/**
 * MCMC — proposed stylesheet, layer 3 of 4: LAYOUT.
 *
 * SEAT REFACTOR-MAP, 10 September 2026 (design). FILLED FROM THE CENSUS BY SEAT REFACTOR-R0
 * the same day and corrected against pixels at 1440, 1024 and 390. PROPOSAL ONLY.
 *
 * ⛔⛔ THIS FILE REPLACES THE WHOLE `.e-con` ENGINE, AND THE ENGINE IS SMALLER THAN IT
 *    LOOKS. `vendor/elementor/assets/css/frontend.min.css` carries 520 rules; over the 79
 *    captures at 1440, 1024 and 390, 59 are alive and 54 win anything at all, and 143 of the
 *    520 sit inside a `(min-width:-1)` / `(max-width:-1)` media query that can never match.
 *
 * ⛔⛔ AND ITS METHOD IS AN INDIRECTION THIS FILE DELETES. Elementor's engine rules read
 *    `display: var(--display)`, `padding-inline: var(--padding-inline-start)`,
 *    `gap: var(--row-gap) var(--column-gap)`, and the per-document stylesheets set those
 *    custom properties per element id. Of the rules that win anything anywhere in this
 *    corpus, 450 — FOUR IN TEN — win NOTHING BUT custom properties. 177 of the 302 distinct
 *    winning property names begin with `--`. The proposed system writes the value where it
 *    applies and the indirection goes with the plugin.
 *
 * ⛔ EVERY DECLARATION BELOW NAMES THE RULE IT COMES FROM AND THAT RULE'S WINNING COUNT, or
 *    it names the pixels it fixes. A declaration with neither does not belong in this file —
 *    REFACTOR-MAP FINDINGS 12, applied to its own proposal.
 *
 * NAMING. One prefix, `mc-`. Block, `__element`, `--modifier`. A layout primitive is a noun
 * (`mc-shell`, `mc-row`, `mc-grid`); a component is what it is on the page (`mc-card`,
 * `mc-facet`, `mc-pager`); a utility says what it does (`mc-hide-mob`). NO ELEMENT CARRIES A
 * GENERATED ID CLASS: today 403 distinct `elementor-element-<hash>` classes exist across 79
 * pages and 4,300 elements wear one.
 */

/* ══ THE SHELL — a full-bleed band with a centred column ══════════════════════════════
   Elementor writes this as `.e-con.e-con-boxed > .e-con-inner`, two elements, and BOTH are
   load-bearing. A first draft of the proposal dropped the inner as "a single child with the
   same box" and un-centred the footer by 20,160 px in one full-width band, twice, on one
   page: the inner is the element carrying `max-width` and the auto margins, and the child
   fills what the parent allows. EQUAL BOXES ARE NOT THE TEST; CONSTRAINING NOTHING IS.

   from `.e-con`                          (323,296 pairs) : position · width · min-width
   from `.e-con` (the second block)       (…)             : padding-inline 10px
   from `:is(…,[data-elementor-id]) > .e-con` (3,026)     : margin-inline auto · max-width
   ⚠ `--padding-left/right` default to 10px on `.e-con` and are consumed by
     `padding-inline-start/end: var(--padding-inline-start/end)` on the same selector. The
     10 px is the kit's `--container-default-padding-*` fallback, not an authored value, and
     it is what puts the story header's inner column at x=150 in a 1440 viewport:
     10 + (1420 − 1140)/2. */
.mc-shell {
  position: relative;
  width: 100%;
  min-width: 0;
  max-width: 100%;
  margin-inline: auto;
  padding-inline: 10px;
  display: flex;
}
/* `.e-con-boxed.e-flex` (2,050): a BOXED flex band is a COLUMN. The row/column of the
   CONTENT is the inner's, and the inner carries `mc-row` or `mc-stack` for it. */
.mc-shell { flex-direction: column; flex-wrap: nowrap; align-content: normal; align-items: normal; justify-content: normal; }
.mc-shell--grid { display: grid; }
/* `.e-con-boxed.e-grid` (2,226) */
.mc-shell--grid { grid-template-columns: 1fr; grid-template-rows: 1fr; justify-items: legacy; }
/* ⛔ AND `.e-con.e-grid` (3,160) GIVES EVERY GRID CONTAINER `--grid-justify-content:start`
   AND `--grid-align-content:start`, WHICH `.mc-grid` CARRIES AND `.mc-shell--grid` DID NOT —
   SEAT REFACTOR-R3, 11 Sept 2026. `.mc-shell` above states `justify-content: normal` and
   `align-content: normal` for the FLEX case, and on a grid those two are not inert: they are
   the grid's own content distribution. Measured on the player page's three boxed grid bands:
   arm A `start`, arm B `normal`, on `a90722a`, `2e37d31` and `4186f06`. Invisible until a
   band's tracks are smaller than the band. */
.mc-shell--grid { justify-content: start; align-content: start; }

