@deriv-com/dds-marketing 0.1.0 → 1.0.0

package/.llm/components/aspectratio.md

@@ -0,0 +1,220 @@
+<!-- GENERATED from aspect-ratio.contract.json. Do not edit. -->
+
+# AspectRatio
+
+A box that holds one shape whatever is put in it, so a page reserves the room a picture will take before the picture arrives. It draws nothing — no fill, no corner, no edge — and owns one decision: the proportion. What it holds fills it and is cropped to fit, because a box asked for a shape and given something a different shape has to answer which of the two wins, and the box is the one the layout was built on.
+
+It exists because the alternative is every page doing the arithmetic. A row of cards whose images arrive at three different shapes is a row that settles on its tallest, and a hero whose art loads late is a page that jumps once the browser learns how tall it is. Both are the same fault: the shape was known and nothing said it.
+
+## Use when
+
+- A picture whose proportions have to match the ones beside it — a row of cards, a grid of markets, a set of logos. One shape asked for once is what keeps a row from settling on whichever image is tallest
+- Room that must be reserved before the thing filling it arrives — art loaded late, an embed, anything fetched. The box is the right height from the first paint, so the page under it never moves
+- An embed that brings its own proportion and no height — a video, a map, a third-party frame. The ratio is the only way to give it one without guessing a pixel measure that is wrong at some width
+
+## Do not use
+
+- The picture is the media region of a card → `CardBody`
+- The thing inside has a size of its own and should keep it → nothing here does this
+- The shape changes with the viewport → nothing here does this
+- Spacing. It reserves a shape, not room around one: a box used to push things apart is a margin written somewhere nobody will look for it
+- Text. Prose has no proportion of its own — it is as tall as it needs to be at the width it is given, and a box that crops it is a box that hides a sentence
+- Anything that must be read in full. It crops, and crops the same way every time regardless of what matters in the picture: a chart with a legend at its foot loses the legend, and nothing reports it
+- Making an image smaller. The box decides a shape and not a size — how much room it takes is the layout's, and a ratio set to make something fit is a width written in the wrong units
+- A box whose height is set and whose width is not. The ratio turns a width into a height, and this one takes all of the width it is given — so a height set on top of that is a second answer to the same question, and the box is drawn off ratio with nothing reporting it. Where the height is the known measure, the width has to be released for the ratio to work backwards from it
+- A parent that sizes to its contents — an inline-flex row, a fit-content box, a float. The width comes from the container, and a container measuring itself by what is inside it has none to give: the box contributes no intrinsic width, resolves to nothing, and collapses. It is the same failure as drawing nothing, and looks identical to not having been asked for
+
+## Anatomy
+
+AspectRatio renders one element.
+
+- `root` — `<div>`, class `dds-aspectratio`. The box, and the whole of what this component is. It paints nothing and holds no corner: a shape is not a surface, and one that brought a fill of its own would be a second surface inside whatever it was put in — the same division Card makes about padding, one level out.
+  
+  It clips. A thing given a shape it does not have must lose something, and the alternative to cropping is letting it out of the box, which is the box not working. `overflow: hidden` is what makes the ratio a promise rather than a suggestion.
+  
+  What it holds fills it, which is stated in aspect-ratio-impl.css rather than bound here. The child is whatever the page passed — an `<img>`, an `<iframe>`, a `<video>` — so there is no part to put a class on, and a rule for it is a relationship between this element and its own child rather than a property of either. An image also needs `object-fit`, which is a keyword no token carries and a property only the thing being cropped can take.
+  
+  The shapes are tokens rather than values written here, which is what the schema insists on and what it should: a structure entry may hold a keyword and not a ratio, and a contract must bind something. It lands in the right place — that this package draws squares, sixteen-by-nines, four-by-fives and twenty-one-by-nines is a decision about the system rather than about one box, and a second component wanting the same shape now reads the same role.
+  
+  It has no size of its own. A ratio and a width give a height; a ratio alone gives nothing, so the box is as wide as whatever holds it and as tall as that width implies. That is why this is not the place to make a picture smaller — the layout owns the width, and this owns only what the width means for the height. It also means the box has two ways to draw nothing, both silent: a parent that sizes to its contents has no width to give it, and a height set from outside is answered by a width it already took in full.
+
+Apply these classes from the generated `aspectRatioParts` constant rather than writing the strings by hand.
+
+## Composition
+
+- Can contain: any content
+- Never contains: `AspectRatio`, `Card` — in children or in any prop that takes a node
+- Holds at least 1 and at most 1
+
+## Props
+
+### ratio — `portrait` | `square` | `landscape` | `widescreen` | `ultrawide` (default `widescreen`)
+
+- `portrait` — Taller than wide, for a picture read on a phone where the width is the constraint — a card whose image carries the card, a person, a product standing up. Four by five rather than anything narrower: past that the image takes the whole screen and the copy under it is below the fold
+- `square` — Equal sides, for a set that repeats — a grid of logos, a row of market tiles, an avatar’s frame. The one shape a grid can tile without the rows disagreeing, and the only one that reads the same in a column as in a row
+- `landscape` — Wider than tall and barely so — the shape a photograph or a screenshot arrives in before anything crops it. Reach for it where the picture was not composed for a wider slot and losing its top and foot would cost the subject
+- `widescreen` — The default, and the shape most moving media already is: an embedded video, a screen recording, a still from one. Sixteen by nine is what a camera and a capture both hand over, so asking for it usually means asking for what the file already was
+- `ultrawide` — Far wider than it is tall, for a hero or a section break where the picture is scenery rather than subject. Twenty-one by nine crops hard — most of a photograph’s height is gone — so it belongs to art chosen for it rather than to a picture that happens to be there
+
+Use when: The shape the box holds, at every width. It is the component's one decision and the reason it exists, so there is no case for leaving it unstated
+Never for: Fitting a particular image. The ratio is a decision about the layout — what shape this slot is — and the picture is what has to answer it. Choosing the value that crops one photograph least is how a row ends up with four slots at four shapes, which is the thing the box was put there to stop
+
+Also accepts everything from `HTMLAttributes<HTMLDivElement>`.
