diff --git a/.convoys/unify-glass-panel-surfaces.md b/.convoys/unify-glass-panel-surfaces.md
index b7a8675..8edeb8b 100644
--- a/.convoys/unify-glass-panel-surfaces.md
+++ b/.convoys/unify-glass-panel-surfaces.md
@@ -361,3 +361,284 @@ build on in-flight migrations.
 - **`storybook-adoption-for-glass-surface`** — once the primitive
   has the full prop matrix (`tint`, `blur`, `rim`, `elevation`,
   `cornerLights`), it's a natural Storybook candidate. P2 DX.
+
+## Architecture
+
+Run date: 2026-06-04. Architect: role-architect (this convoy).
+Reads: this convoy file, `AGENTS.md`, `.cursor/rules/*.mdc`,
+`components/ui/GlassSurface.js`, `components/ui/Modal.js`,
+`components/ui/StatCard.js`, `components/Layout.js`,
+`components/ui/TopSearchBar.js`, `components/BulkSelectionToolbar.js`,
+`pages/login.js`, `pages/signup.js`, `pages/index.js`,
+`pages/profile.js`, `pages/settings.js`,
+`pages/community/collections.js`,
+`components/CollectionsPageView.js`, `styles/globals.css`
+(`.card`, `.glass-panel`, `.glass-panel-strong`, `.page-header-glass`,
+the `--corner-light-*-subtle` tokens added in PR #118, the
+`--bg-secondary-rgb` token used by today's auth cards),
+`.github/workflows/ci.yml`, `test/components/ui-primitives.test.js`,
+`test/components/Layout.test.js`.
+
+### Decisions ratified
+
+The convoy file's four open Decisions are resolved as follows.
+Implementer briefs cite these and MUST NOT renegotiate them mid-flight
+(if a brief discovers a reason to revisit, escalate via the mid-convoy
+scope-expansion process in `role-architect.md`).
+
+**D1. `<GlassSurface>` `cornerLights` prop shape.** Add
+`cornerLights = 'subtle' | 'chrome' | 'none'`, **default `'subtle'`**.
+- `'subtle'` → applies the 4-layer gradient with
+  `--corner-light-warm-subtle` / `--corner-light-cool-subtle` (matches
+  what `.glass-panel-strong` ships today; PR #118).
+- `'chrome'` → applies the 4-layer gradient with full-intensity
+  `--corner-light-warm` / `--corner-light-cool`. Used by the Layout
+  sidebar nav-chip and TopSearchBar header — those keep their inline
+  styles for now; this prop value exists so future floating chrome
+  doesn't have to re-handroll the gradient.
+- `'none'` → no transparent border, no radials, no `--chip-border-base`
+  layer. Identical to today's primitive output. Used for the CardItem
+  grid path and any other GPU-budget-constrained tile that legitimately
+  must skip the gradient-border treatment.
+
+The gradient-border technique is the verbatim 4-layer recipe from
+`.glass-panel-strong` (`styles/globals.css` post-PR-#118): a
+padding-box solid linear-gradient of the fill, two border-box radial
+gradients for the corner catch-lights (warm 0%↔100%, cool 100%↔0%),
+and a border-box `var(--chip-border-base)` base layer. Combined with
+`border: 1px solid transparent` so the gradient renders through the
+border.
+
+**D2. `.card` class retire-or-keep.** **Retire the class entirely.**
+All 8 consumers migrate to `glass-panel rounded-3xl p-{6|4}`. The
+`.card` block in `styles/globals.css` is deleted. Rationale: a single
+glass vocabulary across the app is the convoy's success metric;
+keeping `.card` as a documented opaque fallback creates two parallel
+panel languages for future agents to choose between, which is exactly
+the kind of design-system bifurcation the convoy was scoped to
+eliminate. The GPU-budget concern that motivated keeping `.card` is
+specific to `<CardItem>` grid-mode (a non-`.card` consumer) — none
+of the 8 `.card` sites face that constraint (they're profile cards,
+settings panels, and a single community-page list tile, not a
+multiplied-per-thumbnail surface).
+
+Aggressive path; **risk** = one of the 8 sites has a hidden reason
+to be opaque that the architect didn't catch. Mitigation: Brief 5
+includes a "if any consumer breaks visually post-migration, revert
+that one consumer to inline `style={{ background:
+'var(--bg-secondary)' }}` and document the exception in the brief's
+post-merge note" escape hatch. That keeps the class deletion in the
+PR while leaving an explicit per-site fallback. If that escape hatch
+fires for more than 1 of the 8 consumers, the brief's reviewer is
+expected to push back and propose holding `.card` after all.
+
+**D3. Mobile-drawer corner-light intensity.** Use
+`.glass-panel-strong` (subtle tier). The drawer is a ~256px-wide
+fixed-position surface that covers ~1/3 of a mobile viewport. At
+full-intensity (`'chrome'`) the corner radials would dominate the
+drawer's interior nav text; at subtle they read as "elevated panel"
+without competing with content. The drawer is also a hidden surface
+most of the time — full intensity buys nothing for the rare moments
+it's open.
+
+**D4. CI gate scope.** Gate **inline-style usage only** —
+`var(--glass-surface-(low|mid|high))` appearing inside a JSX
+`style={{ background: ... }}` (or `backgroundColor:`, `background:`
+in template-string form) under `components/**` or `pages/**`. CSS
+class definitions in `styles/globals.css` (where the tokens are
+LEGITIMATELY chained to compose `.glass-panel` / `.glass-panel-strong`
+/ `.page-header-glass`) are not gated. Allowlist entries: the
+Layout sidebar nav-chip block (`components/Layout.js` L858-861) and
+the TopSearchBar header block (the equivalent inline-style block in
+`components/ui/TopSearchBar.js`) — both intentionally retain
+`'chrome'`-tier handrolled gradients and were ratified as such in
+PR #116. The `<GlassSurface>` primitive itself is also allowlisted
+(it sets the background internally; that's its job). Brief 7 ships
+this gate as a separate `forbidden-bespoke-glass-surface` job in
+`ci.yml`, modeled on the existing `forbidden-modal-shell-without-primitive`
+gate's grep-and-allowlist shape.
+
+### File plan
+
+| File | Action | Purpose | Brief |
+| --- | --- | --- | --- |
+| `components/ui/GlassSurface.js` | modified | Add `cornerLights` prop (`'subtle' \| 'chrome' \| 'none'`, default `'subtle'`); compose 4-layer gradient + `border: 1px solid transparent` when not `'none'` | 1 |
+| `test/components/ui-primitives.test.js` | modified | Add 3 corner-light assertions covering the 3 `cornerLights` values | 1 |
+| `pages/login.js` | modified | Replace L82-88 inline glass imitation with `<div className="glass-panel-strong rounded-2xl p-8">` | 2 |
+| `pages/signup.js` | modified | Replace L227-232 inline glass imitation with `<div className="glass-panel-strong rounded-2xl p-8">` | 2 |
+| `components/Layout.js` | modified | Sidebar profile dropdown (L88-96): drop inline background+blur+shadow, add `className="glass-panel-strong rounded-xl"`. Mobile drawer (L698-709): drop inline background+blur+shadow, add `className="glass-panel-strong"` and keep the slide-in transform classes. Sidebar nav-chip block (L858-861) untouched (chrome tier). | 3 |
+| `components/ui/TopSearchBar.js` | modified | UserMenu dropdown (L214-222): drop inline background+blur+shadow, add `className="glass-panel-strong rounded-xl"`. Header block (chrome tier) untouched. | 3 |
+| `test/components/Layout.test.js` | modified | Add regression-lock assertion: sidebar profile dropdown and mobile drawer carry `.glass-panel-strong` class. | 3 |
+| `components/BulkSelectionToolbar.js` | modified | L47 toolbar: replace `bg-white border-gray-200` with `glass-panel-strong rounded-2xl`. L117 dropdown: same swap on `rounded-lg` → `rounded-xl glass-panel-strong`. Sweep interior `text-gray-{600,700}` to `var(--text-primary/secondary)`; `hover:bg-gray-{50,100}` to `nav-item-hover`; `text-red-600 hover:bg-red-50` to `var(--accent-danger)` + ember-tinted hover. | 4 |
+| `pages/profile.js` | modified | 3 `.card` → `glass-panel rounded-3xl p-6` swaps. | 5 |
+| `pages/settings.js` | modified | 2 `.card` → `glass-panel rounded-3xl` swaps (one `p-4`, one `p-6`). | 5 |
+| `pages/community/collections.js` | modified | 1 `.card` → `glass-panel rounded-3xl` swap (preserve `group cursor-pointer hover:shadow-lg transition-all`). | 5 |
+| `components/CollectionsPageView.js` | modified | 1 `.card` → `glass-panel rounded-3xl` swap (preserve `hover:shadow-xl transition-all cursor-pointer group`). | 5 |
+| `styles/globals.css` | modified | **Delete** the `.card { ... }` rule (the class block that lives near L757). No other CSS changes. | 5 |
+| `pages/index.js` | modified | Landing nav bar (L61-72): replace handrolled inline style with `<nav className="page-header-glass border-b">`. Drop the malformed `boxShadow: 'inset 0 1px 0 var(--rim-light-inner)'` (the `.page-header-glass` class already sets the correct inset rim). | 6 |
+| `.github/workflows/ci.yml` | modified | Add `forbidden-bespoke-glass-surface` job per D4. | 7 |
+
+### API surface
+
+N/A. This convoy is pure UI/CSS surface composition. No new routes,
+no Zod schema additions, no rate-limit changes.