/* ══ FLOW PRIMITIVES ══════════════════════════════════════════════════════════════════
   from `.elementor-element:where(.e-con-full, .elementor-widget)` (61,833), which is the
   rule that turns `--flex-direction` / `--justify-content` / `--align-items` / `--gap` into
   real declarations, and `.e-con.e-flex` (27,058) whose `--flex-direction` default is
   `column`. In a system that writes the value where it applies, that is two class names. */
.mc-stack { display: flex; flex-direction: column; }
.mc-row   { display: flex; flex-direction: row; }
/* ⛔ BANDED, 13 SEPT 2026, SEAT REFACTOR-R9: `.mc-row--wrap` and `.mc-stack--wrap` (`flex-wrap: wrap`, the OPT-IN the
   note below calls "the two `--wrap` modifiers above") stood here. Each is on 0 of the 82 captured pages, in no string of any shipped
   script and in no code token of any PHP file (`reports/REFACTOR-R9/<ts>-tools/deadrules-r9.py`,
   step 0); deleted by owner decision of 12 Sept 2026. */
/* ⛔⛔ EVERY FLEX CONTAINER WRAPS BELOW 768 AND IT IS A DEFAULT, NOT A CHOICE — SEAT
   REFACTOR-R3, 11 Sept 2026. `frontend.min.css` carries
   `@media (max-width:767px){.e-con.e-flex{--width:100%;--flex-wrap:var(--flex-wrap-mobile)}}`
   over `.e-con{--flex-wrap-mobile:wrap}`, so the mobile default of EVERY flex container on
   this site is `wrap`; `--flex-wrap-mobile` exists to let a single container opt out, and
   `post-497.css` uses it on fourteen. The two `--wrap` modifiers above carry the OPT-IN and
   there was nothing carrying the default.
   ⛔ IT IS INVISIBLE AT THE THREE GATE WIDTHS AND AT NINE OF THIS UNIT's TWELVE. Measured on
   `/player/drake-maye/`: 11 of 12 widths pixel zero and **71,526 px at 600**, where fourteen
   flex containers read `nowrap` against the served `wrap` and every box below them sat
   2.53 px high. 390 and 375 are zero because the columns there are single-item anyway.
   This is REFACTOR-R1's 576–767 band a second time, from the other side. */
@media (max-width: 767px) {
  .mc-stack, .mc-row { flex-wrap: wrap; }
}
.mc-block { display: block; }