+
+## Choosing a variant
+
+Placement in, props out. When one of these describes what you are building, take
+its props — do not re-derive them from the variant descriptions above.
+
+- **A row of market cards whose images arrive at whatever shape the photographer left them** — `ratio="widescreen"`
+  One shape asked for once. Without it the row settles on its tallest image and every card below the fold is a different height from the one beside it
+- **A hero still that loads after the copy** — `ratio="ultrawide"`
+  The box is the right height from the first paint, so the page under it does not jump when the art arrives
+- **A grid of partner logos on a phone** — `ratio="square"`
+  A square is the one shape a grid can repeat without the rows disagreeing, and a logo that reads at any orientation is what it is for
+
+## Accessibility
+
+Guaranteed by the component. Write none of it by hand: the ids these attributes point at are generated per instance, so an attribute set from outside either says the same thing twice or points at nothing.
+
+- the box announces nothing of its own. It is a shape, and a shape is not something a reader needs told — what is inside carries the meaning, and an image inside it still needs its own alt text whether or not it is being cropped
+- what the crop removes is never the only place something is said. The box cuts the same way every time regardless of what is in the picture, so a chart, a diagram or anything with words in it loses whatever sat outside the shape — and loses it silently
+
+## Rules
+
+- **On a phone a section's actions stack and each fills the width, including when there is only one** — On a phone a page's actions stack, and each one fills the width. Not most of it, and not the width its label happens to need — the whole column.
+
+Two side by side is what this replaces. On the narrowest screen each gets half a column minus the gap, which is rarely enough for a real label: the words wrap inside the pills, the two buttons come out different heights, and the pair reads as a broken row rather than as a choice. Stacked, each has the whole width and neither has to shorten its words to fit.
+
+A single action fills too, and that is the part worth saying, because it is the one that looks unnecessary. A lone button has the room to sit at its natural width and still should not: a full-width control is the phone's convention for the thing the screen is asking you to do, and a pill hugging its label reads as one option among others that are not there. It also makes the target predictable — the whole width, wherever the thumb lands — which a centred pill does not.
+
+Filling is the row's job rather than the button's. A button has no width of its own and is not given one; the row it sits in becomes a column and stretches its children, so nothing about the control changes and the same button is full width here and its own size on a desktop.
+
+This is the phone's breakpoint, size.breakpoint.md, and not the wider one that decides whether a heading is centred. Those are different questions and they break at different widths: a heading stops being centred when the measure is the screen, while a button stops being full width much sooner than that — stretched across a tablet it is a long way from its own label, and a label centred in a metre of pill is harder to hit than one in a pill the size of the words.
+
+What this is not about is every row of controls. A set of filter chips, a pagination strip, a toolbar, a link sitting after a sentence — none of these is the page asking for something, and stacking them to full width would turn a strip the eye skims into a list it has to read. The test is whether the row is the call to action the section was written for.
+  Guidance only — cannot be checked automatically.
+- **Never nest one interactive element inside another** — Nested interactive elements produce ambiguous hit targets and an unreachable inner control for keyboard and screen-reader users. This is why the rule cannot live on any single component — it constrains Button, Link, Input and Switch simultaneously.
+  Guidance only — cannot be checked automatically.
+- **A section that paints a ground sets the ink to go with it** — A page sits on color.bg.page, and a section that separates itself from it takes one of three grounds. color.bg.subtle is the light one and needs nothing further — the page's own ink still reads on it. The other two are dark and do not: color.bg.brand-strong takes color.fg.on-brand-strong, and color.bg.brand takes color.fg.on-solid. Both are white, and both are written as the role rather than the white they resolve to. Set one of those backgrounds without its ink and the text inherits whatever the page was already using, which is near-black on a dark band — legible to nobody, and invisible to every check here, because each pair passes contrast on its own and nothing can tell whether a section used both.
+
+color.bg.brand carries a second limit that is not about forgetting. Its ink measures 3.39:1, which clears the 3:1 a heading needs and misses the 4.5:1 running copy needs. A coral band may carry a headline and a call to action; a paragraph on one is text the reader has to work through at a ratio the waiver on that role does not cover — the waiver says so itself. A section with copy in it takes color.bg.brand-strong, which is near-black and carries white at full contrast.
+
+color.bg.strong is not a section ground. It is the inverted fill for an action that has to outrank what surrounds it, and the components that paint it — Button and ButtonIcon in their inverse form — already set both halves in one rule, which is the pattern a hand-painted section has to copy.
+  Guidance only — cannot be checked automatically.
+- **Every gap inside a section names the relationship it expresses, and the same relationship keeps its measure when it turns on its side** — A section's spacing is named by the relationship it expresses, not chosen by size. Every gap in a section has a role token, and reaching for a step off the generic scale because it happens to be the right number is how two sections written by two people stop lining up: the numbers agree today and drift the first time one of them is tuned.