+
+### Schema diff
+
+N/A. No DB changes. No `migrations/` files added; no
+`scripts/add-*.js` follow-up. The schema-map-fresh CI job will not
+fire on any of these PRs (the conditional in `ci.yml` L47-52 checks
+`migrations/` / `scripts/add-*` / `scripts/fix-*` /
+`scripts/setup-neon-db.js` / `docs/SCHEMA_MAP.md` — none touched).
+
+### Test plan
+
+| Brief | Unit | Component | Smoke | Visual diff | Notes |
+| --- | --- | --- | --- | --- | --- |
+| 1 | — | **3 new assertions** in `test/components/ui-primitives.test.js` covering `cornerLights='subtle'` (default, expects `border: 1px solid transparent` in the rendered style attribute), `cornerLights='chrome'` (expects `--corner-light-warm` token reference, not `-subtle`), `cornerLights='none'` (expects no `border` declaration, no gradient layers — same shape as today). | Existing smoke tests cover Modal open/close; nothing new. | **Baseline refresh required** after Brief 1 lands — Modal/StatCard/landing feature cards all gain corner lights. Plan: queue the refresh as a single Linux-baselined commit immediately after Brief 1 merges, before Briefs 3/4 ship. | The most consequential brief; the assertions are the regression lock. |
+| 2 | — | None needed (presentational class swap). | Existing `tests/smoke/auth.spec.js` covers login/signup happy path; nothing new. | Auth pages diff expected. | — |
+| 3 | — | Regression-lock in `test/components/Layout.test.js`: render Layout with `isMobileMenuOpen={true}` (mocked) and assert the drawer container's `className` includes `glass-panel-strong`. Render with `<UserProfileDropdown>` open (mocked) and assert its dropdown container's `className` includes `glass-panel-strong`. | Existing smoke covers nav clicks. | Layout dropdown/drawer visual diff expected. | — |
+| 4 | — | None needed (rare to bulk-select in a smoke test). | — | Expected visual diff on the bulk-select toolbar — but only fires when one of the visual-diff `tests/visual/*` specs actually triggers bulk-select. Probably won't surface in the diff at all; rely on manual QA + dark-mode side-by-side screenshot for review. | The dark-mode invisibility is the regression we're closing; verify manually post-merge. |
+| 5 | — | None needed (className swap). | — | 4 pages worth of visual diff expected. | — |
+| 6 | — | None needed. | Existing smoke `tests/smoke/landing.spec.js` covers landing nav rendering. | Expected on landing. | — |
+| 7 | **Negative test** documented in the brief's acceptance criteria: an implementer dry-run that adds a scratch `style={{ background: 'var(--glass-surface-low)' }}` to a `components/` file should fail the new `forbidden-bespoke-glass-surface` job. Verified once by the architect via the boot-the-brief check below; not committed. | — | — | — | The grep gate runs in CI; no test code needed. |
+
+Reference test files for the brief authors:
+- `test/components/ui-primitives.test.js` (Button/Input/SearchBar test
+  patterns, jsdom env, cleanup pattern; Brief 1 extends in-file).
+- `test/components/Layout.test.js` (5 regression-lock assertions
+  established by `fix-layout-default-user` PR #15; Brief 3 extends).
+
+### Risk list
+
+1. **Brief 1 ripple effect** (**MED**). The `<GlassSurface>` primitive
+   feeds `<Modal>`, `<StatCard>`, and 4-5 landing-page feature cards.
+   Adding `border: 1px solid transparent` to every consumer changes
+   the rendered geometry by 2px in both axes — most consumers don't
+   notice (their `padding`/`gap` absorbs it), but any consumer that
+   relied on a 0-border box-model for pixel-exact layout will shift.
+   The `<Modal>` panel sets `maxWidth` in pixels and is centered via
+   flex — it's not sensitive. `<StatCard>` is in a 4-up grid — also
+   not sensitive. The landing feature cards use `grid` with `gap`
+   — not sensitive. But the visual-diff baseline refresh is
+   non-negotiable; queue it before Briefs 3 + 4 ship.
+2. **Brief 1 + GPU-budget constraint** (**LOW-MED**). AGENTS.md
+   forbids `backdrop-filter` per `CardItem` thumbnail. `<CardItem>`
+   does NOT use `<GlassSurface>` today, so adding gradient-border to
+   `<GlassSurface>` doesn't pull it in. The `'none'` cornerLights
+   value exists as a defensive escape hatch for any future tile
+   surface that DOES want `<GlassSurface>` without the gradient
+   layers. Brief 1's brief explicitly forbids changing CardItem
+   anywhere in this convoy.
+3. **Brief 3 popovers' z-index + box-shadow chain** (**LOW-MED**).
+   `.glass-panel-strong` ships with
+   `var(--rim-light-inner), var(--elevation-ambient)`. Today's
+   inline-style on the profile dropdown adds
+   `var(--ember-rim-subtle)` as a third shadow for ember emphasis;
+   today's mobile drawer uses `var(--elevation-pronounced)` (heavier
+   elevation than ambient). The brief MUST preserve these via an
+   inline-style override chained onto the class's box-shadow — DON'T
+   just drop them. Verbatim shape documented in Brief 3.
+4. **Brief 4 dark-mode regressions** (**LOW**). The bulk-select
+   toolbar is currently `bg-white` — invisible in dark mode. The
+   migration FIXES that regression but Brief 4's reviewer should
+   capture before/after dark-mode screenshots so the fix is
+   visible in PR review.
+5. **Brief 5 `.card` deletion risk** (**LOW-MED, conditional**).
+   Per D2, all 8 consumers migrate to `.glass-panel`. If any one of
+   them breaks visually (the architect doesn't expect this; the
+   8 sites are all "panel with content inside" surfaces), the
+   brief includes an escape hatch to revert that one site to
+   inline `style={{ background: 'var(--bg-secondary)' }}` rather
+   than holding the whole class deletion. If MORE than 1 of 8
+   needs the escape hatch, the brief's reviewer is expected to
+   push back and propose keeping `.card`.
+6. **Brief 6 `.page-header-glass` already-applied check** (**LOW**).
+   The implementer must verify `pages/index.js`'s nav bar isn't
+   already inside another `.page-header-glass` ancestor (it isn't —
+   landing has no shared Layout); otherwise nested
+   `backdrop-filter`s would compound.
+7. **Brief 7 false-positive risk** (**LOW**). The grep gate must
+   correctly distinguish JSX inline-style usage of
+   `var(--glass-surface-*)` from CSS class definitions in
+   `styles/globals.css`. The architect-prescribed approach is to
+   restrict the grep to files under `components/**` and `pages/**`
+   (excluding `.css`), which by construction means JSX inline-style
+   only. Brief 7 includes a verbatim grep command shape and the
+   2-entry allowlist.
+8. **PR #117 + PR #118 not yet merged** (**LOW, dependency**). This
+   convoy's `depends_on:` frontmatter lists both. If either fails
+   review, the briefs may need to refactor against a different
+   baseline. Recommended: hold Brief 1 dispatch until PR #117 + PR
+   #118 are both squash-merged to main.
+
+### Decomposition
+
+| Brief # | Title | Files | Depends on | LOC est. |
+| --- | --- | --- | --- | --- |
+| 1 | Upgrade `<GlassSurface>` primitive with `cornerLights` prop | `components/ui/GlassSurface.js`, `test/components/ui-primitives.test.js` | — | ~80 |
+| 2 | Migrate auth form cards to `.glass-panel-strong` | `pages/login.js`, `pages/signup.js` | — | ~20 |
+| 3 | Migrate floating popovers to `.glass-panel-strong` | `components/Layout.js`, `components/ui/TopSearchBar.js`, `test/components/Layout.test.js` | 1 | ~90 |
+| 4 | Migrate `BulkSelectionToolbar` to `.glass-panel-strong` + token sweep | `components/BulkSelectionToolbar.js` | 1 | ~70 |
+| 5 | Retire `.card` class; migrate 8 consumers to `.glass-panel` | `pages/profile.js`, `pages/settings.js`, `pages/community/collections.js`, `components/CollectionsPageView.js`, `styles/globals.css` | — | ~30 |
+| 6 | Migrate landing nav bar to `.page-header-glass` | `pages/index.js` | — | ~10 |
+| 7 | `forbidden-bespoke-glass-surface` CI gate | `.github/workflows/ci.yml` | 1, 2, 3, 4, 5, 6 | ~40 |
+
+All briefs target <400 LOC; the largest (Brief 3) is ~90 LOC. Files
+across briefs are disjoint — verified against the slice_dependencies
+block above. No two parallel writers on the same file.