/* ══ THE CONTAINER DEFAULT — SEAT REFACTOR-R1's FIX, 11 September 2026 ════════════════
   ⛔⛔ THIS FILE CARRIED THE `.e-con` PADDING FOR THE SHELL AND THE INNER AND STOPPED
      THERE, AND THE GAP IT LEFT WAS INVISIBLE WHILE A MEASURED `width` SAT ON THE
      ELEMENT. Elementor's `.e-con` consumes `--padding-top/right/bottom/left` — whose kit
      fallback is 10px — and `--gap` (the kit's `--widgets-spacing`, 20px) on EVERY
      container, not only on a top-level band and its inner.
      MEASURED by SEAT REFACTOR-R6B on the served chrome at 1440: the header's logo column
      is 114 px wide and its widget 94 (10 px each side), the footer's logo column 81 and
      its widget 61; without the padding the header logo renders **112 px wide instead of
      75.19**, and without the gap the footer is **213.844 px tall where the served one is
      233.844**. R6B restated both on the chrome's six column classes in `05-chrome.css`
      because fixing them was outside its authorisation; those two rules are GONE from that
      file as of this unit and this is the fix they pointed at.
   ⛔ THE THREE FLOW PRIMITIVES ARE THE RIGHT CARRIER BECAUSE THEY ARE ONLY EVER A
      CONTAINER. `skeleton.py` assigns `mc-stack` / `mc-row` / `mc-grid` from `.e-con` and
      from nothing else; a widget wrapper is `mc-block` or `mc-media` and a component is
      what it is on the page. ⚠ ONE PROPOSED EXCEPTION EXISTS AND IT IS R3's, NAMED HERE
      RATHER THAN LEFT TO BE DISCOVERED: the generator also puts `mc-grid` on
      `.team-logos-grid`, which is a DIV INSIDE A WIDGET and not a container
      (`mcmc-sitewide.css` gives it its own `display:grid`, `gap:14px` and
      `align-items:center`). The player and home templates must let
      `04-components.css`'s `.mc-team-logos` carry those three declarations and drop
      `mc-grid` from that element — not undo this rule on it.
   ⚠ `.mc-grid` RESTATES `gap` BELOW with the same value from the same source; the later
      one wins and they agree. The duplication is left rather than removed because the
      `.mc-grid` block is a transcription of `.e-con.e-grid > .e-con-inner` (10,030) and
      editing it here would make that citation false. */
.mc-stack, .mc-row, .mc-grid { padding: 10px; gap: var(--mc-gap); }

/* ⛔⛔ AND `width: 100%` IS THE SAME OMISSION ONE PROPERTY ALONG — SEAT REFACTOR-R3, 11 Sept
      2026. `frontend.min.css` gives EVERY `.e-con` `width: var(--width)` and `.e-con` defaults
      `--width` to `100%`, so a child container fills its parent whatever its content is. The
      three primitives had no `width` at all, so a container that is a GRID ITEM shrank to its
      content: measured on `/player/arik-armstead/` at 1440, the ISSUE heading's column
      rendered **47.5 px wide where the served one is 907.19**, because its grid parent carries
      `justify-items: center`.
   ⛔ IT COULD NOT SHOW UP BEFORE THIS UNIT, AND THAT IS THE POINT. R6B, R1 and R2 gated the
      chrome, the story and the archive family — and the `<main>` block of a story and of an
      archive contains NO Elementor container at all, so `mc-stack`/`mc-row`/`mc-grid` appeared
      only in the header and the footer, where every one of them is a flex ITEM that stretches
      and where a missing `width` changes nothing. **The player page is the first page whose
      whole content region is containers.** R1's and R2's sets are re-gated with this change. */
.mc-stack, .mc-row, .mc-grid { width: 100%; }

/* `.e-con.e-grid, .e-con.e-grid > .e-con-inner` (10,030) + `.e-con.e-grid` (3,160):
   `--grid-justify-content:start; --grid-align-content:start; --grid-auto-flow:row`. */
.mc-grid {
  display: grid;
  grid-template-columns: repeat(var(--mc-cols, 1), minmax(0, 1fr));
  grid-auto-flow: row;
  justify-content: start;
  align-content: start;
  gap: var(--mc-gap);
}
/* ⛔ `.mc-grid--2`, `.mc-grid--3` AND `.mc-gap-0` WERE HERE AND ARE GONE. A census of the
   five proposed pages read all three DEAD at all three widths: the column count on this site
   is per-section geometry (the loop grids differ page by page and width by width) and it is
   stated by the page layer where it belongs. `--mc-cols` is kept because the shorthand above
   reads it, and a template that genuinely has a fixed two-column primitive can set it in one
   declaration. FINDINGS 12: nothing inert. */