+
+The heading group is three steps, not one gap. An eyebrow sits nearer its title than the title sits to its description, and the actions sit further from the description than either — space.heading.eyebrow-title, space.heading.title-description, space.heading.description-actions, in that order. A single `gap` on the group cannot say this; it sets all three the same and the block reads as evenly spaced lines rather than as a mark, a name and a paragraph. The eyebrow has to identify itself for this to work, because a group that opens with a title must not give its description the eyebrow's tighter step.
+
+The step between a title and its description follows the size of the title, which is why there are two of it. What a reader sees is not the gap but the gap plus the leading above it, and leading scales with the type: under a display title the leading is most of the space, under a card's title it is half as deep. One value for both makes a card look loose at the size that made a section look right. So space.heading.title-description is the display step and space.heading.title-description-compact is the step at a card's size, and the choice is made by how big the title is rather than by what kind of block it sits in.
+
+Every one of those steps is smaller than the gap below the group, and that ordering is what holds the group together as one thing. space.heading.content is the step between a heading group and the content it introduces, and it is the widest measure inside a section: the one gap the eye reads as a change of subject rather than as a continuation.
+
+That step is the same whether the content is under the heading or beside it. A split is the same relationship turned on its side — a heading group and the content it introduces — so the space between the two halves is space.heading.content, not a gutter. When the split collapses to one column on a phone the gap becomes vertical again, which is where it started, and nothing has to change for that to be right.
+
+space.grid.gutter is between two things of the same kind. Two cards in a row are the same kind; the two halves of a split are not, which is why the gutter is the wrong measure there even when the number would have looked acceptable. The gutter runs on both axes, so a grid reads as an even field rather than as rows with a seam down them.
+
+space.inset.card is the card's, not the page's. A row of cards padded differently is a row of different cards, and the fill is the only edge they have. A page composing a row does not get to choose it.
+
+Column count is not a breakpoint decision. A row holds as many columns as fit before a column would be narrower than size.grid.column-min, and drops one when they would not. Written as a fixed count that collapses at a breakpoint, a four-up row keeps four columns at every width above the phone and sets its copy two or three words to a line through the whole of the tablet range — and on a narrow desktop, which no breakpoint sits on at all. The threshold is a property of the row rather than of a card: a card fills whatever cell it is given.
+
+What this does not cover is the space above and below a section, which is space.section.default and space.section.compact, and the page's own side margin, which is space.inset.page. Those are the section's outside; everything here is its inside.
+  Guidance only — cannot be checked automatically.
+- **A section's title is centred over stacked content, starts beside content in columns, and starts on a phone** — Two arrangements, and the title's alignment is decided by which one it is introducing rather than chosen per section.
+
+Content stacked under the title — a row of cards, an accordion, a set of steps — takes a centred title. The heading group and the block below it share one axis, so a centred title is symmetric with the thing it names, and the eye drops from the middle of the words to the middle of the set. A title starting at the left over a centred or full-width block reads as belonging to the first column rather than to all of them.
+
+Content beside the title — a split, a picture in the other half, anything laid out in columns — takes a title that starts where its content starts. The two share a left edge, and that edge is what says they are one thing read across rather than two stacked. Centring the title there breaks the only alignment the reader has to hold the halves together, and it drifts further from its content the wider the page gets.
+
+The heading group is what carries this, not the heading alone. An eyebrow, a title and a description are read as one block and align together; aligning the title without them leaves a centred heading over start-aligned copy, which reads as a mistake rather than as a choice.
+
+Centred means all of it, not just the words. A badge, a row of actions, a picture — these are flex children, and a flex child is placed by align-items rather than by text-align, so a block that only sets the text leaves everything that is not text at the start. A hero is the case that shows it: its title and copy centred while its badge and its buttons did not.
+
+A capped block centres itself as well as its words. A heading group takes size.heading-group.max, and a cap that is not centred leaves the measure at the start of the column with the words centred inside it — composed, and a long way left of the middle it appears to be in.
+
+Both arrangements go to the start on a phone. A centred block reverts below the breakpoint, and the reason is the measure: on a phone the measure is the screen, so a centred title wraps to three or four ragged lines and the eye loses the left edge it was tracking — the same symmetry that reads as composure on a desktop is what makes it hard to scan on a phone. Centring there has to be asked for twice, and a section title is not one of the things that asks: it is centred because there is room for it, and on a phone there is not. Going back to the start means all of it too, and a capped block is the one that hides a leftover: its words go to the start while its measure stays in the middle of the column, which reads as a title still centred over a badge and buttons that are not — the box in the middle and the line inside it at the edge.
+
+A hero is a section like any other: its badge, title, copy and actions are one stacked block, so it is centred on a desktop and starts on a phone with everything else.
+
+What this is not about is a page whose body is a form or a document. A title over a sign-up form or a guide is not introducing a set of content — it is the first line of the thing itself, and it keeps the alignment the body has, which for running text is the start. The distinction is whether the block below is content the title names or content the title is part of.
+  Guidance only — cannot be checked automatically.
+- **Anything a pointer has to hit is at least 24px in both directions** — A control sized by its contents is as big as its contents, and a glyph is chosen for how it reads rather than for how easily it is hit. An icon-only button with no padding is a 16px target — smaller than the pad of a finger, and smaller than the tremor of a hand using a mouse.
+
+WCAG 2.2 asks for 24 by 24 CSS pixels at AA. The exceptions are narrower than they look: inline targets inside a sentence are exempt, and so is a small target with 24px of clear space around it — but a clear button sits against the field it empties, and a field is itself a target, so the spacing exemption does not apply to the case that keeps arising.
