← Back to Scanner Test Hub

Greenwashing Scan Test Hub

This hub links to a set of fixtures for the environmental-marketing wording compliance scanner. Each sub-page below isolates one scanner behaviour — a wording bucket, a disclaimer distance, a stripped region, an alt-text caption, a character encoding, a word boundary, a language, or an output limit. The Expected column states a hard, checkable outcome for the automatic pass; anything phrased as a review outcome is a should-level expectation and may occasionally differ in practice.

FixturePurposeExpected
red-bareMarketing lines using red-bucket wording with nothing else nearby.8 red-bucket flags, all violation.
yellow-bareMarketing lines using yellow-bucket wording with no explanation anywhere on the page.5 yellow-bucket flags, no disclaimer, all violation.
yellow-disclaimedThe same style of yellow-bucket wording, each followed immediately by an explanation.Yellow-bucket flags with a disclaimer present, verdict compliant.
yellow-link-onlyYellow-bucket wording whose explanation exists only behind a link, not in the surrounding wording.2 yellow-bucket flags, no disclaimer, violation.
yellow-disclaimer-farA yellow-bucket claim followed by its explanation many paragraphs later.1 yellow-bucket flag, disclaimer out of range, violation.
permitted-termsPermitted-bucket wording attached to specific, defined claims.6 permitted-bucket flags, all compliant.
longest-matchOverlapping wording patterns where only the longer pattern should register.3 flags total (one permitted-bucket, two red-bucket); shorter overlaps excluded.
normalisationThe same wording written with different punctuation and character encodings.8 flags, one per variant, all treated as identical wording.
word-boundariesLook-alike wording that shares letters with flagged terms but means something else.Zero flags; page still counted as reviewed.
chrome-strippedFlagged wording placed only inside navigation, banners, overlays, and similar regions.Zero flags; only the plain body text is read.
header-keptA page header without a landmark role, holding one flagged line.1 red-bucket flag from the header text.
alt-textFlagged wording placed in an image caption versus a link title, a meta tag, and structured data.1 red-bucket flag from the image caption only.
split-elementsFlagged wording split across separate inline containers, with and without bold styling.1 yellow-bucket flag; the bold-styled line is a known gap.
tiny-mainA landmark region too short to count, forcing a fallback to the full page.1 red-bucket flag, found via the fallback.
germanGerman-language wording covering word endings, alternate spelling, and combined words.6 flags (four red-bucket, two yellow-bucket), German language.
swedishSwedish-language wording the current word list does not cover yet.Zero flags under the current word list; language gap noted.
known-gapsOne recognised claim alongside several phrases the word list does not cover yet.1 flag from the recognised claim; the rest are gaps.
context-verdictsFlagged wording used inside a denial and inside a quoted attribution.2 flags; explanation should clear both on review.
stressA single page with far more flagged lines than a reviewer can process in one pass.330 flags; page marked for manual review.