Redesign the RSIHub page as a product landing page (#11)

* Refresh the RSIHub project page against the latest repository

- Replace the benchmark section with the v2 figure and the combined
  Train / Full Benchmark table, matching the current README numbers.
- Add the four-part composition model (stage, operator, recipe,
  evaluate) and the operator CLI workflow to How RSIHub works.
- Link the showcase paragraph to the result snapshot, frozen rubric,
  and minimal seed Skill.
- Sync the subtitle and descriptions to the new framing (evaluator-
  driven learning, controllable modification).
- Add the RSIHub ring mark above the page title and page styles for
  the mark and multi-line code blocks.
- Bump the sitemap lastmod and update the expected asset list in the
  migration tests.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Rebuild the RSIHub page header with the lockup and a rotating tagline

- Replace the ring mark + text title with the inline repo lockup so the
  wordmark follows the site's manual theme toggle (the standalone SVG
  only responds to prefers-color-scheme).
- Add an animated hero line — evolve your <target> on <benchmark> —
  whose slots rotate through Codex, Claude Code, a DeepSeek harness,
  and MiniSWE against Terminal-Bench 2, Tau³ Banking, and your own
  eval; plain text swaps under prefers-reduced-motion.
- Remove the lineage figure and the now-unused evolve-lineage and
  ring-mark assets; sync the migration-test asset list.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Redesign the RSIHub page as a product landing page

Replace the README-style article with a proper product homepage:

- Hero: repo lockup (ring + wordmark, theme-aware inline SVG), a large
  two-line rotating headline (evolve your <target> / on <benchmark>)
  with per-language word lists, sub-copy, and CTA buttons.
- A terminal card walking through the real quick-start commands with
  published seed/best train scores, plus an honesty note.
- Stats strip (best full-benchmark gains, recipe count, license) with
  ramp-gradient numerals.
- Sections rebuilt as landing blocks: loop pills + architecture,
  composition-model cards + operator CLI, audience cards, mutable-
  surface cards, recipe rows, showcase, benchmark chart with the full
  table folded into a details element, trust grid, and a closing CTA
  band with doc links.
- All styling is page-scoped; the shared stylesheet is restored to its
  upstream state. Bilingual parity and the manual theme toggle are
  preserved throughout; reduced-motion swaps text without animation.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Make the hero rotors layout-stable and drop their chip backgrounds

Each rotor now stacks hidden sizer spans for every item it can show, so
the slot always reserves the width of its longest word and the
surrounding text never moves during a swap — the word simply fades in
place. The chip background is removed (color alone marks the slots) and
the headline loses its trailing period, which otherwise floated apart
from short words in the reserved space.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Left-align the hero so rotor spacing gaps disappear

With the headline, sub-copy, CTAs, and terminal card left-aligned and
rotor words left-justified in their reserved slots, the reserved width
for longer items trails invisibly at the line end instead of opening
gaps mid-line. The Chinese headline now breaks after the benchmark slot
(在 <bench> / 上进化你的 <target>) so both slots sit at line ends.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Rework the first screen into a full-height split hero

- The hero now fills the first viewport: lockup and headline on top,
  then a two-column row with copy + CTAs on the left and the terminal
  card on the right, vertically centered as a unit.
- Subtle brand-color radial washes behind the hero (tuned separately
  for dark), staggered rise-in entrance for lockup, headline, copy, and
  card, and the terminal's generation rows fade in sequentially; all
  animation is disabled under prefers-reduced-motion.
- Terminal commands recomposed to fit the narrower card without
  scrolling; the caption stays under the card.
- Responsive: the row stacks under 960px with minmax(0, 1fr) so the
  card cannot overflow, and the headline shrinks further under 640px so
  the rotors' reserved widths fit narrow viewports.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Fix the blurry hero lockup on high-DPI screens

The infinite ring rotation promoted the whole lockup SVG to an
animated compositor layer, which Chrome rasterized at 1x — leaving the
wordmark soft on Retina displays. The ring and wordmark are now
separate SVGs so only the ring joins the animated layer, and the hero
entrance animations use fill-mode backwards instead of both so no
element stays composited (and cached blurry) after the animation ends.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Condense the page flow to five sections with benchmarks up front

New flow: hero → benchmark results → how it works (loop + architecture
merged with the five recipe rows under 'One loop. Five strategies.') →
showcase → CTA. The stats strip, composition-model cards + CLI block,
audience cards, and mutable-surface / trust sections are removed — the
landing now keeps only what answers a visitor's next question, and the
strongest proof asset sits right after the promise. Sections gained
anchor ids (results, how-it-works, showcase, get-started) and the CSS
for removed blocks is cleaned up.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Paged scrolling, an animated benchmark chart, and an arXiv placeholder