+
+The target and the mark are separate decisions, and conflating them is what produces both faults. Growing the glyph to 24px makes it shout; leaving the button at 16px makes it hard to hit. So the glyph keeps the size that reads correctly beside its text, and the box around it declares a minimum of its own — which is what size.target.min is for.
+
+Declared rather than emergent. Padding plus contents does clear 24px in several components today, and it is the weaker statement: it is a size nobody wrote down, so it moves when the type scale or the glyph does, and nothing says it was meant to be a target at all.
+  Guidance only — cannot be checked automatically.
+- **A component is written along the reading axis, not along the page's left and right** — Every measure, inset and placement in this package names an axis rather than a side: inline-size and block-size, padding-inline and padding-block, inset-inline-start and inset-block-end, margin-inline: auto, text-align: start. Written that way a component follows whichever direction the page is read in, and a band whose claim ranges to the leading edge with its subject on the trailing one is the same composition in Arabic as in English, mirrored, with nothing restated. Written as left and right it is the same pixels in both, which is a layout nobody designed for the second language.
+
+The enum values follow the same rule and for the same reason. A prop that names where a picture sits takes start and end, never left and right — FeatureSplit, FeatureBullets and FeatureScroll all declare mediaPosition that way, and the flex directions they generate flip on their own because row already means the reading order. A value called left would have to be re-read as right by every author working in a mirrored page, which is a translation this system should be doing rather than asking for.
+
+The guidance a contract carries is part of the component, so it follows too. Describing a claim as ranging left is describing one of the two languages, and it was written that way here for a while: three of the opening bands said so in their own intent. A reader in a mirrored page meets that sentence and has to decide whether the component is wrong or the sentence is, and neither answer is one documentation should force.
+
+Two things genuinely have no logical form and are not covered by this. A CSS gradient takes a physical direction — there is no to inline-end — so a pattern that must mirror has to be stated twice under a direction query, and one that is periodic and centred, like PageHeader's rule grid, does not need to. A glyph that points the way onward is the other: an arrow is a picture of a direction, and mirroring it belongs to the icon set rather than to the component holding it.
+  Guidance only — cannot be checked automatically.
+
+## Anti-examples
+
+❌ AspectRatio for: The picture is the media region of a card
+   CardBody already holds a picture and takes an `aspect` of its own, so a ratio box around it is a second opinion about the same shape. Reach for this where the picture is not part of a card's anatomy — a hero, a grid cell, a band of art
+   → use `CardBody` instead.
+
+❌ AspectRatio for: The thing inside has a size of its own and should keep it
+   This crops to fill rather than fitting inside, so a logo that must not be cut is the wrong thing to put in one. Give it the room it needs and let the layout hold it — a shape forced onto something whose proportions are the point is the one case where the box is doing harm
+   → nothing in this system does this. Not a gap waiting to be filled — the case belongs outside it.
+
+❌ AspectRatio for: The shape changes with the viewport
+   The ratio is one value and holds at every width, which is what makes a row agree. A picture that should be wide on a desktop and square on a phone is two decisions, and the page is what knows the breakpoint — a contract holds no media query
+   → nothing in this system does this. Not a gap waiting to be filled — the case belongs outside it.
+
+## Example
+
+```tsx
+import { AspectRatio } from '@deriv-com/dds-marketing'
+
+<AspectRatio ratio="widescreen" />
+```
+
+## In use
+
+### The five shapes
+
+One decision, held at every width, and they run tallest to widest. The picture fills the box and is cropped to it, because a box asked for a shape has to win that argument.
+
+```tsx
+import { AspectRatio } from '@deriv-com/dds-marketing'
+
+// portrait — read on a phone, where width is the constraint
+<AspectRatio ratio="portrait" />
+
+// square — a grid that tiles without its rows disagreeing
+<AspectRatio ratio="square" />
+
+// landscape — barely wide, and the shape a photo arrives in
+<AspectRatio ratio="landscape" />
+
+// widescreen — the shape most moving media already is
+<AspectRatio ratio="widescreen" />
+
+// ultrawide — scenery rather than subject
+<AspectRatio ratio="ultrawide" />
+```
+
+### A row that agrees
+
+Three pictures at whatever shape they arrived, each given the same one. Without it the row settles on its tallest and every card below sits at a different height.
+
+```tsx
+import { AspectRatio } from '@deriv-com/dds-marketing'
+
+// First of three, all asked for the same shape
+<AspectRatio ratio="widescreen" />
+
+// Second
+<AspectRatio ratio="widescreen" />
+
+// Third
+<AspectRatio ratio="widescreen" />
+```

package/.llm/components/badge.md

@@ -0,0 +1,239 @@
+<!-- GENERATED from badge.contract.json. Do not edit. -->
+
+# Badge
+
+A short word marking something as new, chosen or reduced. It is promotion rather than state: a marketing page badges the plan it wants read first and the feature it shipped last week, where an app badges what an account is currently doing. That difference decides everything else here — there is no success or warning ground, because nothing on a landing page has a status, and the colour a badge takes says how loudly the page is asking rather than what it is reporting.
+
+Use it sparingly, and this is not a nicety. A badge works by being the one thing on a card that is not the copy; a page with a badge on every card has a row of ordinary labels in coloured pills and nothing marked at all.