+
+Parallel dispatch recommended for Briefs 1, 2, 5, 6 (no deps, disjoint
+files). Briefs 3 + 4 wait for Brief 1 to land (per D1 — they may
+simplify by adopting the upgraded primitive's `cornerLights='subtle'`
+default rather than handrolling). Brief 7 runs LAST so the grep gate
+doesn't fail the build on in-flight migrations.
+
+### Boot-the-brief check
+
+Read-only verification against the post-PR-#117 + post-PR-#118 tree.
+
+1. **Dep-set check.** No new npm dependencies added by any brief. The
+   gradient-border recipe is pure CSS; the `cornerLights` prop is
+   pure React composition; the CI gate is a shell `grep` invocation
+   in a YAML job. No `pnpm view` / `package.json` audit needed.
+2. **Verbatim code-shape check.**
+   - Brief 1's 4-layer recipe matches `.glass-panel-strong`'s current
+     shape verbatim — verified against `styles/globals.css` post-PR
+     #118 (the `background:` block uses
+     `linear-gradient(var(--glass-surface-high), ...) padding-box,
+     radial-gradient(at 0% 100%, var(--corner-light-warm-subtle) ...) border-box,
+     radial-gradient(at 100% 0%, var(--corner-light-cool-subtle) ...) border-box,
+     var(--chip-border-base) border-box`). The primitive's
+     `'chrome'` value uses the non-`-subtle` tokens; both pairs exist
+     in `:root` + `[data-theme="dark"]` (verified).
+   - Brief 3's box-shadow preservation: the sidebar profile dropdown
+     ships `var(--rim-light-inner), var(--ember-rim-subtle),
+     var(--elevation-ambient)`; the mobile drawer ships
+     `var(--rim-light-inner), var(--rim-light-outer),
+     var(--elevation-pronounced)`. Both are preserved as inline
+     `style={{ boxShadow: '...' }}` overrides on the
+     `.glass-panel-strong` element (the class's default
+     `var(--rim-light-inner), var(--elevation-ambient)` would otherwise
+     lose the ember-rim and the pronounced elevation). Brief 3
+     documents the verbatim chain.
+   - Brief 5's class deletion: the `.card` block in
+     `styles/globals.css` is a single rule; deleting it is a clean
+     diff (verified by reading L750-L770 area — no other `.card`
+     compound selectors exist).
+   - Brief 7's grep command shape: modeled on the existing
+     `forbidden-modal-shell-without-primitive` job (`ci.yml`
+     L200-L226), which uses
+     `grep -lE 'pattern' pages components -r --include='*.js'`.
+     Same shape works here. Allowlist is implemented via
+     `! -path 'components/ui/GlassSurface.js'` and an explicit
+     line-range exclusion for `components/Layout.js` and
+     `components/ui/TopSearchBar.js` (the chrome blocks).
+3. **Cross-brief commitments.** None. Each brief lands a complete
+   migration of its scope; no brief introduces a stub or forward
+   declaration that another brief resolves. Brief 7's grep gate
+   passes ONLY AFTER Briefs 1-6 land — that's a dispatch ordering
+   constraint (encoded in `depends_on:`), not a cross-brief
+   commitment in the role-architect sense.
+
+All checks passed. No briefs need revision.
diff --git a/.convoys/unify-glass-panel-surfaces/brief-1-upgrade-glass-surface-primitive.md b/.convoys/unify-glass-panel-surfaces/brief-1-upgrade-glass-surface-primitive.md
new file mode 100644
index 0000000..28a2c88
--- /dev/null
+++ b/.convoys/unify-glass-panel-surfaces/brief-1-upgrade-glass-surface-primitive.md
@@ -0,0 +1,82 @@
+---
+convoy: unify-glass-panel-surfaces
+brief_number: 1
+depends_on: []
+files:
+  - components/ui/GlassSurface.js
+  - test/components/ui-primitives.test.js
+---
+
+# Brief 1: Upgrade `<GlassSurface>` primitive with `cornerLights` prop
+
+## Goal (1 sentence)
+
+Add a `cornerLights = 'subtle' | 'chrome' | 'none'` prop to `<GlassSurface>` and compose the 4-layer gradient-border pattern (padding-box fill + two border-box corner radials + `--chip-border-base` base) on top of the existing `tint` / `blur` / `rim` / `elevation` props so every consumer (Modal, StatCard, landing feature cards) picks up corner catch-lights automatically.
+
+## Files in scope (do not edit anything else)
+
+- `components/ui/GlassSurface.js` — add the prop, compose the 4-layer `background` and the `border: 1px solid transparent` when the prop is not `'none'`.
+- `test/components/ui-primitives.test.js` — add 3 corner-light assertions to the existing `describe('GlassSurface', ...)` block (or add the block if it doesn't exist; today the file covers Button / Input / SearchBar only, so a new top-level `describe('GlassSurface', ...)` block at the end of the file is correct).
+
+**Out of scope:** every consumer of `<GlassSurface>` — `components/ui/Modal.js`, `components/ui/StatCard.js`, `pages/index.js` landing cards. Their behavior changes via the new default; no edits required. Do NOT modify them in this PR.
+
+## Conventions to follow
+
+- **Verbatim 4-layer recipe** (mirrors `.glass-panel-strong` post-PR #118; see `styles/globals.css`):
+  ```js
+  // When cornerLights === 'subtle' (default):
+  background:
+    `linear-gradient(var(--glass-surface-${tint}), var(--glass-surface-${tint})) padding-box,
+     radial-gradient(at 0% 100%, var(--corner-light-warm-subtle) 0%, transparent 42%) border-box,
+     radial-gradient(at 100% 0%, var(--corner-light-cool-subtle) 0%, transparent 42%) border-box,
+     var(--chip-border-base) border-box`
+  // When cornerLights === 'chrome':
+  // Same recipe; swap --corner-light-warm-subtle → --corner-light-warm
+  // and --corner-light-cool-subtle → --corner-light-cool.
+  // When cornerLights === 'none':
+  // Today's behavior — single layer background: var(--glass-surface-${tint}).
+  // No border declaration.
+  ```
+- **Border declaration**: when `cornerLights !== 'none'`, also set
+  `border: '1px solid transparent'` on the composed style. This is what
+  reveals the border-box gradient layers. When `'none'`, omit the
+  border entirely (preserve today's box-model).
+- **Shadow stack**: unchanged. The existing `RIM_SHADOWS` /
+  `ELEVATION_SHADOWS` composition logic stays exactly as written. The
+  corner-light layers live in `background`, not `box-shadow`.
+- **Default value**: `cornerLights = 'subtle'`. This is the new default
+  for every consumer that doesn't pass the prop explicitly.
+- **Backdrop filter**: unchanged. `backdropFilter` /
+  `WebkitBackdropFilter` continue to compose from `blur` / `tint` exactly
+  as today. Do not touch this line.
+- **`style` prop merging**: today the composed style spreads
+  `...style` LAST so caller overrides win. Preserve that exactly — a
+  caller passing `style={{ border: '1px solid #f00' }}` overrides the
+  transparent border. This is the explicit escape hatch any consumer
+  who needs a real border can use without setting `cornerLights='none'`.
+- **AGENTS.md / no-go zones**: `styles/globals.css` is untouched in
+  this brief (Brief 5 owns the `.card` deletion; no other CSS edits in
+  this convoy). The `--corner-light-warm-subtle` /
+  `--corner-light-cool-subtle` /  `--corner-light-warm` /
+  `--corner-light-cool` / `--chip-border-base` tokens all exist
+  in `:root` and `[data-theme="dark"]` per PR #118 — verified.
+
+## Acceptance criteria
+
+- [ ] `<GlassSurface>` accepts a `cornerLights` prop with values `'subtle'` (default), `'chrome'`, `'none'`. Any other value falls back to `'subtle'` (mirror the existing `RIM_SHADOWS[rim] ?? []` defensive fallback shape).
+- [ ] When `cornerLights='subtle'`, rendered output includes:
+  - `background:` containing both `var(--corner-light-warm-subtle)` and `var(--corner-light-cool-subtle)`.
+  - `border:` value `1px solid transparent`.
+- [ ] When `cornerLights='chrome'`, rendered output includes:
+  - `background:` containing `var(--corner-light-warm)` and `var(--corner-light-cool)` (NOT the `-subtle` variants).
+  - `border:` value `1px solid transparent`.
+- [ ] When `cornerLights='none'`, rendered output is identical to today: single-layer `background: var(--glass-surface-${tint})`, no `border` declaration in the composed style.
+- [ ] `tint` / `blur` / `rim` / `elevation` / `as` / `style` / `className` props all behave exactly as before. Existing consumers continue to compile + render without prop changes.
+- [ ] `test/components/ui-primitives.test.js` has 3 new assertions (one per `cornerLights` value) verifying the inline style attribute or the rendered element's `style.background` / `style.border` accordingly. Use the existing `render` + `screen.getByRole` / `getByText` / `container` patterns from the file; don't pull in new test deps.
+- [ ] `npm run lint` passes.
+- [ ] `npm run test:run` is 22+/22+ (today's count is 21; this brief adds at least 3 — exact count depends on whether you split assertions across multiple `it(...)` blocks).