/* ══ THE WIDGET WRAPPER ═══════════════════════════════════════════════════════════════
   6,529 `.elementor-widget` wrappers exist over the corpus and most of them constrain
   nothing and are unwrapped. The ones that survive are a plain block with three
   declarations, and they are NOT decoration:
     `.elementor-widget`                                   (4,384) : position: relative
     `.e-con .elementor-widget`                            (4,394) : min-width: 0
     `.e-con.e-con > .e-con-inner > .elementor-widget, …`  (4,365) : max-width: 100%
     `.e-con .elementor-widget.elementor-widget`           (4,394) : margin-block-end: 0
   ⛔ `position: relative` IS THE ONE THAT COSTS PIXELS IF IT IS DROPPED, and it does not
     paint. It is the containing block for every absolutely-positioned descendant, and the
     screen-reader spans inside the footer's social icons are exactly that: with the wrapper
     static, all seven resolved to ONE static position instead of seven, 40 px apart.
     Measured by SEAT REFACTOR-R0 on the story page at 390 as a 40/80/120/160/200/240 px
     ladder of box differences that a pixel diff cannot see, because the spans are at
     y = −199,289. A box that never paints can still be wrong, and a defect a pixel gate
     cannot see is a defect the NEXT change makes visible. */
/* ⛔⛔ `max-width: 100%` LEFT THE THREE CONTAINER PRIMITIVES HERE — SEAT REFACTOR-R3, 11 Sept
     2026 — AND THE CITATION ABOVE IS WHY. The four served rules this block transcribes are
     WIDGET rules: `.e-con.e-con > .e-con-inner > .elementor-widget, …` (4,365) is what carries
     `max-width: 100%`, and a `.e-con` has NO max-width at all — measured on 23 containers of
     the player page, where arm A reads `none` and arm B read `100%`. On a container inside a
     flex row whose siblings are wider than their basis, a `max-width: 100%` is not inert: it
     caps the box at the PARENT's content width and the row stops distributing. The widgets
     keep it, because it is theirs. */
/* ⚠ `.mc-spacer` JOINED THIS LIST — SEAT REFACTOR-R3, 11 Sept 2026. `frontend.min.css`'s
   `.elementor-widget{position:relative}` reaches every widget wrapper, and the spacer is one:
   arm A reads `relative` on `ecfe5ac`, arm B read `static`. */
.mc-block, .mc-media, .mc-spacer, .mc-stack, .mc-row, .mc-grid { position: relative; min-width: 0; margin-block-end: 0; }
.mc-block, .mc-media, .mc-spacer { max-width: 100%; }

/* ══ THE INNER, AND ITS POSITION IN THIS FILE IS THE FIX — SEAT REFACTOR-R1, 11 Sept 2026
   ⛔⛔ THIS BLOCK USED TO SIT AT LINE 66, ABOVE THE RESET DIRECTLY ABOVE IT, AND THE RESET
      THEREFORE BEAT ITS `max-width` BY SOURCE ORDER AT EQUAL SPECIFICITY. Both selectors
      are ONE class: `min(100%, var(--mc-shell-max))` lost to a flat `max-width: 100%` on
      every boxed shell whose inner is also a row or a grid — which is every one of them.
      MEASURED by SEAT REFACTOR-R6B: the header's inner rendered **1420 px wide at x=10
      where the served one is 1140 at x=150**, and the footer's the same. REFACTOR-R0's
      arm B could not show it because its page layer PINNED `width: 1140px` on that
      element; R6B restated the `max-width` in `05-chrome.css`, which loads last, and
      that restatement is GONE as of this unit.
   ⭐ MOVING THE BLOCK AND NOT RAISING ITS SPECIFICITY IS THE POINT. The inner is the last
      word on its own box; nothing else in this file changes meaning, because the reset and
      this block overlap in `max-width` and in no other property. It also makes the
      CONTAINER DEFAULT above correct for free: the inner wears `mc-row` or `mc-grid`, so
      it takes that rule's `padding: 10px` and then states its own `padding-inline: 0` and
      `padding-block: 10px` here, which is what the served page has. */
/* from `.e-con > .e-con-inner` (12,672) and `.e-con-full, .e-con > .e-con-inner` (16,974)
   ⚠ `padding-block: 10px` and `height: 100%` were BOTH missing from the first draft. The
     block padding is the `--padding-top/bottom` default the inner consumes, and it is what
     puts the header's logo row at y=10 with height 70 inside a 90 px band. */
.mc-shell__inner {
  width: 100%;
  max-width: min(100%, var(--mc-shell-max));
  height: 100%;
  margin-inline: auto;
  padding-inline: 0;
  padding-block: 10px;
  gap: var(--mc-gap);
  text-align: initial;
  /* ⛔ `position` AND `min-width` ARE STATED HERE BECAUSE THE RESET ABOVE REACHES THIS
     ELEMENT AND ELEMENTOR'S DOES NOT — SEAT REFACTOR-R3, 11 Sept 2026. The inner wears
     `mc-grid` / `mc-stack` / `mc-row`, so `.mc-block, .mc-media, .mc-stack, …` gives it
     `position: relative; min-width: 0`. `frontend.min.css` gives `.e-con-inner` NEITHER —
     grep: every `min-width:0` in that file is on `.e-con` or on `.elementor-grid-item`, and
     the only `.e-con-inner` position rules are for `.elementor-absolute`/`.elementor-fixed`
     children. Measured on four inners of the player page: arm A `static`/`auto`. A static
     ancestor is the R0 screen-reader-span defect pointed the other way, so it is stated
     rather than left to coincide. */
  position: static;
  min-width: auto;
}
/* `.e-con.e-flex > .e-con-inner` (3,280) — the inner of a flex band is itself a flex item
   that grows. */
/* ⛔ THE SELECTOR IS EVERY FLEX CONTAINER'S INNER AND NOT ONLY A BAND'S — SEAT REFACTOR-R3,
   11 Sept 2026. The served rule is `.e-con.e-flex > .e-con-inner` (3,280) and a BOXED flex
   container that is not a top-level band is still `.e-con.e-flex`: the player page has two of
   them (`493163f`, `b5f3b7f`), whose inner read `flex-grow: 0` in arm B against arm A's 1. */
.mc-shell.mc-shell--flex > .mc-shell__inner,
.mc-stack > .mc-shell__inner,
.mc-row > .mc-shell__inner { align-self: auto; flex: 1 1 auto; }
.mc-shell--grid > .mc-shell__inner { display: grid; }

