diff --git a/bundle-budget.json b/bundle-budget.json
index cb161380b..607b72b9a 100644
--- a/bundle-budget.json
+++ b/bundle-budget.json
@@ -1,5 +1,5 @@
 {
-  "$comment": "Client-bundle budgets, enforced by scripts/check-bundle-budget.mjs (--check) after `next build` in CI. Values are KB gzip and include every emitted lazy route and locale chunk, while route budgets include the shared rootMainFiles baseline. Re-measured at v1.37.2: / 455, /insights 443, /measurements 441, /insights/mood 453, total 3328, baseline 130, lazy catalogs 849. Every per-route ceiling is unchanged. The /insights overview crept a kilobyte over its 445 ceiling when the ECG teaser began importing its overview cap from the classification module, dragging that module's label map and helpers into the eager page; the cap now lives in its own value-only module, and the Health Score's single provenance entry is split out of the full metric-provenance map, so the overview imports the bare number and citation it renders rather than every metric's label map behind them. Both are back under 445 with room to spare. The 42 KB aggregate rise from 3286 (v1.35.0) is all locale-catalog strings: v1.36 and v1.37 added the encounter, mood-context, mood-prognosis, practitioner, workout-batch and account-sharing surfaces, and their copy lands across the six locale catalogs the whole-build sum counts independently (804 to 849 KB). The non-catalog client chunks are flat against v1.35.0, so nothing a person downloads on a given route grew beyond its own ceiling. The aggregate cap therefore moves from 3310 to 3352 (24 KB headroom, matching the previous margin); per-route ceilings, the static-catalog guard and the one-Recharts-chunk guard remain unchanged. v1.37.4's honest-error-state charge then added recoverable read-failure handling to thirteen surfaces (the shared QueryErrorCard and QueryErrorRow primitives plus an isError branch per surface) and rewrote the MCP scope note and the managed-record error copy across the six locales; the measured total rose to 3404 while every per-route ceiling stayed met (/page 457, /insights 445, /measurements 443, /insights/mood 455). The aggregate cap moves from 3352 to 3428, keeping the same margin; nothing a person downloads on a given route grew beyond its ceiling. v1.37.15's async-truth charge then added the refresh settling state with its two-locale-key copy, the index-outcome plain-words line (seven keys across the six catalogs) and the reindex-count wiring; the measured total rose to 3429 while every per-route ceiling stayed met (/page 456, /insights 445, /measurements 442, /insights/mood 455). The aggregate cap moves from 3428 to 3452, keeping the same margin; nothing a person downloads on a given route grew beyond its ceiling. v1.37.27 found /insights sitting a few hundred bytes under 445 on main and a few hundred over it on the release branch — the page had been at its ceiling for three releases and the release only had to breathe on it. The weight it was carrying was not its own: the authenticated shell imported one destination predicate from the `@/lib/record-settings` barrel, and the barrel also re-exports the managed settings-patch schemas, so Zod, the module registry and three validation modules rode into the eager bundle of every authenticated route for a function that reads a URL slug. The shell imports the `classification` leaf now. Measured across all 585 client-rendered routes: none grew by a byte, 100 fell, by 4.7 KB at the median and 8.1 KB at the most; /insights 444 to 437, /measurements 442 to 435, / 456 to 450, /insights/mood 455 to 450. The whole-build sum moved the other way, 3451 to 3476, and that direction is the metric rather than the change: the aggregate counts each emitted chunk once, so a graph that leaves the chunk group every route shares gets counted again wherever it lands — here inside the deferred settings chunks that actually use it. Bytes on disk rose 25 KB; bytes a reader downloads only ever fell. The aggregate cap moves from 3452 to 3500, the same 24 KB headroom as the four re-measurements before it. Per-route ceilings keep their old values rather than tightening to the new floor: the CI runner measures each route about 1.5 KB heavier than a local arm64 build does, and a ceiling set from a local number would fail