+- [ ] No edits to files outside the two listed in `files:` above. **No consumer migrations in this PR.** Brief 3 + 4 will compose against the new default; Briefs 2 + 5 + 6 don't use the primitive.
+
+## Rationale (≤3 sentences)
+
+The primitive is the single largest leverage point in the convoy — upgrading it ripples through `<Modal>`, `<StatCard>`, and the landing-page feature cards in one commit. Defaulting to `'subtle'` matches the PR #118 token-tier decision and means downstream consumers don't need to opt in. The `'chrome'` + `'none'` values exist so the prop is future-proof: floating chrome can opt up without handrolling, and any GPU-constrained tile (e.g. a future <CardItem>-like surface) can opt out cleanly.
diff --git a/.convoys/unify-glass-panel-surfaces/brief-2-auth-form-cards.md b/.convoys/unify-glass-panel-surfaces/brief-2-auth-form-cards.md
new file mode 100644
index 0000000..aa7074b
--- /dev/null
+++ b/.convoys/unify-glass-panel-surfaces/brief-2-auth-form-cards.md
@@ -0,0 +1,79 @@
+---
+convoy: unify-glass-panel-surfaces
+brief_number: 2
+depends_on: []
+files:
+  - pages/login.js
+  - pages/signup.js
+---
+
+# Brief 2: Migrate auth form cards to `.glass-panel-strong`
+
+## Goal (1 sentence)
+
+Replace the handrolled `rgba(var(--bg-secondary-rgb), 0.85)` + `backdrop-blur-sm` glass imitation on the login + signup form cards with the canonical `.glass-panel-strong rounded-2xl p-8` className so the auth flow shares the rest of the app's surface treatment (corner catch-lights, system blur tier, gradient border).
+
+## Files in scope (do not edit anything else)
+
+- `pages/login.js` — replace the form-card container `<div>` (currently L82-88, look for the `className="p-8 rounded-2xl shadow-2xl backdrop-blur-sm border border-opacity-20"` + inline `backgroundColor: 'rgba(var(--bg-secondary-rgb), 0.85)'`).
+- `pages/signup.js` — same migration on the form-card container (currently L227-232 in the symmetric `<div className="p-8 rounded-2xl shadow-2xl backdrop-blur-sm ...">` block).
+
+**Out of scope:** the surrounding layout/header on either page, the `<Input>` / `<Button>` children, error/success banners, any of the legend/divider/social-button styling below the form. Do not touch them.
+
+## Conventions to follow
+
+- **Verbatim replacement** for each card container:
+  ```jsx
+  // Before:
+  <div
+    className="p-8 rounded-2xl shadow-2xl backdrop-blur-sm border border-opacity-20"
+    style={{
+      backgroundColor: 'rgba(var(--bg-secondary-rgb), 0.85)',
+      borderColor: 'var(--border)',
+    }}
+  >
+
+  // After:
+  <div className="glass-panel-strong rounded-2xl p-8">
+  ```
+  The `style={{}}` block is removed entirely. `.glass-panel-strong`
+  already composes the background, blur, rim, elevation, and the
+  gradient-border treatment.
+- **No new imports.** This brief does NOT use `<GlassSurface>`
+  directly — the className path is correct because (a) it's an HTML
+  div with no compositional requirements, (b) the existing
+  `.glass-panel-strong` class is the documented canonical shape per
+  the design audit, and (c) using the class keeps the diff minimal.
+- **Preserve children verbatim.** The `<form>`, every `<Input>`, every
+  `<Button>`, the error banner, the social-sign-in divider, the
+  "Don't have an account?" footer link — all stay byte-identical.
+- **Do not adjust the surrounding header block** (Deck Hearth logo +
+  greeting text); only the form-card `<div>` itself migrates.
+- **Tailwind safelist note:** `.glass-panel-strong` is defined in
+  `styles/globals.css` as a plain CSS class (not a Tailwind
+  utility). It's already used in `<Layout>` and elsewhere, so the
+  build picks it up via the `@layer` block. No `tailwind.config.js`
+  edit needed.
+
+## Acceptance criteria
+
+- [ ] `pages/login.js` form-card `<div>` uses
+      `className="glass-panel-strong rounded-2xl p-8"` and carries
+      no inline `backgroundColor` / `borderColor` style.
+- [ ] `pages/signup.js` form-card `<div>` uses
+      `className="glass-panel-strong rounded-2xl p-8"` and carries
+      no inline `backgroundColor` / `borderColor` style.
+- [ ] All form fields, labels, buttons, error banners, and footer
+      links render identically post-migration (manual smoke;
+      `tests/smoke/auth.spec.js` continues to pass without edits).
+- [ ] No edits to other files (e.g. no token additions in
+      `styles/globals.css`, no new components in `components/ui/`).
+- [ ] `npm run lint` passes; `npm run test:run` is unchanged
+      (no test additions needed for a pure className swap).
+- [ ] Dark mode: card remains legible against the body background
+      gradient (verify manually; before-shot vs after-shot screenshot
+      attached to the PR description).
+
+## Rationale (≤3 sentences)
+
+These two cards are the first surfaces a new user sees and they predate the `.glass-panel-strong` system; aligning them is both a correctness fix (the `rgba(--bg-secondary-rgb, 0.85)` shape doesn't compose corner lights) and a consistency win. The migration is a pure className swap with zero behavioral change — the lowest-risk brief in the convoy.
diff --git a/.convoys/unify-glass-panel-surfaces/brief-3-floating-popovers.md b/.convoys/unify-glass-panel-surfaces/brief-3-floating-popovers.md
new file mode 100644
index 0000000..69383aa
--- /dev/null
+++ b/.convoys/unify-glass-panel-surfaces/brief-3-floating-popovers.md
@@ -0,0 +1,154 @@
+---
+convoy: unify-glass-panel-surfaces
+brief_number: 3
+depends_on: [1]
+files:
+  - components/Layout.js
+  - components/ui/TopSearchBar.js
+  - test/components/Layout.test.js
+---
+
+# Brief 3: Migrate floating popovers to `.glass-panel-strong`
+
+## Goal (1 sentence)
+
+Migrate three floating surfaces — the sidebar profile dropdown, the mobile drawer, and the TopSearchBar `<UserMenu>` dropdown — from inline `var(--glass-surface-*)` + `backdropFilter` styles to the canonical `.glass-panel-strong` className, while preserving their existing box-shadow chains (ember rim for the profile dropdown; pronounced elevation for the drawer) via inline override.
+
+## Files in scope (do not edit anything else)
+
+- `components/Layout.js`
+  - Sidebar profile dropdown panel (currently around L88-96 inside `<UserProfileDropdown>`).
+  - Mobile drawer slide-in panel (currently around L698-709 — the `<div>` that takes `${isMobileMenuOpen ? 'translate-x-0' : '-translate-x-full'}` and the `style={{ background: 'var(--glass-surface-mid)', backdropFilter: '...', ... }}` block).
+  - **Do NOT touch** the sidebar nav-chip block (around L858-861); that block intentionally uses full-intensity chrome corner lights and stays handrolled — Decision 4 of this convoy's Architecture allowlists it explicitly.
+- `components/ui/TopSearchBar.js`
+  - `<UserMenu>` dropdown panel (around L214-222 — the absolutely-positioned `<div>` with the `background: 'var(--glass-surface-high)'` + `backdropFilter` block).
+  - **Do NOT touch** the TopSearchBar `<header>` block itself; that block also intentionally uses full-intensity chrome corner lights and is allowlisted.
+- `test/components/Layout.test.js`
+  - Add regression-lock assertions that the sidebar profile dropdown and mobile drawer render with `.glass-panel-strong` in their className.
+
+**Out of scope:** the rest of `Layout.js` (`<DesktopSidebar>`, `<MobileNavigation>`, `<UserProfileDropdown>` trigger button, `<DailyEmberWidget>`, etc.), the rest of `TopSearchBar.js` (search input, `<CommandPaletteModal>` link, etc.), and any deeper menu-item styling (e.g. `hover:bg-gray-50 dark:hover:bg-gray-700` cleanup — that's owned by PR #117 / `design-sweep-pass`).
+
+## Conventions to follow
+
+- **Verbatim replacement** for each floating-surface `<div>`:
+
+  ### A. Sidebar profile dropdown panel (Layout.js)
+  ```jsx
+  // Before:
+  <div
+    className="absolute bottom-full left-0 right-0 mb-2 rounded-xl z-20"
+    style={{
+      background: 'var(--glass-surface-high)',
+      backdropFilter: 'blur(var(--glass-blur-mid)) saturate(var(--glass-saturate))',
+      WebkitBackdropFilter: 'blur(var(--glass-blur-mid)) saturate(var(--glass-saturate))',
+      boxShadow: 'var(--rim-light-inner), var(--ember-rim-subtle), var(--elevation-ambient)',
+    }}
+  >
+
+  // After:
+  <div
+    className="glass-panel-strong absolute bottom-full left-0 right-0 mb-2 rounded-xl z-20"
+    style={{
+      boxShadow:
+        'var(--rim-light-inner), var(--ember-rim-subtle), var(--elevation-ambient)',
+    }}
+  >
+  ```
+  The `background` + `backdropFilter` + `WebkitBackdropFilter` inline
+  styles are removed (the class provides them). The `boxShadow` is
+  **preserved as an inline override** because `.glass-panel-strong`
+  ships only `var(--rim-light-inner), var(--elevation-ambient)` —
+  we need the `var(--ember-rim-subtle)` middle entry to keep the
+  ember emphasis the dropdown is known for.