+
+## Use when
+
+- One of a set is the one the page wants read first — the plan most people choose, the market most people start with
+- Something is new or has changed, and saying so is worth a word rather than a sentence
+- A number the page is promoting is worth marking as a saving rather than left in the copy — a percentage off, a period free
+
+## Do not use
+
+- What it says is a state the reader's account is in → nothing here does this
+- The words are the point rather than the mark → `CardBody`
+- A status. There is no success, warning or danger ground here on purpose: those roles report what happened to a reader who did something, and a landing page's reader has not done anything yet
+- Metadata that should be plain text. A date, a category, a count — a pill around each one turns a card into a row of labels and the badge into decoration
+- More than one per thing. Two badges on one card are two claims about which is the important one, and a reader believes neither
+- A control. It does nothing when pressed and takes no focus — something a reader acts on is CardButton, which is a target rather than a mark on one
+
+## Anatomy
+
+Badge renders 2 elements. A token binding names the part it applies to.
+
+- `root` — `<span>`, class `dds-badge`. The mark, and the whole of it. A <span> rather than a box holding a label: without a glyph beside the words there is nothing to group, and a second element would be a part that only ever repeats what its parent already said.
+  
+  A pill, at `radius.pill`, which is what separates it at a glance from the cards and bodies around it — every other surface in this package takes `radius.surface` or tighter, so a fully rounded end reads as a mark rather than as a small card.
+  
+  `inline-flex` with the items centred, so a badge sitting in a line of text sits on the line rather than on the baseline of its own contents. `white-space: nowrap`, because a badge is one or two words and a badge that wrapped would be a paragraph in a pill — if it is long enough to wrap it is the wrong component.
+  
+  It takes `font.label.sm`: fourteen, and the step the scale names for interactive text in a supporting position. A badge is not interactive, but it is supporting and it is a label, and the alternative at that size is `body.sm`, which is for disclaimers.
+  
+  Fourteen rather than sixteen because the copy it sits beside is sixteen: a badge set at the size of the sentence next to it has only its weight left to be read by, and a mark that has to be read is not a mark. Smaller and heavier is what makes it register first.
+  
+  It is a row now rather than a box of words: `inline-flex` was already here for the line box, and a glyph beside the label is what makes the gap worth naming. The words are not a part: they carry nothing the root does not already say, and a flex row treats bare text as an item like any other, so the gap falls between the glyph and them without an element in between to hang it on.
+  
+  `inline-size: fit-content`, which is the difference between a mark and a bar. `inline-flex` says how the badge lays out what is inside it and nothing about how it is laid out itself — put one in a flex column, which is what a hero is, and its display is blockified and it stretches the full width of the section with a pill drawn round it. A definite inline size is what stretch has nothing to do with.
+  
+  Twelve either side, and eight at the leading end when there is a glyph. Eight alone was the step for a badge that was only words, and it is short for a pill: at `radius.pill` the corner is half the height, so the ends are curving through twelve pixels and an eight-pixel inset starts the first letter inside the arc. Twelve is also what product's badge takes, and two packages drawing the same mark at different insets is a difference a reader can see and nobody chose.
+  
+  The glyph end keeps the eight, because a glyph is not a letter and does not need the same room. It arrives as a sixteen-pixel box with its mark drawn inside it, so it carries a margin a letter does not, and twelve in front of it reads as more air than the label's own end has. Eight is tight against the arc without being on it — four, which is the block inset, would put the mark hard against the curve at the one place the pill is narrowest. The words keep the twelve, so the pill is asymmetric exactly as far as its contents are.
+  
+  That asymmetry is the one thing here a binding cannot hold. What decides it is whether the page passed a glyph, which is a slot rather than an axis — there is no value to switch on — so `:has()` asks the question directly in the sheet beside the generated one, the way BottomSheet's header asks whether it has a leading end.
+  
+  This rather than `align-self`, which would also stop it: that would be the badge claiming where it sits in somebody else's layout, and the only thing it knows is that it is as wide as its own words.
+- `icon` — `<span>`, class `dds-badge__icon`. The glyph, when there is one. A <span> around it rather than the glyph itself, because Icon clones onto the SVG and this part is the place in the row rather than the mark in it.
+  
+  Drawn at `icon.size.xs` — sixteen in a twenty-four pill, which leaves four either side and is the largest step that does not make the badge a button. The step is the badge's: an Icon arriving already sized is unwrapped rather than obeyed. The root insets eight in front of it rather than the twelve it gives the words: a glyph's box carries its own margin, so the same inset would read as more air at the mark's end than at the label's.
+  
+  Alongside the label and never instead of it. A badge whose meaning is a shape is a badge a screen reader reads as nothing, and the glyph here is decorative for that reason.
+  
+  Regular weight, and nothing enforces it. A Quill glyph's weight is part of the name it is imported under, so `Solid` arrives here indistinguishable from `Regular` and draws heavier than the semibold words it sits beside — the mark outweighing the label it supports. The rule is in the prop's guidance because that is the only place it can be.
+  
+  Absent means absent, and `null` is how React says it. The slot is tested against both `null` and `undefined` rather than `undefined` alone: `icon={hasGlyph ? <GlobeRegularIcon /> : null}` is the ordinary way a page writes a conditional glyph, and read as present it renders an empty sixteen-pixel box and a four-pixel gap inside a pill only twenty-four tall — a dent in the mark with nothing in it. Rendered only when `icon` is set.
+
+Apply these classes from the generated `badgeParts` constant rather than writing the strings by hand.
