valid-looking match. Pairing it with a longer, harder-to-guess unique key means a forger can't just try sequential task numbers and hope one returns a real result; both values have to be correct together, which in practice means both have to come from a genuine report. That's a reasonable design choice for a lookup tool that has no login wall protecting it, and it's worth knowing why the two fields exist rather than treating them as redundant.

What the verification page actually looks like

We navigated to the page directly to confirm this rather than describing it from memory. Janoshik's site runs a standard cookie-consent banner on first visit; past that, the verification page is minimal by design. Its main navigation lists Home, Services, Verify, Public Results, About Us, and Contact Us. The Verify page itself has one short line of copy — "Here you can verify the authenticity of any test conducted by our company" — two text fields labeled "Task number" and "Unique key," and a single "Verify" button. There's no account creation, no email requirement, and no CAPTCHA standing between you and a result. That simplicity is worth noting: a verification tool that's slow or gated behind a login is a tool people skip. One that takes ten seconds is one people might actually use.

Step by step

1. Find the task number and unique key on the report you were given. These are the two values the vendor's document should display. If a document claiming to be a Janoshik report doesn't show both, that's already worth pausing on — see fake peptide COAs for what a missing or cropped identifier usually means.

2. Go to janoshik.com/verification directly. Type the address yourself rather than following a link the vendor gave you. A link supplied by the party you're trying to verify isn't a trustworthy way to reach a verification tool.

3. Enter the task number and the unique key. The page has exactly two fields — "Task number" and "Unique key" — and a "Verify" button. There's no account or login required to use it.

4. Read the result. We tested this directly with a made-up task number and key. The page returned a plain, unambiguous message: "No test found for the given task number and unique key." A report whose identifiers don't produce a match at this tool should not be treated as genuine, full stop — there's no ambiguous middle ground in that response.

A report whose numbers do resolve should show you the actual test record tied to those identifiers. What matters at that point is whether the record matches what you were told: does it correspond to the product you were shown, and does anything about it look inconsistent with the document in front of you?

Common mistakes that produce a false non-match

Before concluding a report is fraudulent because the lookup failed, rule out the boring explanations first. The most common one is a transcription error: the unique key is a long string, and a single mistyped character, an extra trailing space copied from a PDF, or a capitalization mismatch (some fields may be case-sensitive) will produce the same "no test found" result as an actual fake. Retype both fields by hand rather than copy-pasting from a source you don't fully trust, and double-check against the source document once before concluding anything. That said, don't let this turn into talking yourself into trusting a report that genuinely doesn't verify — if you've checked carefully and it still doesn't resolve, the caution in this post applies.

Field by field: what to compare

Once a lookup resolves, the comparison that actually matters is between what the vendor's document claims and what the verified record says. Specifically:

• The product or compound. This is exactly where one documented case (covered in fake peptide COAs) fell apart — a certificate presented as an NAD+ result turned out, once checked, to correspond to a bacteriostatic-water test. A verified task number tied to the wrong product is not a technicality; it means the document you were shown does not describe what it claims to.
• The purity and amount figures. Confirm the numbers on the vendor's copy match what the verified record shows. An edited PDF built from a real report will usually still verify correctly at this stage if the forger didn't bother changing the key — the mismatch shows up in the numbers, not the lookup.
• The date. A verified, genuine report from a year ago tells you about a vial tested a year ago. It doesn't tell you anything about the batch being sold today.

Scanning the QR code versus typing manually

Janoshik's own homepage says reports are verifiable "via Verify tab in the menu or QR code on the report," which implies the QR code is simply a shortcut to the same lookup — pre-filling the task number and key rather than making you type them. Scanning it is faster when you have the physical document or a clean image of it. Typing the fields manually still matters in two situations: when you only have a photo or screenshot where the QR code doesn't scan cleanly, and when you specifically want to confirm that the QR code itself hasn't been swapped for one pointing somewhere else. If you have both a printed QR code and the typed task number and key on the same document, and they lead to different results when checked independently, that mismatch is itself informative — it would suggest the document was assembled from more than one source.

If the key doesn't resolve

Treat it as unverified, not as inconclusive. A non-match means one of a few things: the numbers were entered wrong (worth double-checking once), the document is fabricated, or a real report for something else has had its identifiers altered or misapplied. None of those are a reason to trust the document. If you believe you copied the numbers correctly and the tool still returns no match, the safest assumption is that the certificate is not what it claims to be.