+
+  ### B. Mobile drawer panel (Layout.js)
+  ```jsx
+  // Before:
+  <div
+    className={`
+      md:hidden fixed inset-y-0 left-0 z-50 w-64 transform transition-transform duration-300 ease-in-out
+      ${isMobileMenuOpen ? 'translate-x-0' : '-translate-x-full'}
+    `}
+    style={{
+      background: 'var(--glass-surface-mid)',
+      backdropFilter: 'blur(var(--glass-blur-mid)) saturate(var(--glass-saturate))',
+      WebkitBackdropFilter: 'blur(var(--glass-blur-mid)) saturate(var(--glass-saturate))',
+      boxShadow: 'var(--rim-light-inner), var(--rim-light-outer), var(--elevation-pronounced)',
+    }}
+  >
+
+  // After:
+  <div
+    className={`
+      glass-panel-strong md:hidden fixed inset-y-0 left-0 z-50 w-64 transform transition-transform duration-300 ease-in-out
+      ${isMobileMenuOpen ? 'translate-x-0' : '-translate-x-full'}
+    `}
+    style={{
+      boxShadow:
+        'var(--rim-light-inner), var(--rim-light-outer), var(--elevation-pronounced)',
+    }}
+  >
+  ```
+  Same pattern. `boxShadow` is preserved as an inline override so the
+  drawer keeps its `var(--elevation-pronounced)` (heavier) elevation
+  + `var(--rim-light-outer)` (which `.glass-panel-strong` doesn't
+  ship by default).
+
+  ### C. TopSearchBar `<UserMenu>` dropdown panel
+  ```jsx
+  // Before — the absolutely-positioned <div> nested inside UserMenu,
+  // around L214-222 of TopSearchBar.js:
+  <div
+    className="absolute right-0 top-full mt-2 w-56 rounded-xl py-2 z-50"
+    style={{
+      background: 'var(--glass-surface-high)',
+      backdropFilter: 'blur(var(--glass-blur-mid)) saturate(var(--glass-saturate))',
+      WebkitBackdropFilter: 'blur(var(--glass-blur-mid)) saturate(var(--glass-saturate))',
+      boxShadow: 'var(--rim-light-inner), var(--ember-rim-subtle), var(--elevation-ambient)',
+    }}
+  >
+
+  // After:
+  <div
+    className="glass-panel-strong absolute right-0 top-full mt-2 w-56 rounded-xl py-2 z-50"
+    style={{
+      boxShadow:
+        'var(--rim-light-inner), var(--ember-rim-subtle), var(--elevation-ambient)',
+    }}
+  >
+  ```
+  **NOTE:** confirm the verbatim shape of the current TopSearchBar
+  UserMenu dropdown against L214-222 before editing — the inline
+  className/style block might already have minor differences from the
+  "before" sample above (e.g. different `w-` width, different
+  `mt-` margin). Migrate the actual block, not this sample; the
+  sample only shows the shape of the swap.
+
+- **Preserve all children verbatim.** Menu items, dividers, icons,
+  click handlers — all stay byte-identical.
+- **Do not touch the existing `hover:bg-gray-50 dark:hover:bg-gray-700`
+  classes on individual menu items.** PR #117 (`design-sweep-pass`)
+  is already migrating those to `nav-item-hover`; this brief MUST NOT
+  duplicate that work or it'll merge-conflict. If PR #117 has already
+  merged when this brief dispatches, the relevant lines may already
+  read `nav-item-hover` — that's fine; just leave them alone.
+- **Test additions in `test/components/Layout.test.js`:**
+  - One assertion for the sidebar profile dropdown: render `<Layout user={mockUser} />`, simulate clicking the user-profile trigger to open the dropdown, assert the dropdown panel `className` contains `'glass-panel-strong'`.
+  - One assertion for the mobile drawer: render `<Layout user={mockUser} />`, simulate opening the drawer (today the test file already has a way to set `isMobileMenuOpen`; if not, render with the relevant prop / state directly), assert the drawer panel `className` contains `'glass-panel-strong'`.
+  - Use the existing test file's patterns (jsdom env, `cleanup()` afterEach, `render` + `screen` from `@testing-library/react`). Do NOT add new test dependencies. Do NOT lower any existing assertion in the file.
+
+## Acceptance criteria
+
+- [ ] Sidebar profile dropdown panel uses `className="glass-panel-strong ..."` and carries an inline `boxShadow` with `var(--rim-light-inner), var(--ember-rim-subtle), var(--elevation-ambient)`.
+- [ ] Mobile drawer panel uses `className="glass-panel-strong ..."` and carries an inline `boxShadow` with `var(--rim-light-inner), var(--rim-light-outer), var(--elevation-pronounced)`.
+- [ ] TopSearchBar `<UserMenu>` dropdown panel uses `className="glass-panel-strong ..."` and carries an inline `boxShadow` with `var(--rim-light-inner), var(--ember-rim-subtle), var(--elevation-ambient)`.
+- [ ] No inline `background: 'var(--glass-surface-*)'` or `backdropFilter` style remains on any of the three migrated `<div>`s.
+- [ ] Sidebar nav-chip block (Layout.js ~L858-861) UNCHANGED. TopSearchBar `<header>` block UNCHANGED. (Architect's allowlist for chrome-tier surfaces.)
+- [ ] `test/components/Layout.test.js` has at least 2 new assertions (sidebar dropdown + mobile drawer carry `glass-panel-strong`).
+- [ ] `npm run lint` + `npm run test:run` (23+/23+, post Brief 1) pass.
+- [ ] Manual: open sidebar profile dropdown, open mobile drawer, open TopSearchBar UserMenu dropdown in light + dark mode — all three render with visible corner catch-lights and the appropriate shadow elevation.
+- [ ] No edits to files outside the 3 listed in `files:` above.
+
+## Rationale (≤3 sentences)
+
+These three popovers are the most visible floating surfaces in the authenticated app and they predate the gradient-border system; migrating them is the largest visual-correctness win of the convoy after Brief 1. Preserving the existing box-shadow chains via inline override (rather than letting them snap to the class default) is non-negotiable — the `var(--ember-rim-subtle)` ember emphasis on the profile menu and the `var(--elevation-pronounced)` weight on the drawer are deliberate design choices that must survive the class swap.
diff --git a/.convoys/unify-glass-panel-surfaces/brief-4-bulk-selection-toolbar.md b/.convoys/unify-glass-panel-surfaces/brief-4-bulk-selection-toolbar.md
new file mode 100644
index 0000000..8c2ad80
--- /dev/null
+++ b/.convoys/unify-glass-panel-surfaces/brief-4-bulk-selection-toolbar.md
@@ -0,0 +1,123 @@
+---
+convoy: unify-glass-panel-surfaces
+brief_number: 4
+depends_on: [1]
+files:
+  - components/BulkSelectionToolbar.js
+---
+
+# Brief 4: Migrate `BulkSelectionToolbar` to `.glass-panel-strong` + token sweep
+
+## Goal (1 sentence)
+
+Replace the bulk-select toolbar's `bg-white border-gray-200` Tailwind shape with `.glass-panel-strong rounded-2xl` so the floating toolbar is legible in dark mode, and sweep its interior hardcoded gray/red Tailwind classes (`text-gray-700`, `hover:bg-gray-100`, `text-red-600 hover:bg-red-50`) to token-driven inline styles + `nav-item-hover` for consistency with the rest of the post-PR-#117 surfaces.
+
+## Files in scope (do not edit anything else)
+
+- `components/BulkSelectionToolbar.js` — the entire file. Specifically:
+  - L47 outer toolbar `<div>` — swap `bg-white rounded-xl shadow-2xl border border-gray-200` to `glass-panel-strong rounded-2xl shadow-2xl`. (Keep `shadow-2xl` for now — the toolbar floats over arbitrary content and the extra weight is intentional. `.glass-panel-strong`'s `var(--rim-light-inner), var(--elevation-ambient)` chain doesn't include it; chain it via inline override.)
+  - L60 divider — swap `bg-gray-300` to a CSS-var-driven divider color (`backgroundColor: 'var(--border)'` inline, since `bg-gray-300` doesn't theme).
+  - L53 selection-count label — swap `text-gray-700` to `style={{ color: 'var(--text-primary)' }}`.
+  - L108 more-actions trigger — swap `text-gray-600 hover:text-gray-800 hover:bg-gray-100` to use `var(--text-secondary)` + `nav-item-hover`, with `rounded-lg` upgraded to `rounded-xl`.
+  - L117 more-actions dropdown — swap `bg-white rounded-lg shadow-xl border border-gray-200` to `glass-panel-strong rounded-xl`. (Drop `shadow-xl`; the class provides shadow. Keep `border` semantics out — the class's gradient border is the new look.)
