Claimline is a claims editor for direct-to-consumer supplement and wellness marketing. It quotes the passage, names the rule, and hands the question back to you — because a rewritten health claim is an unsubstantiated claim with a new author.
$ python3 tests/verify.py PASS 7/7 cases, 238 checks run $ python3 tests/verify.py --selftest PASS 5/5 known-bad reviews correctly rejected
Catch what's actually there. Refuse to fix it for you. Say nothing when there's nothing to say. Click any finding to see the passage it came from.
If an editor rewrites a health claim, it has authored a claim nobody substantiated.
The replacement would be written by a reviewer that has never seen your study, doesn't know the dose in your finished product, doesn't know your study population, and can't read your label.
You'd ship language a machine invented. That's the exact failure the review existed to prevent, reintroduced by the reviewer.
So: you hold the substantiation file, so you write the claim. Claimline tells you what the current line does, which rule it runs into, and what would have to be true to keep it.
The one exception is quoting an agency's own published wording. When FDA states in the Federal Register that helps to maintain cholesterol levels that are already within the normal range is acceptable, quoting that is citing a source — and it's marked as one, never adapted to your product.
Severity maps to the body of law and the remedy, not to how bad the copy feels.
Matches a documented disease-claim criterion under 21 CFR 101.93(g)(2), or a per-se prohibition in 16 CFR Part 465.
Fixed by: nothing. A supplement may not make the claim, so no evidence rescues it.
May be permissible in form, but asserts a result, mechanism, or level of proof requiring evidence Claimline can't see.
Fixed by: your substantiation file. The finding names the specific evidence.
An endorsement, testimonial, or review missing a required disclosure, or making a typicality claim without the data behind it.
Fixed by: a disclosure, or the numbers to support one.
Sits on a line the agencies themselves say isn't bright, and a fact you hold would settle it.
Fixed by: answering. The finding names which fact decides it, and where each answer lands.
GRAY is not a soft STOP. STOP says the copy matches a criterion. GRAY says the reviewer isn't entitled to say — and names what would change that.
The most common error in this domain is returning one verdict. FTC's Health Products Compliance Guidance is explicit that DSHEA structure/function provisions "don't govern the FTC's assessment of those claims in advertising."
A line can be a compliant structure/function claim in labeling and a deceptive advertisement, because it's unsubstantiated or its net impression misleads.
So every finding names which regime it sits under. A passage that fails both produces two findings, with two citations and two resolution paths.
It's also why Claimline stops and asks whether your copy runs as labeling or as advertising before it reads anything. That's the only gap that blocks — everything else it's missing goes into the LIMITS block and the review proceeds.
A fabricated finding in this domain reads exactly like a real one. Same structure, same citation format, same confident tone. The only reliable way to tell them apart is whether the copy it describes actually exists — so that comparison is done mechanically, not by a reader already inclined to trust fluent regulatory prose.
python3 tests/verify.pyChecks seven shipped reviews against the contract in rules.md. The load-bearing check is Rule 0: every quoted passage must appear verbatim in the submitted draft.--selftestRuns five deliberately broken reviews and asserts each fails on the check it should. A verifier that passes everything proves nothing.--case 06 --output run.mdCheck your own run of the editor against the same contract.cold-test/Five blind runs by fresh instances that never saw the answer key, the nine defects they reported, and the seven that got fixed. One caught a finding the answer key had missed.The green highlight in the demo above is that check running in your browser — each quoted passage is searched for in the draft text, and reported as grounded or not.
A compliance tool that oversells itself is worse than none.
This one’s free and yours to keep. If you’d rather not do it yourself — tell me the job it would be doing for you.
Tell me the job →