/* ══ MEDIA ════════════════════════════════════════════════════════════════════════════
     `.elementor img`             (27,197) : border none · border-radius 0 · box-shadow none
                                             · height auto · max-width 100%
     `.elementor-widget-image img` (2,152) : display inline-block · vertical-align middle
     `.elementor-widget-image`       (707) : text-align center
     `.elementor-widget-image a`     (620) : display inline-block
   ⛔⛔ THE FIRST DRAFT WROTE `display:block; width:100%` HERE AND BOTH ARE INVENTIONS. No
     rule in the corpus makes an image `display:block` or `width:100%` inside a media box;
     `width:100%` on the site belongs to `.post .wp-post-image` and `.team-logo img`, which
     are components. It cost 2,050 px on the story header logo at 1440 and 143,374 px on the
     footer logo at 390 — one wrong declaration, two thirds of the page's whole diff. */
.mc-media { display: block; text-align: center; }
.mc-media > a { display: inline-block; }
.mc-media img {
  display: inline-block; vertical-align: middle;
  height: auto; max-width: 100%;
  border: none; border-radius: 0; box-shadow: none;
}
/* ⛔ `.mc-media--fit img { object-fit: cover }` WAS HERE AND IS GONE — DEAD on all five
   pages at all three widths. The one place this site crops an image is the parent theme's
   `.post .wp-post-image`, which is `04-components.css`'s `.mc-post .mc-post__img`
   (renamed from `.mc-card__img` by SEAT REFACTOR-R2 — see that file). */