+  - L118-141 menu items — swap `text-gray-700 hover:bg-gray-100` to `var(--text-primary)` + `nav-item-hover` className.
+  - L144 menu divider — swap `border-gray-200` to `borderColor: 'var(--border)'`.
+  - L146-157 destructive menu item — swap `text-red-600 hover:bg-red-50` to `color: 'var(--accent-danger)'` (verify the token exists in `globals.css`; if not, fall back to `rgb(239, 68, 68)` and document) + `nav-item-hover` className.
+  - L160-end "Clear Selection" button — sweep any remaining `text-gray-*` / `bg-gray-*` hardcodes the same way.
+
+**Out of scope:** the parent components that consume `BulkSelectionToolbar` (e.g. `pages/my-cards.js`, `components/CollectionPageView.js`); the props contract (`selectedCount`, `selectedCards`, all the `onBulk*` callbacks); any of the SVG icon paths.
+
+## Conventions to follow
+
+- **Outer toolbar:**
+  ```jsx
+  // Before:
+  <div className="bg-white rounded-xl shadow-2xl border border-gray-200 px-3 sm:px-6 py-3 sm:py-4 flex items-center justify-between sm:justify-start sm:space-x-4 sm:min-w-96">
+
+  // After:
+  <div
+    className="glass-panel-strong rounded-2xl px-3 sm:px-6 py-3 sm:py-4 flex items-center justify-between sm:justify-start sm:space-x-4 sm:min-w-96"
+    style={{
+      // Chain shadow-2xl-equivalent depth onto the class's existing
+      // rim+ambient stack so the floating toolbar still reads as
+      // elevated over arbitrary page content.
+      boxShadow:
+        'var(--rim-light-inner), var(--elevation-pronounced)',
+    }}
+  >
+  ```
+  Rationale: `shadow-2xl` is a hardcoded RGB; `var(--elevation-pronounced)` is the system equivalent and respects the theme.
+
+- **More-actions dropdown:**
+  ```jsx
+  // Before:
+  <div className="absolute bottom-full right-0 mb-2 bg-white rounded-lg shadow-xl border border-gray-200 py-2 min-w-48">
+
+  // After:
+  <div className="glass-panel-strong absolute bottom-full right-0 mb-2 rounded-xl py-2 min-w-48">
+  ```
+  No inline-style override needed — the class's default
+  `var(--rim-light-inner), var(--elevation-ambient)` is the right
+  weight for an inner dropdown.
+
+- **Menu items** (the `<button>` rows inside the dropdown):
+  ```jsx
+  // Before (non-destructive item):
+  <button className="w-full px-4 py-2 text-left text-sm text-gray-700 hover:bg-gray-100 flex items-center space-x-2">
+
+  // After:
+  <button
+    className="nav-item-hover w-full px-4 py-2 text-left text-sm flex items-center space-x-2"
+    style={{ color: 'var(--text-primary)' }}
+  >
+
+  // Before (destructive item):
+  <button className="w-full px-4 py-2 text-left text-sm text-red-600 hover:bg-red-50 flex items-center space-x-2">
+
+  // After:
+  <button
+    className="nav-item-hover w-full px-4 py-2 text-left text-sm flex items-center space-x-2"
+    style={{ color: 'var(--accent-danger)' }}
+  >
+  ```
+
+- **Verify `--accent-danger` exists** before using it. Quick check:
+  search `styles/globals.css` for `--accent-danger`. If absent, the
+  PR #117 sweep likely defines it; if it's still absent post-#117,
+  use `color: 'rgb(239, 68, 68)'` (the literal Tailwind `red-500`
+  RGB) and add a note in the PR description requesting follow-up to
+  introduce the token in a separate PR. Do NOT add the token in this
+  brief — token additions belong in a design-system PR.
+
+- **Verify `nav-item-hover` exists** — it's defined in
+  `styles/globals.css` and used widely post-PR #117. If for some
+  reason this brief dispatches before PR #117 lands, escalate to the
+  conductor — that's a dependency violation (Brief 4 depends_on: [1]
+  but transitively depends on the post-PR-#117 token tier).
+
+- **Verbatim children:** SVG icon paths, button text labels, click
+  handler bindings, the `VOCAB.MY_COLLECTION` / `VOCAB.REMOVE_FROM_MY_COLLECTION`
+  imports, the `setShowActions(false)` flow — all stay byte-identical.
+
+- **`px-2 sm:px-3 py-2` action buttons** (the 3 colored quick-action
+  buttons: Collection / Deck / My Collection): leave them alone. They
+  use `--accent-flame`, `--accent-gold`, `--accent-wood` already and
+  the `onMouseEnter`/`onMouseLeave` swap is a known pattern. Touching
+  them is scope expansion.
+
+## Acceptance criteria
+
+- [ ] Outer toolbar `<div>` uses `glass-panel-strong rounded-2xl` and an inline `boxShadow` matching the verbatim shape above.
+- [ ] More-actions dropdown `<div>` uses `glass-panel-strong rounded-xl` and no inline `background` / `border` / `shadow` style.
+- [ ] All `text-gray-{600,700,800}` classes within the toolbar are replaced with `style={{ color: 'var(--text-{primary,secondary}) ' }}` or `nav-item-hover` className.
+- [ ] All `hover:bg-gray-{50,100}` classes are replaced with `nav-item-hover` className.
+- [ ] The destructive "Delete Selected" item uses `var(--accent-danger)` (or the documented `rgb(239,68,68)` fallback) — NOT `text-red-600`.
+- [ ] The L60 divider uses `style={{ backgroundColor: 'var(--border)' }}` — NOT `bg-gray-300`.
+- [ ] The 3 quick-action buttons (Collection / Deck / My Collection at L64-101) are UNCHANGED. The "Clear Selection" button at the end may need a small text-color swap; that's in scope. Everything else is left alone.
+- [ ] **Dark-mode verification**: open the bulk-select toolbar (select 2+ cards on `/my-cards` or `/collection/[id]`), confirm it renders legibly with corner catch-lights and proper text contrast in BOTH light and dark themes. Attach before/after dark-mode screenshots to the PR description.
+- [ ] `npm run lint` + `npm run test:run` (23+/23+, post Brief 1) pass.
+- [ ] No edits to files outside `components/BulkSelectionToolbar.js`.
+
+## Rationale (≤3 sentences)
+
+The toolbar is currently invisible in dark mode because `bg-white` doesn't theme — this PR is a correctness fix as much as a design unification. Sweeping the interior hardcoded grays + reds in the same PR is cheap (the file is small, ~200 lines) and prevents a follow-up convoy from having to revisit the file. The 3 colored quick-action buttons stay as-is because they use the right tokens already and a token sweep there is scope expansion.
diff --git a/.convoys/unify-glass-panel-surfaces/brief-5-retire-card-class.md b/.convoys/unify-glass-panel-surfaces/brief-5-retire-card-class.md
new file mode 100644
index 0000000..35a09fb
--- /dev/null
+++ b/.convoys/unify-glass-panel-surfaces/brief-5-retire-card-class.md
@@ -0,0 +1,87 @@
+---
+convoy: unify-glass-panel-surfaces
+brief_number: 5
+depends_on: []
+files:
+  - pages/profile.js
+  - pages/settings.js
+  - pages/community/collections.js
+  - components/CollectionsPageView.js
+  - styles/globals.css
+---
+
+# Brief 5: Retire `.card` class; migrate 8 consumers to `.glass-panel`
+
+## Goal (1 sentence)
+
+Delete the legacy `.card` class from `styles/globals.css` and migrate all 8 consumers (3 in `pages/profile.js`, 2 in `pages/settings.js`, 1 in `pages/community/collections.js`, 1 in `components/CollectionsPageView.js`, plus 1 likely-missed inline `className="card hover:..."` instance) to `glass-panel rounded-3xl p-{4|6}` so the app converges on a single panel vocabulary.
+
+## Files in scope (do not edit anything else)
+
+- `styles/globals.css` — delete the entire `.card { ... }` rule block (currently 4 lines at L729-733):
+  ```css
+  .card {
+    @apply rounded-3xl shadow-lg p-6 transition-all duration-300;
+    background-color: var(--bg-primary);
+    border: 1px solid var(--border);
+  }
+  ```
+- `pages/profile.js` — 3 `className="card p-6 [...]"` swaps. Likely sites (verify line numbers against current file): L290 (`card p-6 text-center` — the profile avatar card), L397 (`card p-6 mt-6` — the stats card), L432 (`card p-6` — the activity card).
+- `pages/settings.js` — 2 `className="card ..."` swaps. Likely sites: L266 (`card p-4`), L290 (`card p-6`).
+- `pages/community/collections.js` — 1 swap. Likely site: L266 (`card group cursor-pointer hover:shadow-lg transition-all duration-200`).
+- `components/CollectionsPageView.js` — 1 swap. Likely site: L125 (`card hover:shadow-xl transition-all duration-300 cursor-pointer group`).
+
+**Out of scope:** everything else in those files (page-level layout, header strips, modal wrappers, button styling, inline content). Only the `.card` className references migrate.