on a build that had not regressed. The static-catalog guard and the one-Recharts-chunk guard remain unchanged. Raised to 3700 at v1.37.32 for the seventh interface language (Korean): the lazy message catalogs went 6->7 chunks (1026 KB gz measured), pushing the aggregate to 3631 against the old 3500. The raise is entirely the new catalog; no route budget moved, and the four eager route ceilings are unchanged and still passing. Most of the aggregate is the dynamic locale-switch and fallback path that a first paint never loads (see issue #253), so this ceiling guards bytes almost no visitor requests; 3700 leaves ~69 KB for ordinary drift until that double-count is cleaned up. v1.38.14 re-measures the four eager routes and changes what the number means. The ceilings had been re-stated but never re-based since v1.37.4, so /insights and /measurements sat within a kilobyte of theirs and / and /insights/mood within one of theirs; e2e run 34043772954 failed on `/insights/page: 445 KB gz exceeds the 445 KB budget` with the other two at 459 of 460, and none of it was a regression anyone had made. A ceiling a route is already touching does not measure the route, it measures whichever dependency bump landed that week. The file states the measured value per route now, plus one drift allowance over it: measured on a local arm64 build at d48bbff01 as / 447, /insights 434, /measurements 433, /insights/mood 447, with a 20 KB allowance, so the effective ceilings are 467 / 454 / 453 / 467 — a raise for all four. The allowance covers the runner delta (the CI runner measures each route a little heavier than a local arm64 build) and ordinary week-to-week drift, while the step change the gate exists for — a statically imported message catalog, a duplicated chart library — is several times larger and still fails. Every run now prints the delta against the baseline, so a route that has crept 15 KB is legible in the log long before it is red; the flat ceiling could only ever say pass or fail. Re-base by re-measuring on a build with nothing else in it and writing the new numbers here with the reason, exactly as a raised ceiling was written before. Raised to 3724 at v1.39.4 on a measured delta, not a feature. The release branch alone measured 3699 KB gz (3698.9) against the 3700 cap, so the ordinary drift room the v1.37.32 raise left had been used up by the releases since; the review fixes on top of it added 1.3 KB gz across the whole build (3700.2), almost all of it new copy in the seven lazy locale catalogs (the total and reused-table labels, the step and chip wording) plus the result-panel states. The four eager routes measured the same on both builds (/ 458, /insights/mood 456, /insights 445, /measurements 442), all inside their ceilings. The aggregate cap moves from 3700 to 3724, the same 24 KB headroom as the re-measurements before it. Raised to 3749 at v1.39.9 on a measured delta. Main measured 3718 KB gz after v1.39.7 and v1.39.8 (unit-aware copy, the phone layouts, the time-zone work and the account-settings restore labels, almost all of it new strings in the seven lazy locale catalogs), leaving 6 KB of the v1.39.4 headroom; the manual workout entry (#1085: the form, its sheet, the delete action and their copy in seven locales) adds about 7 KB and measured 3725. The four eager routes stay inside their drift allowances (/ +12.5, /insights +12.1, /insights/mood +10.2 KB of 20 on main). The aggregate cap moves from 3724 to 3749, the same 24 KB headroom as the re-measurements before it. Raised to 3790 at v1.40.0 on a measured delta. With the background document reads on main (#1103: the run polling hook, the still-reading states and their copy in seven locales), the key backup step and the plain-http sign-in notice (#1104) and the v1.40 fixes (#1106: the duration reader, the marker unit lock, the chart window) each branch measured 3752 KB gz on its own against the 3749 cap; together they add a few KB more. The four eager routes stay inside their drift allowances (/ +18.4, /insights +16.5, /measurements +14.9, /insights/mood +14.7 KB of 20). The aggregate cap moves from 3749 to 3790, about 24 KB above the combined build, the same headroom as the re-measurements before it. Re-based at v1.40.0 when the quick-entry forms began loading on demand. Custom symptoms added a