Merge pull request 'docs(link): the shard-name fallback, the atlas places shape, and two traps' (#76) from docs/protocol-3-smoke-findings into edge

Reviewed-on: #76

android/PLAN.md

@@ -970,14 +970,22 @@ push, and Play (M6–M8) follow the designed app.
         (`atlas`). Note the path: `/public/atlas`, **not** `/public/shard` — the atlas is static shard
         *content*, not live shard *state*, and unlike `/shard/*` it **is** `siteMode`-gated like
         `/posts` and `/wiki`, so a site in maintenance mode withholds it independently of the sidecar.
-        Two units/naming traps from `v3.md` §6.3: respawn delays are **seconds** throughout, and
-        `points` is a *count* on the search route while `spawners` is the *list* on the detail route.
+        Three traps. Two are units/naming, from `v3.md` §6.3: respawn delays are **seconds**
+        throughout, and `points` is a *count* on the search route while `spawners` is the *list* on
+        the detail route. The third is a **shape**: `places` is a list of
+        `{facet, label, spawners, maxAlive}` **objects**, not of place-name strings — it is the
+        aggregate the screen exists to show ("Shrines, Isamu-Jima, Yew"), it arrives only on the
+        detail route, and typing it `List<String>` makes that whole route fail to decode while the
+        request itself returns `200`.
     - **Verification** — the five-rung walk (`anonymous`, `logged_in`, `player`, `staff`, `admin`)
       against a local website on the cutover branch, per
       [`../link/v3.md`](../link/v3.md) §11 and the shard-visibility smoke harness; plus one pass with
       **every feature disabled** in Admin → Shard Visibility, confirming the app *hides* each surface
       instead of erroring on it. Unit tests cover the menu filter (role × feature set), the
-      `404`/`403`/`503` mapping, and DTO decode for each new shape.
+      `404`/`403`/`503` mapping, and DTO decode for each new shape. **Decode tests must feed real
+      captured JSON**, not DTOs built in Kotlin: the fakes under `data/api/fake/` construct objects
+      directly, so they can never catch a wire/type mismatch — which is how the `places` shape above
+      shipped past a green suite.
     - **Excluded**, in the same class as M10's exclusions: the admin *configuration* panels — Shard
       Visibility, Spawn Atlas and Cliloc import — alongside the hero/CMS block editor, Discord-bot
       config, uo-link config and OAuth-provider setup.

link/v3.md

@@ -350,8 +350,11 @@ Sidecar — `store.rs`: singleton `ruleset(id CHECK(id=1), rev, json, updated_t)
 `main.rs`: new arm in the board-projection match; `web.rs`: `GET /ruleset` served from the store, so
 it answers during a shard outage (`PROTOCOL_2.md` §12.2).
 
-Website — `uoLinkClient.getRuleset()`; `uoLinkSocket.backfill()` (object-shaped, so follow the
-`getPresence()` block's explicit form, not the array-only `snapshot()` helper); `shardIngest.js` →
+Website — `uoLinkClient.getRuleset()`; `uoLinkSocket.backfill()` (object-shaped, so it cannot use the
+array-only `snapshot()` helper — but it **must still go through `shardIngest.ingest()`**, as
+`ingestEach` does, rather than calling `shardState.setRuleset` directly: the two arrival orders have
+to produce the same stored frame, and a direct call quietly made backfill a second writer that
+skipped the normalization below); `shardIngest.js` →
 `shardState.setRuleset`, **not** in `LOGGED_KINDS` (it re-arrives every reconnect and `server.hello`
 already marks those); `KIND_FEATURE['world.ruleset'] = 'ruleset'`; `shard_ruleset` singleton table
 (`rev`, `expansion`, `payload JSON`, `t`); `GET /public/shard/ruleset` behind
@@ -360,6 +363,17 @@ already marks those); `KIND_FEATURE['world.ruleset'] = 'ruleset'`; `shard_rulese
 Client — NEW `routes/public/Rules.jsx` at `/site/rules`, alongside
 `/site/champs|guilds|governors|houses`; live via `useShardFeed({ filter: new Set(['world.ruleset']) })`.
 
+**The `shard` field falls back to the instance's own name.** ServUO ships `Server.cfg` with
+`Name=My Shard`, so an operator who never edited it publishes that verbatim — which is the shard
+saying *unnamed*, not naming anything, and the rules page then reads "My Shard" under a header
+carrying the real one. `shardIngest` substitutes `settings.getInstanceName()` (the admin-editable
+site title, else `BRAND_NAME` — the same resolution `getPublic().brand.name` uses, so one install
+never shows two names) when `shard` is absent, blank, or exactly the stock default, matched
+case-insensitively and trim-tolerantly but only as a **whole** value: a shard genuinely called
+*"My Shard Reborn"* has named itself and keeps it. Applied at **ingest**, not on read, because the
+ruleset is also broadcast live — the same object goes to the SSE fan-out, so a read-time
+substitution would be undone by the next reconnect's frame.
+
 ### 5.4 Risk
 
 Perf is nil (~3 KB per connect). The only real risk is publishing a secret, mitigated by the explicit
@@ -638,6 +652,15 @@ Client — NEW `routes/public/Leaderboards.jsx` at `/site/leaderboards`; a "Loya
 added to `components/CharacterSheet.jsx`, one edit serving both `PlayerCharacter.jsx` and
 `AdminCharacter.jsx`.
 
+**An unscored board still renders a row.** Most systems on a young shard have `top: []`, and a page
+of blank cards reads as broken rather than as new — so a board with no entries shows a single
+placeholder bearing the **instance's own name** with an em dash where a score goes, above the
+existing "nobody has earned points here yet" line. It is deliberately **not** shaped like an entry —
+no rank, no medal, no bar, muted — because a placeholder that looked like a real standing would be a
+fabricated one; the first real entry replaces it outright. Purely presentational: the API keeps
+sending an empty `top`, so no consumer ever receives an invented row. Web and app render it the same
+way (`Leaderboards.jsx`, `LeaderboardsScreen.kt`).
+
 ### 7.5 What the run against a real shard changed
 
 The plan above was written from reading `PointsSystem.cs`. Booting the actual shard (ServUO 57.4, a