Food additive scanner app: four tools and the E number gap

A food additive scanner app names the additives in a product, either from a barcode or from ingredient text you paste. Four are worth comparing in the US: Yuka, Open Food Facts, Bobby Approved and LabelPeek. All four run into the same wall — Red 40 on a US label is E129 in Europe.

This page covers what each app matches, how the two naming systems line up, and why a barcode that resolves in one country can return nothing for the same food in another. LabelPeek is our own tool, so its limits get a named section rather than a footnote at the bottom.

What a food additive scanner app can tell you

A food additive scanner app does one job well: it names substances. Whether it reads a barcode or pasted text, the useful output is a list of additives found, matched against some reference. Everything after that — a score, a colour band, an approve-or-avoid verdict — is an editorial layer sitting on top of the match.

Those three things separate one scanner from another far more than the camera does. The wider field, including tools built around nutrition rather than additives, sits in best food scanner apps, and every head-to-head on this site is indexed in the scanner comparison hub.

Red 40 and E129 are one dye with two names

Red 40 and E129 name the same synthetic dye. The US declares certified colours by their FD&C name, the EU by an E number, and a product crossing the Atlantic changes its printed ingredient list without changing its formula. An app indexing only one naming system misses the other on sight.

The documents behind the first three rows are named, not summarised: the EU statement requirement is Regulation (EC) No 1333/2008, Annex V; the California date comes from the California School Food Safety Act, AB 2316 (2024); the titanium dioxide removal is Commission Regulation (EU) 2022/63, against 21 CFR 73.575 in the US. The longer arguments live in what the Red 40 rules say and in titanium dioxide under two regulators.

That ratio is the practical shape of the vocabulary problem. Colours, sweeteners and preservatives mostly hold numbers on both sides. Composite entries — refined oils, sugars under a dozen trade names, blanket phrases like natural flavors — often carry no number anywhere, so number-matching alone would miss the part of the label that is hardest to read.

Why a barcode that scans at home returns nothing abroad

A barcode identifies a pack, not a recipe. The same brand sold in two countries is two products with two codes, two label formats and two database records, so a scan that resolves in one market can return nothing in another — even when the food inside the wrapper is identical.

Three consequences follow for anyone choosing a scanner. Coverage is regional even when a database is global, so an app strong in Europe may be thin on US private-label goods. Naming follows the market, so a European record answers in E numbers a US shopper may not recognise. And a missing record cannot be argued with: no app can say anything about a product it has never seen. A text-based app to scan food ingredients trades the convenience of the camera for immunity to all three.

Four food additive scanners and the names each one knows

Four apps answer four different questions about additives. Yuka folds them into a score, Open Food Facts publishes the transcribed label, Bobby Approved applies one curator’s standard, and LabelPeek matches pasted text against a list printed on the same page. Each entry below states what it reads and where its vocabulary stops.

Yuka

Yuka scans a barcode and returns a 0–100 score with a colour band, covering food and cosmetics in one app, with 28 million users in the United States out of 85 million worldwide on its own press page. Additives appear inside that score as one weighted component rather than as the answer itself: the app names what it found, then folds the finding into a single figure. Premium, around $10 a year in the US, unlocks search without scanning, offline use, unlimited history and dietary filters, and neither tier carries advertising. For additive vocabulary the binding limit is the one every barcode app shares — the additive list comes from the product record, so a pack with no record produces no additive list. Yuka alternatives compared sets that model against three others.

Open Food Facts

Open Food Facts is a crowd-sourced database rather than a verdict engine, showing 4.6 million products on its homepage counter and publishing the whole dataset for download under the Open Database License 1.0. Contributors transcribe the ingredient list from the pack in front of them, which makes it the closest thing here to a multilingual additive dictionary: a substance appears as it was actually printed, in whichever market the photograph came from. The non-profit runs Open Beauty Facts as a sibling project for cosmetics. Searches for alternatives to it split in two directions, shoppers wanting an app and developers wanting the API, and the right answer depends on which you are. US private-label coverage depends entirely on whether a contributor has reached that shelf yet.

Bobby Approved

Bobby Approved is the FlavCity grocery scanner built by Bobby Parrish, and it needs separating from an unrelated subscription-tracking app also trading as Bobby before any comparison begins. It returns a curated approve-or-avoid verdict drawn from one person’s published ingredient standards, applied consistently across the American grocery aisle, with seed oils, synthetic dyes and certain preservatives among the criteria. Because the standard was written for US shelves, its working vocabulary is US label names rather than E numbers, which fits the market it serves. Whether one curator’s rule set is what you want is the real decision: seed oil scanner apps covers the criterion that draws the most argument, and alternatives to Bobby Approved covers what changes when a criteria list is published in full.

LabelPeek

LabelPeek has no app in any store as of 10 August 2026, and the web checker is what exists today. It takes pasted ingredient text and matches it against a hand-written reference of 78 entries, 45 of which carry an E number stored as a matching form beside the US name — paste E129, Allura Red AC or Red 40 Lake and the same entry answers. The output has three states rather than two: matched, recognised but not on the list, and not recognised at all, with the count of unrecognised entries printed instead of absorbed into a clean result. All 74 markers are printed on the same page as the ultra-processed food checker, carrying a version date. Nothing is uploaded and no account is required, because the match runs in your browser.

What naming an additive does not tell you

An app that names an additive has told you what is present, not what it does. Those are two different claims with two different standards of proof: the first is checkable against the printed list in front of you, the second belongs to toxicology and to the regulators reading it, and no scanner adjudicates that.

This is why the reference behind our checker separates two fields that most tools merge. One field records whether a substance is a marker of industrial production — a statement about manufacturing, verifiable from the ingredient list. A second records a regulatory fact with its jurisdiction, its effect and its date. Only 25 of the 78 entries carry anything in that second field, and an empty field means no restriction was recorded, not that a substance was cleared of anything.

Present and harmful are two sentences, not one. The first can be checked against the packet in your hand. The second needs a document, a jurisdiction and a date — and for most additives on most labels, no such document exists.

Where LabelPeek is the wrong choice

LabelPeek is the wrong choice in the aisle with a packet in your hand. There is no camera, no barcode lookup and no released app, so using it in a store would mean typing an ingredient list into a phone, which nobody does. A working barcode scanner wins that job outright, and no amount of citation quality changes it.

It is also the wrong tool for allergen decisions, where a purpose-built filter beats a general additive reference, and for anyone cooking from raw ingredients, who is buying food with no ingredient list to paste. The case it fits is narrow and specific: one label, read closely, against a list you can read yourself.

Checking one additive on a label you already have

Checking a single additive takes under a minute and needs no app at all. The method matters more than the tool: copy the whole list rather than the part that worries you, search both naming systems, and treat anything the reference did not recognise as an open question rather than a pass.

1. Step 1

   Copy the whole ingredient list

   Not the three names that caught your eye. Additives cluster at the end of a list, and colours are usually the last line of all, so a partial copy drops exactly the part you were checking.

2. Step 2

   Search both names for the same substance

   Look for the US name and the E number: Red 40 and E129, Yellow 5 and E102, titanium dioxide and E171. Imported and dual-market packaging frequently prints only one of the pair.

3. Step 3

   Read the unrecognised entries last

   Whatever a checker could not identify is the honest remainder of the label. Trade names, blends and collective phrases land here, and they are unresolved rather than absent.

None of this is medical advice, and none of it is a verdict on any named brand’s product. This page describes what four apps do, which names they match and where each one stops; whether a particular food suits you depends on your own circumstances, and on a professional when the question is a health one.