fifth quick-entry form, and with all five forms bundled eagerly every one of the four eager routes measured over its ceiling (/ 472, /insights 457, /insights/mood 468, /measurements 454 KB gz, each 21 to 25 KB over the baseline). Loading the forms on demand brings them to / 429, /insights 368, /insights/mood 379, /measurements 411: 43 to 90 KB less on first load. The four route baselines move down to those measured values, so the gain is protected by the same 20 KB allowance rather than spent. The aggregate moves the other way, 3774 to 4077, for the reason described at v1.37.27: the aggregate counts each emitted chunk once per route group, and on-demand form chunks land in every route group that can open a quick-entry sheet. Bundling the five forms into one shared on-demand module measured the same (4078), so the split is not the cause. The aggregate cap moves from 3790 to 4100. Bytes on disk rose; bytes a reader downloads fell on every eager route. Raised to 4155 at v1.40.2 on a measured delta. The Coach redesign (the right conversation panel, its row menu and the quick-settings overlay, all loaded on demand), the answer action row, the dashboard's Today overview, the document procedure filter and their copy in seven locales took the whole-build sum from 4080 to 4129 KB gz; most of it is new strings in the seven lazy locale catalogs and the on-demand Coach chunks, which a first paint does not load. The four eager routes stay well inside their drift allowances (/ +1.9, /measurements +1.9, /insights/mood +1.4, /insights +1.0 KB of 20). The aggregate cap moves from 4100 to 4155, about 26 KB above the build, the same headroom as before; per-route baselines are unchanged. Raised to 4182 at v1.42.0 on a measured delta. Main measured 4146 KB gz (lazy catalogs 1177, other chunks 2969); the release branch's shared contract change measured 4158 (lazy catalogs 1188, other chunks 2969): the whole 12 KB is new strings in the seven lazy locale catalogs (the Health Connect import card, air quality and pollen, the managed-profile handover and claim, own vaccines, the lab conversion note, the per-type Apple Health list), and no eager route moved (/ +3.6, /measurements +1.5, /insights/mood +1.1, /insights +0.7 KB of 20). The aggregate cap moves from 4155 to 4182, about 24 KB above the build, the same headroom as before; per-route baselines are unchanged. Re-based at v1.42.0 when the settings sections and Zod left the shared bundles. The settings section route imported all twenty-one sections statically; each now loads on its own, and /settings/[section] measured 611 KB gz before and 403 after. The account-access parse that runs on every boot was the shell's only Zod import, about 63 KB gz on every route; it is plain TypeScript now, held to the published schema by an equivalence test, and the layout entry fell from 199 to 136 KB gz. The canonical unit table moved out of the measurement validation module, which took Zod off the routes whose only reason for it was the display transform: /insights 369 to 304, /medications 428 to 365, /labs 400 to 337. Across the 627 client routes 62 fell, none grew by more than 2 KB, and the per-route sum fell by 1844 KB gz. The /insights baseline moves down to 304 and /settings/[section] joins the watched routes at 403, so both gains are held by the same 20 KB allowance rather than spent. The aggregate moves the other way, 4167 to 4311, for the reason described at v1.37.27 and v1.40.0: the aggregate counts each emitted chunk once, and modules that used to sit in one shared chunk are now emitted into each route group that still needs them (the measurement validation module and the unit table into about twenty-five each, the settings sections into their own chunks). Bytes on disk rose; bytes a reader downloads fell on every route that moved. The aggregate cap moves from 4182 to 4335, about 24 KB above the build, the same headroom as before. Raised to 4400 at v1.42.0 on a measured delta. The release branch after its tickets merge measured 4365 KB gz (lazy catalogs 1206, other chunks 3159), already 30 KB over the 4335 cap; the day view and timeline contract adds 10 more (catalogs 1215, other 3160), almost all of it the base copy for the day, the timeline and life events in seven locales. The aggregate cap moves from 4335 to 4400, about 25 KB above the build, the same headroom as before; per-route baselines are unchanged.",