Public results: Janoshik's own listing

Janoshik also publishes a searchable public archive at public.janoshik.com, linked from the main site as "Public results." Browsing it directly, it lists individual test entries — each with a task number, a short product description, who submitted the test, and, where given, a "Made By" line for the manufacturer. It also has its own search box for finding entries by keyword. The page carries its own disclaimer: "We do not endorse, sponsor, or have any affiliation with any of the companies, organizations, or products listed."

It's worth being precise about what this listing is and isn't. It's a public log of tests Janoshik has run and that someone chose to make visible — it is not a vendor ranking, and a listing there is not an endorsement of the product, the vendor, or the claims in the listing's title. Browsing a sample of entries makes this obvious quickly: many titles read like product listings, with shipping regions, purity claims and even contact details, because the vendor is often the one who submitted the sample and titled the entry. The "tested by" name on these listings is frequently the vendor itself rather than an independent buyer, which is a meaningfully different thing from a buyer-commissioned report — it tells you the vendor chose to have this specific sample tested and made the result public, which is useful information, but it isn't the same claim as "an independent buyer tested a vial they purchased and the result matched the label." Reading the "tested by" field before assuming a public listing represents independent buyer testing is worth the extra few seconds.

What Janoshik verification does not tell you

This is the part that's easy to skip past. A verified report, one that genuinely resolves and genuinely matches the product, still only tells you about the specific vial that was tested, on the day it was tested. It doesn't tell you:

• Whether the vial in front of you is the same batch. Verification confirms the report is real; it doesn't confirm which physical vial the report describes unless you also check the lot or batch number against what's printed on your vial.
• Whether the vendor who showed it to you is the vendor who actually sold that batch. A real, verified report can be genuinely misapplied to the wrong seller — attached to a listing for a vendor who didn't submit the sample, or reposted by someone other than the buyer who commissioned it. Verifying the report confirms the paperwork; it doesn't independently confirm the chain of custody between the lab and the listing you're looking at.
• Anything about a different batch, a different order, or a future purchase. One good result is a data point about one vial, not a guarantee about the next one.

Why this matters more than it might seem

It's tempting to treat verification as a formality — a box to check after you've already decided a vendor looks legitimate. In practice it's closer to the opposite: verification is cheapest and most useful to do before you've formed an opinion about the vendor, because that's exactly when you're least likely to notice something's wrong on your own. Roughly seven in ten reports in our data are Janoshik results, which means for most vendors most readers will look up, this two-field, ten-second check is available and free. There's very little reason not to run it on any Janoshik-attributed report before treating it as evidence of anything.

A short note on other labs

Roughly three in ten reports in our data come from labs other than Janoshik — Peptide Test/TrustPoint, Freedom Diagnostics, Chromate, SR Biolabs, Testides, Vanguard, and others. Several of them do publish a lookup: Freedom Diagnostics has a COA lookup that searches by accession number or company name, and the Chromate and Peptide Test sites each list a Verify page in their menus. We haven't tested how those tools work, so we won't describe them in detail. The same general principle from this post applies regardless: look up the lab independently of the vendor, and if the lab has any kind of public verification system, use it before trusting a document you didn't obtain yourself. If it doesn't, contacting the lab directly and asking them to confirm a specific report number is the fallback, as described in fake peptide COAs.

Verification is not the same as reading the report

It's worth separating two skills this site covers in two different posts, because it's easy to conflate them. Verification, the subject of this post, answers whether a document is genuine and belongs to the sample it claims to describe. Reading the report — covered in how to read a peptide COA — is about understanding what the numbers on a genuine report actually mean: what counts as a good purity figure, why a 99%-pure vial can still be badly underdosed, and what identity testing does and doesn't confirm. A report can pass verification and still describe a bad vial. Both skills matter, and neither substitutes for the other — a well-verified report of a failed test is still a failed test, and a beautifully clean-looking result that fails verification tells you nothing at all.

The takeaway

Verification answers one question well: is this specific document real, and does it describe the product it claims to? It's a narrow question, and it's worth answering before you look at a single purity number — because a beautifully formatted report that fails this check is worth nothing, and a plain one that passes it is worth taking seriously. These are compounds sold for research use only either way; verification tells you what the paperwork actually says, not whether using an unregulated compound is a good idea.