- The page now scrolls screen by screen: sections snap (y proximity)
  and are sized to the viewport, and each section's children rise in
  with a stagger the first time it enters view (IntersectionObserver;
  everything appears instantly under prefers-reduced-motion or no JS).
- The static benchmark SVG is replaced by an HTML/CSS chart in the
  original blue/orange palette (MiniSWE blue, Codex orange, green
  deltas, truncated axes, rotated method labels) carrying the current
  v2 full-benchmark numbers. On entry the dark seed sections grow
  first, the light improvement sections follow, and the best (change)
  labels fade in last.
- The hero CTA row gains a disabled arXiv button with a Coming soon
  badge as a placeholder for the paper link.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Rebuild the architecture figure as an animated HTML diagram

The static architecture.svg (and the redundant loop-pill row above it)
is replaced by an in-page diagram in the benchmark chart's visual
language: blue method pills plus a dashed your-recipe slot, a
blue-tinted composable-loop box whose orange dashed border marks the
declared mutable surface, six stage cards (Mutate and Gate highlighted
orange) with a next-generation return loop, evidence/contract
connectors, and dark slate protected-mechanism cards. On section entry
the pills, connector, stages, surface, and locks appear in sequence; a
soft highlight ring then cycles through the six stages continuously.
All motion is disabled under prefers-reduced-motion.

Also: the CTA column now centers its items, fixing the git-clone chip
stretching to full width, and the two now-unused SVG assets are
removed with the migration-test asset list updated.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Link recipe rows to their original method papers

The right-hand mutable-surface labels read as floating fragments, so
each recipe row now ends with a reference link instead: ahe, gepa, and
hyperagents point to their arXiv papers, aevolve to the A-Evolve
repository, and hill_climb (a classic technique with no single source)
carries none. The label column is gone along with its styles.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Stop recipe code chips stretching to the grid column width

Grid items stretch by default, so short recipe names dragged a wide
empty pill background behind them; justify-self: start lets each chip
hug its text while descriptions stay aligned on the fixed column.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Label hill_climb as a classic loop in the recipe list

The row's empty reference slot now carries a muted non-link tag
(Classic loop / 经典循环) so the column reads consistently.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Recolor the architecture diagram with the lab's brand ramp

The diagram now uses Simple Agent Lab's palette instead of the legacy
chart colors: method pills in brand blue (#3c8cff), the declared
mutable surface and the Mutate/Gate stages in ramp green (#2fa844
light / #78e85c dark), the cycling stage highlight as a teal #00cbd4
ring with a soft glow, and the protected-mechanism cards in ink
(var(--fg) on var(--bg)), matching the primary button and inverting
cleanly in dark mode. The blue/orange palette remains exclusive to the
benchmark chart.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Lighten the diagram connector and protected-mechanism cards

The bare 1px vertical line between the methods row and the loop box
read as a rendering artifact — it is now a small faint down arrow. The
protected-mechanism cards were the only large filled blocks in an
otherwise outlined diagram, so they switch from ink fill to a 1.5px
ink border, keeping the sealed-vault contrast without the weight.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Render the showcase posters at equal heights within one screen

The 2fr:1fr grid split makes the 1.6:1 and 0.8:1 posters come out at
identical heights from their natural aspect ratios — no cropping — and
capping the grid at 840px keeps the whole section inside a single
viewport. Below 700px the posters stack.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Drop 你的 from the Chinese hero headline

在 <bench> 上进化你的 <target> becomes the tighter 在 <bench> 上进化
<target>; the English line keeps its idiomatic your.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Rework the how-it-works copy in plainer terms

Hand-edited Chinese copy toward a plainer voice: the hero sub now
positions RSIHub as an environment for the RSI process with known
evolution methods and benchmarks integrated; the loop section reads as
a typical iteration loop with lighter stage and mechanism wording. The
English spans whose message changed (hero sub, section title and lead,
surface label, Mutate and Gate stages) are retranslated to match;
spans where the Chinese was only simplified keep their English.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Plain-voice pass over the remaining Chinese copy

Continues the hand-edited tone across the page: benchmark section
(kicker, title, lead, table headers), terminal caption, architecture
notes, the recipe bridge line and all five descriptions, the showcase
lead, captions, and closing line, and the CTA lead. Translationese is
dropped in favor of everyday phrasing, and common English terms (eval,
benchmark, prompt, skill, agent) stay in English, matching the voice
of the manual edits. English spans are unchanged — the message is the
same, only the Chinese register moved.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* Keep candidate in English across the Chinese copy

候选/候选方案 becomes candidate everywhere (Mutate stage, hill_climb
description, the loop lead, and the outside-candidate-control note),
matching the voice that keeps technical terms untranslated. The hero
sub's English is synced to the hand-edited Chinese (multiple RSI
methods rather than well-known evolution methods).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

Author: Octane0411