+
+## Composition
+
+- Takes no children. Everything it renders comes from props
+- Never contains: `Badge`, `Card`, `CardButton`, `Accordion`, `AccordionItem` — in children or in any prop that takes a node
+
+## Props
+
+### variant — `accent` | `neutral` (default `accent`)
+
+- `accent` — The default, and the one a marketing page nearly always wants. A brand-coloured mark rather than a coral block: the tint carries the colour and the ink carries the words, which is what lets a badge sit on a card without competing with the card
+- `neutral` — The badge is naming a kind rather than making a claim — a category on an article, a market type on a card. It marks without asking, which is what lets a page use several without the set becoming loud
+Never for: Ranking two badges against each other. The variants are two different jobs rather than two volumes of one: neutral names a kind and accent makes a claim, so a set mixing them is not a set with a most-important member
+
+### label — `string` **Required.**
+
+Use when: Always — it is the whole component. One or two words: 'New', 'Most popular', 'Save 20%'. A string rather than a node, for the reason CardBody's title is one: a node prop cannot be given text in an examples row, and a badge nobody can show is worse than one that cannot hold a link
+Never for: A sentence. What does not fit on one line is not a badge, and `white-space: nowrap` will let it run out of its own card rather than wrap — which is the component saying so
+
+### icon — `node`
+
+Accepts: QuillSystemIcon
+Use when: A glyph that says what the badge is about faster than its word does — a globe on a badge reading 'For platforms', a spark on one reading 'New'. Pass the glyph; the badge draws it at its own step. Alongside the label and never instead of it.
+
+The regular weight, always: `GlobeRegularIcon` rather than `GlobeSolidIcon` or `GlobeLightIcon`. Icon has no weight prop and cannot have one — a Quill glyph's weight is chosen when it is imported — so this is the call site's to get right and there is nothing that will catch it
+Never for: Carrying the meaning. A badge is one or two words and a mark beside them; a badge that is only a mark says nothing to a screen reader and nothing to a reader who does not already know the shape.
+
+A solid or light glyph. Solid is a filled shape and reads heavier than the semibold words beside it, so the mark outweighs the label it was put there to support — and light loses its thinnest strokes at sixteen pixels, which is a mark drawn at a size it was not cut for
+
+Also accepts everything from `HTMLAttributes<HTMLSpanElement>`.
+
+## Accessibility
+
+Guaranteed by the component. Write none of it by hand: the ids these attributes point at are generated per instance, so an attribute set from outside either says the same thing twice or points at nothing.
+
+- The badge carries no role and no name of its own. Its words are read in the order they sit on the page, which is what a reader wants — a mark announced as a mark is a reader being told about the design rather than about the thing.
+- What it says is legible without its colour. A coral pill reading 'Most popular' says the same thing in a screen reader, in a printout and to a reader who cannot separate it from the ground; a badge whose meaning is the colour has not been written.
+
+## Rules
+
+- **On a phone a section's actions stack and each fills the width, including when there is only one** — On a phone a page's actions stack, and each one fills the width. Not most of it, and not the width its label happens to need — the whole column.
+
+Two side by side is what this replaces. On the narrowest screen each gets half a column minus the gap, which is rarely enough for a real label: the words wrap inside the pills, the two buttons come out different heights, and the pair reads as a broken row rather than as a choice. Stacked, each has the whole width and neither has to shorten its words to fit.
+
+A single action fills too, and that is the part worth saying, because it is the one that looks unnecessary. A lone button has the room to sit at its natural width and still should not: a full-width control is the phone's convention for the thing the screen is asking you to do, and a pill hugging its label reads as one option among others that are not there. It also makes the target predictable — the whole width, wherever the thumb lands — which a centred pill does not.
+
+Filling is the row's job rather than the button's. A button has no width of its own and is not given one; the row it sits in becomes a column and stretches its children, so nothing about the control changes and the same button is full width here and its own size on a desktop.
+
+This is the phone's breakpoint, size.breakpoint.md, and not the wider one that decides whether a heading is centred. Those are different questions and they break at different widths: a heading stops being centred when the measure is the screen, while a button stops being full width much sooner than that — stretched across a tablet it is a long way from its own label, and a label centred in a metre of pill is harder to hit than one in a pill the size of the words.
+
+What this is not about is every row of controls. A set of filter chips, a pagination strip, a toolbar, a link sitting after a sentence — none of these is the page asking for something, and stacking them to full width would turn a strip the eye skims into a list it has to read. The test is whether the row is the call to action the section was written for.
+  Guidance only — cannot be checked automatically.
+- **Never nest one interactive element inside another** — Nested interactive elements produce ambiguous hit targets and an unreachable inner control for keyboard and screen-reader users. This is why the rule cannot live on any single component — it constrains Button, Link, Input and Switch simultaneously.
+  Guidance only — cannot be checked automatically.
+- **A section that paints a ground sets the ink to go with it** — A page sits on color.bg.page, and a section that separates itself from it takes one of three grounds. color.bg.subtle is the light one and needs nothing further — the page's own ink still reads on it. The other two are dark and do not: color.bg.brand-strong takes color.fg.on-brand-strong, and color.bg.brand takes color.fg.on-solid. Both are white, and both are written as the role rather than the white they resolve to. Set one of those backgrounds without its ink and the text inherits whatever the page was already using, which is near-black on a dark band — legible to nobody, and invisible to every check here, because each pair passes contrast on its own and nothing can tell whether a section used both.
+
+color.bg.brand carries a second limit that is not about forgetting. Its ink measures 3.39:1, which clears the 3:1 a heading needs and misses the 4.5:1 running copy needs. A coral band may carry a headline and a call to action; a paragraph on one is text the reader has to work through at a ratio the waiver on that role does not cover — the waiver says so itself. A section with copy in it takes color.bg.brand-strong, which is near-black and carries white at full contrast.