+
+## Conventions to follow
+
+- **Swap pattern (default):**
+  ```jsx
+  // Before:
+  <div className="card p-6 [...whatever-else]">
+
+  // After:
+  <div className="glass-panel rounded-3xl p-6 [...whatever-else]">
+  ```
+  Verbatim: replace the standalone token `card` with `glass-panel rounded-3xl`. **Preserve any sibling tokens** (`p-4` / `p-6` / `text-center` / `mt-6` / `group` / `cursor-pointer` / `hover:shadow-lg` / `transition-all duration-200`, etc.). The legacy `.card` `@apply`'d `rounded-3xl shadow-lg p-6 transition-all duration-300`; we keep `rounded-3xl` explicitly and let `.glass-panel`'s composed shadow stack replace `shadow-lg` (which is the right move — `.glass-panel`'s tokens theme correctly, `shadow-lg` doesn't).
+
+- **`p-6` handling:** the legacy class baked in `p-6`. If the consumer wrote `card p-4` (`pages/settings.js` L266), the explicit `p-4` already wins via Tailwind's cascade — `p-4` overrides the `@apply rounded-3xl shadow-lg p-6` because they're at the same specificity and `p-4` is the later-defined declaration in compiled output. After migration, the explicit `p-4` continues to win because there's no longer any baked-in `p-6`. **Bottom line:** preserve whatever padding token the consumer wrote; don't normalize to `p-6`.
+
+- **`transition-all duration-300` handling:** the legacy class baked in `transition-all duration-300`. Most consumers ALSO wrote it explicitly (e.g. `CollectionsPageView.js` L125: `transition-all duration-300`). If a consumer relied on the baked-in version, the swap loses the transition — add `transition-all duration-300` explicitly to that consumer's className post-swap. **Audit each site for whether the transition is referenced; add it back where needed.**
+
+- **`shadow-lg` handling:** the legacy class baked in `shadow-lg`. `.glass-panel`'s composed `box-shadow` (rim + elevation-ambient) is the right replacement; do NOT carry `shadow-lg` over.
+
+- **`pages/community/collections.js` site (L266):** the consumer adds `hover:shadow-lg` on top of the baked `shadow-lg`. The base shadow becomes `.glass-panel`'s shadow stack; the `hover:shadow-lg` is a hardcoded Tailwind shadow. Decision: replace `hover:shadow-lg` with no `hover:` override — the corner-light gradient on `.glass-panel` is the new affordance. (If review shows this regresses hover legibility, fall back to inline `style={{}}` boxShadow override on `:hover` — but the simpler clean diff is to drop it.)
+
+- **`components/CollectionsPageView.js` site (L125):** same logic. Drop `hover:shadow-xl`.
+
+- **Escape hatch (per Decision D2):** if any one of the 8 consumers breaks visually post-swap (the architect doesn't expect this), revert that one site to inline `style={{ background: 'var(--bg-secondary)' }}` rather than holding the entire class-deletion. Document each escape-hatch use in the PR description with a screenshot. If MORE than 1 of 8 needs the escape hatch, the brief's reviewer is expected to push back and hold the deletion entirely — flag in PR comments.
+
+- **CSS deletion order:** delete the `.card { ... }` rule LAST, after all 8 className swaps are in place. (Pure ordering hygiene for diff-readability; CSS doesn't actually care.)
+
+- **Boundaries:** do NOT touch any other class in `globals.css` (no renaming `.glass-panel` → `.opaque-panel`, no adding new classes). Do NOT touch any pages or components not in the 4 listed.
+
+- **Grep verification before PR:**
+  ```bash
+  # Should return 0 matches anywhere in pages/ or components/:
+  rg "className=[\"\\'\\\`]card\\b" pages/ components/ --type js
+  # And the class definition itself:
+  rg "^\\.card \\{" styles/
+  ```
+  Both should return zero. If either returns anything, the migration is incomplete.
+
+## Acceptance criteria
+
+- [ ] `styles/globals.css` no longer contains a `.card { ... }` rule block.
+- [ ] Zero `className=".*\bcard\b.*"` references remain under `pages/` or `components/` (verified by the grep command above).
+- [ ] Each of the 8 migrated consumers renders with `.glass-panel rounded-3xl` and the original `p-{4|6}` padding intact.
+- [ ] Wherever the consumer relied on baked-in `transition-all duration-300` (audit by reviewing each site), the transition is added explicitly post-migration.
+- [ ] `hover:shadow-{lg,xl}` Tailwind overrides on community/CollectionsPageView are dropped (replaced by `.glass-panel`'s corner-light affordance).
+- [ ] No escape-hatch sites or — if any — at most 1 escape-hatch site, documented with a screenshot in the PR description.
+- [ ] `npm run lint` + `npm run test:run` (no test count change expected; 22+/22+ post Brief 1) pass.
+- [ ] Manual: visit `/profile`, `/settings`, `/community/collections`, `/community/collections` collection-detail in light + dark mode and confirm all panels render with corner catch-lights and theme-correct backgrounds. Attach before/after screenshots to PR.
+- [ ] No edits to files outside the 5 listed in `files:` above.
+
+## Rationale (≤3 sentences)
+
+`.card` predates the gradient-border system and is the only remaining "opaque solid panel" pattern in user-facing pages — a single panel vocabulary across the app is the convoy's success metric. The 8 consumers are all "panel-with-content-inside" surfaces with no GPU-budget constraint, so the migration is a clean className swap with no compositional surprises. Deleting the class outright (rather than keeping it as a documented fallback) prevents future agents from picking the wrong pattern by accident.
diff --git a/.convoys/unify-glass-panel-surfaces/brief-6-landing-nav-bar.md b/.convoys/unify-glass-panel-surfaces/brief-6-landing-nav-bar.md
new file mode 100644
index 0000000..614f525
--- /dev/null
+++ b/.convoys/unify-glass-panel-surfaces/brief-6-landing-nav-bar.md
@@ -0,0 +1,91 @@
+---
+convoy: unify-glass-panel-surfaces
+brief_number: 6
+depends_on: []
+files:
+  - pages/index.js
+---
+
+# Brief 6: Migrate landing nav bar to `.page-header-glass`
+
+## Goal (1 sentence)
+
+Replace the handrolled translucent `<nav>` on the public landing page with the canonical `.page-header-glass border-b` className, and drop the malformed `boxShadow: 'inset 0 1px 0 var(--rim-light-inner)'` (the token already contains its own `inset 0 1px 0`; double-wrapping breaks the cascade — documented in `styles/globals.css` L748-761).
+
+## Files in scope (do not edit anything else)
+
+- `pages/index.js` — the landing `<nav>` block, currently L61-72 (around the `border-b` declaration with inline `background: 'var(--glass-surface-mid)'` + `backdropFilter` + `boxShadow: 'inset 0 1px 0 var(--rim-light-inner)'`).
+
+**Out of scope:** everything else on the landing page (the Deck Hearth heading, the Sign In / Get Started buttons inside the nav, the hero section, feature cards, footer, the `if (user) return null;` redirect logic, etc.). The migration is a single `<nav>` element's `className` + `style` swap.
+
+## Conventions to follow
+
+- **Verbatim replacement:**
+  ```jsx
+  // Before:
+  <nav
+    className="border-b"
+    style={{
+      background: 'var(--glass-surface-mid)',
+      backdropFilter:
+        'blur(var(--glass-blur-mid)) saturate(var(--glass-saturate))',
+      WebkitBackdropFilter:
+        'blur(var(--glass-blur-mid)) saturate(var(--glass-saturate))',
+      borderColor: 'var(--border)',
+      boxShadow: 'inset 0 1px 0 var(--rim-light-inner)',
+    }}
+  >
+
+  // After:
+  <nav
+    className="page-header-glass border-b"
+    style={{
+      borderColor: 'var(--border)',
+    }}
+  >
+  ```
+  The class provides `background`, `backdrop-filter`, and the
+  correct `box-shadow` (which IS `var(--rim-light-inner), 0 1px 0
+  var(--border)` per its definition in `globals.css` L300-306). The
+  `border-b` Tailwind class needs the `borderColor` style to remain
+  themed; everything else collapses into the class.
+
+- **The malformed boxShadow is the bug:** `var(--rim-light-inner)` is
+  already a complete `inset 0 1px 0 <color>` declaration. Wrapping
+  it in another `inset 0 1px 0 ...` produces invalid CSS that the
+  browser silently drops. The class's own `box-shadow:
+  var(--rim-light-inner), 0 1px 0 var(--border)` does the right
+  thing. Removing the inline `boxShadow` line is the fix.
+
+- **Children verbatim:** the `<div className="max-w-7xl mx-auto ...">`
+  inside the `<nav>`, the `<AnimatedFireLogo>`, the heading, and the
+  Sign In / Get Started buttons all stay byte-identical. Do not
+  touch them.
+
+- **`.page-header-glass` placement check:** the class is defined in
+  `styles/globals.css` L300-306 and is already used by authenticated
+  pages' header strips (dashboard, my-cards, cards, etc.). It's safe
+  to reuse on the public landing page — the class has no auth
+  coupling.