/* ══ THE MAIN COLUMN — the parent theme's, not Elementor's ════════════════════════════
   `body:not([class*=elementor-page-]) .site-main` and its five siblings in `theme.css`.
   ⛔⛔ THIS IS A FOUR-STEP LADDER AND THE 800 px STEP EXISTS ONLY BETWEEN 992 AND 1199 px.
     REFACTOR-MAP measured at 1440 and 390 and its census therefore called the `min-width:
     992px` rule DEAD; SEAT REFACTOR-R0's census at 1024 reads it EFFECTIVE, 58 winning
     pairs. It is one of 114 rules and 5,604 winning pairs that live only in the tablet band
     — 9.5% of the effective set, invisible to a two-width gate. This is what the owner's
     third width bought. */
/* ⭐⭐ AND THE LADDER IS SERVED TO TWO ELEMENTS ON A STORY, NOT ONE — SEAT REFACTOR-R1,
   11 September 2026. The rule's selector list is six selectors long and `.site-main` is
   only one of them; `.page-header .entry-title` is another, so the parent theme applies
   the SAME four-step ladder to the page's `<h1>` INSIDE the main column. Measured on the
   served story at 1440: `main` is 1140 wide at x=150 and the `<h1>` is 1140 wide with
   `max-width: 1140px` of its own, won by that rule.
   ✅ AND R2 TOOK THE CONDITION NAMED HERE RATHER THAN DISCOVERING IT — 11 September 2026.
      The child combinator is only right for a template whose title is a DIRECT child of the
      main column. `single-stories.php` drops the `.page-header` wrapper (it carries no
      declaration of its own); `taxonomy.php`, `category.php` and `404.php` drop it too, for
      the same measurement taken again on their own page type — `wins.mjs` at 1440 reads the
      archive's `.page-header` winning NOTHING but the universal `box-sizing`, and R0's
      `skeleton.py` unwrapped it mechanically on the same page (`unwrapped: 1`). **So this
      selector did NOT have to grow a second form, and the six lines below reach all four
      page types unchanged.**
   ⚠ THE NEXT UNIT TO MEET IT IS THE ONE WHOSE TITLE IS NOT A DIRECT CHILD. Nothing about
      the ladder has been generalised; it has been kept true by keeping the markup flat.
   ⛔ IT IS SCOPED `.mc-main > .mc-entry__title` AND NOT BARE, because the story CARDS on
      home, the directory and the player pages wear `mc-entry__title` too
      (`04-components.css`'s `.mc-post .mc-entry__title a`, which R2 renamed from
      `.mc-card--story …`) and the served rule does not reach them: it needs an
      `.entry-title` inside a `.page-header`, and a card title is in neither. The child combinator is the same scope one element cheaper, which is
      why `single-stories.php` does not carry the `.page-header` wrapper.
   ⛔⛔ AND THE BREAKPOINT FOR THE PADDING IS 575, WHICH IS THE MEASURED LAYER'S OWN
      COUNTER-EXAMPLE. REFACTOR-R0's machine-produced story layer carries
      `padding-left/right: 10px` on the title inside `@media (max-width: 767px)`, because
      it was measured at 390 and emitted at the harness's own breakpoint. The SERVED rule
      is `@media (max-width: 575px)`. Both are right at 1440, 1024 and 390 — the three
      widths the gate shoots — and they disagree at every width from 576 to 767, where the
      pinned layer indents the title by 10 px that the site does not. R6B's FINDINGS 1,
      one rule at a time: keep the RULE, not the number. */
.mc-main, .mc-main > .mc-entry__title { width: 100%; margin-inline: auto; }
@media (min-width:  576px) { .mc-main, .mc-main > .mc-entry__title { max-width:  500px; } }
@media (min-width:  768px) { .mc-main, .mc-main > .mc-entry__title { max-width:  600px; } }
@media (min-width:  992px) { .mc-main, .mc-main > .mc-entry__title { max-width:  800px; } }
@media (min-width: 1200px) { .mc-main, .mc-main > .mc-entry__title { max-width: 1140px; } }
@media (max-width:  575px) { .mc-main, .mc-main > .mc-entry__title { padding-inline: 10px; } }