+
+color.bg.strong is not a section ground. It is the inverted fill for an action that has to outrank what surrounds it, and the components that paint it — Button and ButtonIcon in their inverse form — already set both halves in one rule, which is the pattern a hand-painted section has to copy.
+  Guidance only — cannot be checked automatically.
+- **Every gap inside a section names the relationship it expresses, and the same relationship keeps its measure when it turns on its side** — A section's spacing is named by the relationship it expresses, not chosen by size. Every gap in a section has a role token, and reaching for a step off the generic scale because it happens to be the right number is how two sections written by two people stop lining up: the numbers agree today and drift the first time one of them is tuned.
+
+The heading group is three steps, not one gap. An eyebrow sits nearer its title than the title sits to its description, and the actions sit further from the description than either — space.heading.eyebrow-title, space.heading.title-description, space.heading.description-actions, in that order. A single `gap` on the group cannot say this; it sets all three the same and the block reads as evenly spaced lines rather than as a mark, a name and a paragraph. The eyebrow has to identify itself for this to work, because a group that opens with a title must not give its description the eyebrow's tighter step.
+
+The step between a title and its description follows the size of the title, which is why there are two of it. What a reader sees is not the gap but the gap plus the leading above it, and leading scales with the type: under a display title the leading is most of the space, under a card's title it is half as deep. One value for both makes a card look loose at the size that made a section look right. So space.heading.title-description is the display step and space.heading.title-description-compact is the step at a card's size, and the choice is made by how big the title is rather than by what kind of block it sits in.
+
+Every one of those steps is smaller than the gap below the group, and that ordering is what holds the group together as one thing. space.heading.content is the step between a heading group and the content it introduces, and it is the widest measure inside a section: the one gap the eye reads as a change of subject rather than as a continuation.
+
+That step is the same whether the content is under the heading or beside it. A split is the same relationship turned on its side — a heading group and the content it introduces — so the space between the two halves is space.heading.content, not a gutter. When the split collapses to one column on a phone the gap becomes vertical again, which is where it started, and nothing has to change for that to be right.
+
+space.grid.gutter is between two things of the same kind. Two cards in a row are the same kind; the two halves of a split are not, which is why the gutter is the wrong measure there even when the number would have looked acceptable. The gutter runs on both axes, so a grid reads as an even field rather than as rows with a seam down them.
+
+space.inset.card is the card's, not the page's. A row of cards padded differently is a row of different cards, and the fill is the only edge they have. A page composing a row does not get to choose it.
+
+Column count is not a breakpoint decision. A row holds as many columns as fit before a column would be narrower than size.grid.column-min, and drops one when they would not. Written as a fixed count that collapses at a breakpoint, a four-up row keeps four columns at every width above the phone and sets its copy two or three words to a line through the whole of the tablet range — and on a narrow desktop, which no breakpoint sits on at all. The threshold is a property of the row rather than of a card: a card fills whatever cell it is given.
+
+What this does not cover is the space above and below a section, which is space.section.default and space.section.compact, and the page's own side margin, which is space.inset.page. Those are the section's outside; everything here is its inside.
+  Guidance only — cannot be checked automatically.
+- **A section's title is centred over stacked content, starts beside content in columns, and starts on a phone** — Two arrangements, and the title's alignment is decided by which one it is introducing rather than chosen per section.
+
+Content stacked under the title — a row of cards, an accordion, a set of steps — takes a centred title. The heading group and the block below it share one axis, so a centred title is symmetric with the thing it names, and the eye drops from the middle of the words to the middle of the set. A title starting at the left over a centred or full-width block reads as belonging to the first column rather than to all of them.
+
+Content beside the title — a split, a picture in the other half, anything laid out in columns — takes a title that starts where its content starts. The two share a left edge, and that edge is what says they are one thing read across rather than two stacked. Centring the title there breaks the only alignment the reader has to hold the halves together, and it drifts further from its content the wider the page gets.
+
+The heading group is what carries this, not the heading alone. An eyebrow, a title and a description are read as one block and align together; aligning the title without them leaves a centred heading over start-aligned copy, which reads as a mistake rather than as a choice.
+
+Centred means all of it, not just the words. A badge, a row of actions, a picture — these are flex children, and a flex child is placed by align-items rather than by text-align, so a block that only sets the text leaves everything that is not text at the start. A hero is the case that shows it: its title and copy centred while its badge and its buttons did not.
+
+A capped block centres itself as well as its words. A heading group takes size.heading-group.max, and a cap that is not centred leaves the measure at the start of the column with the words centred inside it — composed, and a long way left of the middle it appears to be in.
+
+Both arrangements go to the start on a phone. A centred block reverts below the breakpoint, and the reason is the measure: on a phone the measure is the screen, so a centred title wraps to three or four ragged lines and the eye loses the left edge it was tracking — the same symmetry that reads as composure on a desktop is what makes it hard to scan on a phone. Centring there has to be asked for twice, and a section title is not one of the things that asks: it is centred because there is room for it, and on a phone there is not. Going back to the start means all of it too, and a capped block is the one that hides a leftover: its words go to the start while its measure stays in the middle of the column, which reads as a title still centred over a badge and buttons that are not — the box in the middle and the line inside it at the edge.
+
+A hero is a section like any other: its badge, title, copy and actions are one stacked block, so it is centred on a desktop and starts on a phone with everything else.
+
+What this is not about is a page whose body is a form or a document. A title over a sign-up form or a guide is not introducing a set of content — it is the first line of the thing itself, and it keeps the alignment the body has, which for running text is the start. The distinction is whether the block below is content the title names or content the title is part of.