+
+- **Nested backdrop-filter check** (boot-the-brief risk #6): the
+  landing page has no enclosing Layout (the `pages/index.js` redirect
+  logic returns `null` for authenticated users and renders its own
+  full-page `<div>` for anonymous visitors — no `<Layout>` wrapper).
+  So there's no risk of nested `backdrop-filter` stacking. Verified.
+
+- **AGENTS.md / no-go zones:** `pages/index.js` is an authored page,
+  not a generated artifact. Allowed.
+
+## Acceptance criteria
+
+- [ ] Landing `<nav>` uses `className="page-header-glass border-b"`.
+- [ ] Inline `style` on the `<nav>` retains only `borderColor: 'var(--border)'` — no `background`, no `backdropFilter`, no `WebkitBackdropFilter`, no `boxShadow`.
+- [ ] The Deck Hearth heading + Sign In / Get Started buttons inside the nav render identically.
+- [ ] Manual: open `/` in an incognito window (anonymous user), verify the nav bar renders with a subtle glassy treatment, an `inset 0 1px 0` highlight (the inner rim), and a 1px bottom border. Verify in both light and dark theme via the theme toggle (which appears on the public landing). Attach light + dark screenshots to PR.
+- [ ] `npm run lint` + `npm run test:run` pass (no test count change expected).
+- [ ] No edits to other files.
+
+## Rationale (≤3 sentences)
+
+The landing nav is the highest-visibility public surface in the app and currently both (a) duplicates `.page-header-glass`'s composition handrolled, and (b) ships a malformed `boxShadow` that silently breaks. Migrating to the canonical class is a one-line cleanup that fixes a latent bug. No corner-light treatment is appropriate here — the nav is full-bleed, so corner radials would land at viewport edges and never read as light sources; `.page-header-glass` (which uses the rim/border-only composition) is the right shape.
diff --git a/.convoys/unify-glass-panel-surfaces/brief-7-forbidden-bespoke-glass-surface-ci-gate.md b/.convoys/unify-glass-panel-surfaces/brief-7-forbidden-bespoke-glass-surface-ci-gate.md
new file mode 100644
index 0000000..6cdd79f
--- /dev/null
+++ b/.convoys/unify-glass-panel-surfaces/brief-7-forbidden-bespoke-glass-surface-ci-gate.md
@@ -0,0 +1,123 @@
+---
+convoy: unify-glass-panel-surfaces
+brief_number: 7
+depends_on: [1, 2, 3, 4, 5, 6]
+files:
+  - .github/workflows/ci.yml
+---
+
+# Brief 7: `forbidden-bespoke-glass-surface` CI gate
+
+## Goal (1 sentence)
+
+Add a `forbidden-bespoke-glass-surface` job to `.github/workflows/ci.yml` that fails the build if `var(--glass-surface-(low|mid|high))` appears in JSX inline-style usage under `components/**` or `pages/**`, with a curated allowlist for the documented chrome exceptions (Layout sidebar nav-chip block, TopSearchBar `<header>` block, and the `<GlassSurface>` primitive itself).
+
+## Files in scope (do not edit anything else)
+
+- `.github/workflows/ci.yml` — append a new job after the existing `forbidden-modal-shell-without-primitive` job (around L200) or alongside it in the `forbidden-*` cluster. Don't reorder existing jobs.
+
+**Out of scope:** every other file in the repo. This brief is pure CI surface; no application code changes.
+
+## Conventions to follow
+
+- **Model after the existing `forbidden-modal-shell-without-primitive` job** (`.github/workflows/ci.yml` L200-226). Same shape:
+  - Single `runs-on: ubuntu-latest` step that runs a `grep` command.
+  - Collects matches into a bash array.
+  - Iterates and emits `::error file=${f}::<reason>` for each match.
+  - Exits non-zero if any match remains.
+  - The grep pattern is bounded to `pages/ components/ -r --include='*.js'`.
+
+- **Verbatim job shape** (drop into `ci.yml` at the bottom of the forbidden-* cluster):
+  ```yaml
+    forbidden-bespoke-glass-surface:
+      name: No bespoke var(--glass-surface-*) inline styles
+      runs-on: ubuntu-latest
+      steps:
+        - uses: actions/checkout@v4
+        - name: Fail if JSX inline styles handroll glass surfaces
+          run: |
+            # unify-glass-panel-surfaces convoy (PR sequence after #118 +
+            # #117). Once Briefs 1-6 land, every panel-shaped surface in
+            # the app composes via .glass-panel / .glass-panel-strong /
+            # .page-header-glass / <GlassSurface>. Inline-style usage of
+            # var(--glass-surface-low|mid|high) under pages/ or
+            # components/ JSX is the regression vector this gate prevents.
+            #
+            # Allowlist (explicit, documented exceptions):
+            #   - components/ui/GlassSurface.js          # the primitive itself
+            #   - components/Layout.js                   # sidebar nav-chip chrome block (L858 area)
+            #   - components/ui/TopSearchBar.js          # <header> chrome block
+            # These three files intentionally compose handrolled chrome
+            # surfaces ratified in PR #116 (gradient borders) + the
+            # unify-glass-panel-surfaces architect decision D4. If you
+            # need to add a fourth allowlist entry, that's a design-system
+            # decision — open a new convoy.
+            ALLOWLIST=(
+              "components/ui/GlassSurface.js"
+              "components/Layout.js"
+              "components/ui/TopSearchBar.js"
+            )
+            FOUND=()
+            while IFS= read -r file; do
+              skip=false
+              for allowed in "${ALLOWLIST[@]}"; do
+                if [ "$file" = "$allowed" ]; then
+                  skip=true
+                  break
+                fi
+              done
+              if [ "$skip" = true ]; then
+                continue
+              fi
+              FOUND+=("$file")
+            done < <(grep -lE "var\\(--glass-surface-(low|mid|high)\\)" \
+                pages components -r --include='*.js' 2>/dev/null \
+                | sort -u || true)
+            if [ ${#FOUND[@]} -gt 0 ]; then
+              echo "::error::Bespoke var(--glass-surface-*) inline styles detected outside the documented allowlist."
+              echo "Use .glass-panel / .glass-panel-strong / .page-header-glass or compose <GlassSurface> from components/ui/ instead."
+              for f in "${FOUND[@]}"; do
+                echo "::error file=${f}::Replace inline var(--glass-surface-*) with the appropriate class or primitive."
+              done
+              exit 1
+            fi
+            echo "OK: no bespoke glass-surface inline styles outside the allowlist."
+  ```
+
+- **Allowlist hygiene:** the 3 allowlist entries are exact paths. If
+  the relevant chrome block in `components/Layout.js` is later
+  refactored into its own sub-component (e.g. `<SidebarNavChip>`),
+  the new file path replaces `components/Layout.js` in the
+  allowlist — that's the kind of edit a future PR would carry.
+
+- **Pre-merge negative test:** before opening this PR, run the
+  allowlist locally — verify that adding a scratch `style={{
+  background: 'var(--glass-surface-low)' }}` to a non-allowlisted
+  file (e.g. `pages/profile.js`) and re-running the grep produces a
+  match. Then revert the scratch change. Document the negative test
+  in the PR description (don't commit the scratch change).
+
+- **CI ordering:** Brief 7 depends on Briefs 1-6 ALL landing first.
+  If this job is added before any of the prior briefs ships, the
+  build will fail on the in-flight migrations (every site this
+  convoy is migrating IS currently a bespoke `var(--glass-surface-*)`
+  inline-style usage). Conductor MUST hold dispatch until 1-6 are
+  all on `main`.
+
+- **Boundaries:** do NOT touch any other CI job. Do NOT edit any
+  application file. Do NOT update AGENTS.md or other docs in this
+  PR (a separate AGENTS.md update can land alongside Brief 1 if
+  the architect wants — Brief 7 is pure CI).
+
+## Acceptance criteria
+
+- [ ] `.github/workflows/ci.yml` contains a new `forbidden-bespoke-glass-surface` job matching the verbatim shape above (job name, `runs-on`, checkout step, grep-and-allowlist shell block).
+- [ ] The allowlist has exactly 3 entries: `components/ui/GlassSurface.js`, `components/Layout.js`, `components/ui/TopSearchBar.js`.
+- [ ] On a clean post-Briefs-1-through-6 `main`, the new CI job is GREEN — no false positives. (Verify by running the grep locally before opening the PR.)
+- [ ] Negative test: adding a scratch `style={{ background: 'var(--glass-surface-low)' }}` to e.g. `pages/profile.js` and re-running the same grep command produces a match. (Documented in PR description; not committed.)
+- [ ] No other CI jobs are reordered, renamed, or modified.
+- [ ] PR description links back to this brief and to the convoy file (`.convoys/unify-glass-panel-surfaces.md`).
+
+## Rationale (≤3 sentences)
+
+The convoy's success metric is unification, but unification without a regression gate is half a fix — a future PR can re-introduce a bespoke `var(--glass-surface-*)` inline style and undo the work. Modeling the gate on the existing `forbidden-modal-shell-without-primitive` job keeps it consistent with the repo's CI vocabulary and makes the failure message actionable. The 3-entry allowlist is small and intentional; growing it requires an explicit design-system decision, which is the right friction for a convention-enforcing gate.