/* ══ VISIBILITY UTILITIES ═════════════════════════════════════════════════════════════
   `.elementor .elementor-hidden-desktop`  (267) @ (min-width:1025px)
   `.elementor .elementor-hidden-tablet`   (397) @ (min-width:768px) and (max-width:1024px)
   `.elementor .elementor-hidden-mobile`   (346) @ (max-width:767px)
   ⛔⛔ THE TABLET ONE IS ALIVE AT EXACTLY ONE OF THE THREE GATE WIDTHS AND REFACTOR-MAP
     COULD NOT SEE IT AT EITHER OF ITS TWO. It is one of only four rules in the whole 2,606
     that are alive at 1024 and at neither 1440 nor 390.
   ⚠ THE SITE IS BUILT TWICE IN PLACES AND THESE ARE HOW. 267 / 397 / 346 elements carry one
     of the three; 121 carry ALL THREE and are `display:none` at every breakpoint the kit
     defines — those 121 are deletions, not renames. The rest are a real authoring pattern:
     the header, three home sections and the footer copy are authored once wide and once
     narrow, and IN THE FOOTER'S CASE THE TWO COPIES ARE NOT THE SAME TEXT (the narrow one
     is split into two paragraphs). Collapsing a duplicated pair is a per-case decision for
     a build seat with the pixel gate as its check, never a sweep. */
@media (min-width: 1025px) { .mc-hide-desk { display: none !important; } }
@media (min-width: 768px) and (max-width: 1024px) { .mc-hide-tab { display: none !important; } }
@media (max-width: 767px) { .mc-hide-mob { display: none !important; } }

/* the accessible-hidden helper — `.elementor-screen-only, .screen-reader-text,
   .screen-reader-text span, .ui-helper-hidden-accessible` in frontend.min.css, 41,570
   winning pairs on 12 elements per page.
   ⛔ `top: -10000em` IS PART OF THE RULE and the first draft dropped it while adding a
     `white-space: nowrap` that no rule in the corpus contains. See `02-base.css` `.mc-skip`
     for the OTHER accessible-hiding rule, which is theme.css's and is not this one.
   ⚠ BANDED, 13 SEPT 2026, SEAT REFACTOR-R9: `.mc-skip` is deleted (no element carries it); theme.css's rule is
     still the one the served skip link gets. */
.mc-sr {
  border: 0; clip: rect(0, 0, 0, 0); height: 1px; margin: -1px; overflow: hidden;
  padding: 0; position: absolute; top: -10000em; width: 1px;
}

/* ══ THE SPACER ═══════════════════════════════════════════════════════════════════════
   `.e-con-inner > .elementor-widget-spacer, .e-con > .elementor-widget-spacer` (540) and
   the two `.elementor-spacer-inner` rules (180 each). 90 spacer widgets in the corpus. In
   the finished system a spacer is padding or a gap on the element that needs it and the 90
   elements go; this class exists so a build seat can land a template before doing that
   arithmetic, and every use of it is a debt with a name. */
.mc-spacer { display: block; width: var(--mc-spacer-w, 100%); height: var(--mc-spacer, 20px); flex-shrink: 0; }
.mc-spacer__inner { height: 100%; }

/* ══ THE LAZY BACKGROUND ══════════════════════════════════════════════════════════════
   ⛔ REPRODUCED, NOT REPLACED, AND IT IS NOT ELEMENTOR PRO'S. This is Elementor FREE's
   lazyload module: an inline `<style>` of 722 B on 79 of 79 pages plus a 1.5 KB inline
   IntersectionObserver that adds `e-lazyloaded`. Its one rule is EFFECTIVE — 1,300 winning
   pairs on 618 elements. It needs no library and reads no Elementor API, so the refactor
   keeps the mechanism and renames the two classes with the script that sets them.
   ⚠ `nth-of-type` COUNTS SIBLINGS, so the class it is written against has to stay on the
   same element in the same position. That is the one place in this file where a markup
   change is not free, and it is invisible before scroll. */
.mc-band:not(.mc-lazyloaded):not(.mc-no-lazyload):nth-of-type(n+4),
.mc-band:not(.mc-lazyloaded):not(.mc-no-lazyload):nth-of-type(n+4) * { background-image: none !important; }
@media screen and (max-height: 1024px) {
  .mc-band:not(.mc-lazyloaded):not(.mc-no-lazyload):nth-of-type(n+3),
  .mc-band:not(.mc-lazyloaded):not(.mc-no-lazyload):nth-of-type(n+3) * { background-image: none !important; }
}
@media screen and (max-height: 640px) {
  .mc-band:not(.mc-lazyloaded):not(.mc-no-lazyload):nth-of-type(n+2),
  .mc-band:not(.mc-lazyloaded):not(.mc-no-lazyload):nth-of-type(n+2) * { background-image: none !important; }
}