+  Guidance only — cannot be checked automatically.
+- **Anything a pointer has to hit is at least 24px in both directions** — A control sized by its contents is as big as its contents, and a glyph is chosen for how it reads rather than for how easily it is hit. An icon-only button with no padding is a 16px target — smaller than the pad of a finger, and smaller than the tremor of a hand using a mouse.
+
+WCAG 2.2 asks for 24 by 24 CSS pixels at AA. The exceptions are narrower than they look: inline targets inside a sentence are exempt, and so is a small target with 24px of clear space around it — but a clear button sits against the field it empties, and a field is itself a target, so the spacing exemption does not apply to the case that keeps arising.
+
+The target and the mark are separate decisions, and conflating them is what produces both faults. Growing the glyph to 24px makes it shout; leaving the button at 16px makes it hard to hit. So the glyph keeps the size that reads correctly beside its text, and the box around it declares a minimum of its own — which is what size.target.min is for.
+
+Declared rather than emergent. Padding plus contents does clear 24px in several components today, and it is the weaker statement: it is a size nobody wrote down, so it moves when the type scale or the glyph does, and nothing says it was meant to be a target at all.
+  Guidance only — cannot be checked automatically.
+- **A component is written along the reading axis, not along the page's left and right** — Every measure, inset and placement in this package names an axis rather than a side: inline-size and block-size, padding-inline and padding-block, inset-inline-start and inset-block-end, margin-inline: auto, text-align: start. Written that way a component follows whichever direction the page is read in, and a band whose claim ranges to the leading edge with its subject on the trailing one is the same composition in Arabic as in English, mirrored, with nothing restated. Written as left and right it is the same pixels in both, which is a layout nobody designed for the second language.
+
+The enum values follow the same rule and for the same reason. A prop that names where a picture sits takes start and end, never left and right — FeatureSplit, FeatureBullets and FeatureScroll all declare mediaPosition that way, and the flex directions they generate flip on their own because row already means the reading order. A value called left would have to be re-read as right by every author working in a mirrored page, which is a translation this system should be doing rather than asking for.
+
+The guidance a contract carries is part of the component, so it follows too. Describing a claim as ranging left is describing one of the two languages, and it was written that way here for a while: three of the opening bands said so in their own intent. A reader in a mirrored page meets that sentence and has to decide whether the component is wrong or the sentence is, and neither answer is one documentation should force.
+
+Two things genuinely have no logical form and are not covered by this. A CSS gradient takes a physical direction — there is no to inline-end — so a pattern that must mirror has to be stated twice under a direction query, and one that is periodic and centred, like PageHeader's rule grid, does not need to. A glyph that points the way onward is the other: an arrow is a picture of a direction, and mirroring it belongs to the icon set rather than to the component holding it.
+  Guidance only — cannot be checked automatically.
+
+## Anti-examples
+
+❌ Badge for: What it says is a state the reader's account is in
+   Nothing here does this, which is the useful thing to know — a reader told to reach for a card reaches for one and is still wrong. A marketing page has no account to report on and no state to poll: something true of a visitor is copy on the page, not a pill. The status grounds exist in this package for a banner or a note that says what happened, which is a sentence rather than a word, and no component wraps one yet.
+   → nothing in this system does this. Not a gap waiting to be filled — the case belongs outside it.
+
+❌ Badge for: The words are the point rather than the mark
+   A badge is read before it is read properly — a glance, a colour and one or two words. Anything a reader has to finish is copy, and a pill around a sentence is a paragraph nobody can scan
+   → use `CardBody` instead.
+
+## Example
+
+```tsx
+import { Badge } from '@deriv-com/dds-marketing'
+import { GlobeRegularIcon } from '@deriv-com/quill-icons-web'
+
+<Badge label="For platforms" icon={<GlobeRegularIcon />} />
+```
+
+## In use
+
+### Two jobs, not two volumes
+
+Both are a tint carrying the colour and dark ink carrying the words. Brand makes a claim and neutral names a kind, so a set mixing them is not a set with a most-important member.
+
+```tsx
+import { Badge } from '@deriv-com/dds-marketing'
+import { ChartCandlestickRegularIcon, GlobeRegularIcon } from '@deriv-com/quill-icons-web'
+
+// accent — the plan the page wants read first
+<Badge variant="accent" label="Most popular" icon={<GlobeRegularIcon />} />
+
+// neutral — naming a kind rather than claiming
+<Badge variant="neutral" label="Forex" icon={<ChartCandlestickRegularIcon />} />
+```
+
+### A glyph, alongside and never instead
+
+The mark says what the badge is about faster than its word does. It is decorative — what it means is already in the label, so a screen reader passes over it.
+
+```tsx
+import { Badge } from '@deriv-com/dds-marketing'
+import { GlobeRegularIcon } from '@deriv-com/quill-icons-web'
+
+// With a glyph
+<Badge label="For platforms" icon={<GlobeRegularIcon />} />
+
+// Without — the ordinary case
+<Badge label="Most popular" />
+```
+
+### One or two words, and no more
+
+A badge is read before it is read properly: a glance, a colour, a word. It does not wrap, so anything needing a second line runs out of its card — which is the component saying it is copy.
+
+```tsx
+import { Badge } from '@deriv-com/dds-marketing'
+
+// New
+<Badge label="New" />
+
+// A saving, marked rather than left in the copy
+<Badge label="Save 20%" />
+
+// Two words is the ceiling
+<Badge label="Most popular" />
+```