+  "$comment": "Client-bundle budgets, enforced by scripts/check-bundle-budget.mjs (--check) after `next build` in CI. Values are KB gzip and include every emitted lazy route and locale chunk, while route budgets include the shared rootMainFiles baseline. Re-measured at v1.37.2: / 455, /insights 443, /measurements 441, /insights/mood 453, total 3328, baseline 130, lazy catalogs 849. Every per-route ceiling is unchanged. The /insights overview crept a kilobyte over its 445 ceiling when the ECG teaser began importing its overview cap from the classification module, dragging that module's label map and helpers into the eager page; the cap now lives in its own value-only module, and the Health Score's single provenance entry is split out of the full metric-provenance map, so the overview imports the bare number and citation it renders rather than every metric's label map behind them. Both are back under 445 with room to spare. The 42 KB aggregate rise from 3286 (v1.35.0) is all locale-catalog strings: v1.36 and v1.37 added the encounter, mood-context, mood-prognosis, practitioner, workout-batch and account-sharing surfaces, and their copy lands across the six locale catalogs the whole-build sum counts independently (804 to 849 KB). The non-catalog client chunks are flat against v1.35.0, so nothing a person downloads on a given route grew beyond its own ceiling. The aggregate cap therefore moves from 3310 to 3352 (24 KB headroom, matching the previous margin); per-route ceilings, the static-catalog guard and the one-Recharts-chunk guard remain unchanged. v1.37.4's honest-error-state charge then added recoverable read-failure handling to thirteen surfaces (the shared QueryErrorCard and QueryErrorRow primitives plus an isError branch per surface) and rewrote the MCP scope note and the managed-record error copy across the six locales; the measured total rose to 3404 while every per-route ceiling stayed met (/page 457, /insights 445, /measurements 443, /insights/mood 455). The aggregate cap moves from 3352 to 3428, keeping the same margin; nothing a person downloads on a given route grew beyond its ceiling. v1.37.15's async-truth charge then added the refresh settling state with its two-locale-key copy, the index-outcome plain-words line (seven keys across the six catalogs) and the reindex-count wiring; the measured total rose to 3429 while every per-route ceiling stayed met (/page 456, /insights 445, /measurements 442, /insights/mood 455). The aggregate cap moves from 3428 to 3452, keeping the same margin; nothing a person downloads on a given route grew beyond its ceiling. v1.37.27 found /insights sitting a few hundred bytes under 445 on main and a few hundred over it on the release branch — the page had been at its ceiling for three releases and the release only had to breathe on it. The weight it was carrying was not its own: the authenticated shell imported one destination predicate from the `@/lib/record-settings` barrel, and the barrel also re-exports the managed settings-patch schemas, so Zod, the module registry and three validation modules rode into the eager bundle of every authenticated route for a function that reads a URL slug. The shell imports the `classification` leaf now. Measured across all 585 client-rendered routes: none grew by a byte, 100 fell, by 4.7 KB at the median and 8.1 KB at the most; /insights 444 to 437, /measurements 442 to 435, / 456 to 450, /insights/mood 455 to 450. The whole-build sum moved the other way, 3451 to 3476, and that direction is the metric rather than the change: the aggregate counts each emitted chunk once, so a graph that leaves the chunk group every route shares gets counted again wherever it lands — here inside the deferred settings chunks that actually use it. Bytes on disk rose 25 KB; bytes a reader downloads only ever fell. The aggregate cap moves from 3452 to 3500, the same 24 KB headroom as the four re-measurements before it. Per-route ceilings keep their old values rather than tightening to the new floor: the CI runner measures each route about 1.5 KB heavier than a local arm64 build does, and a ceiling set from a local number would fail on a build that had not regressed. The static-catalog guard and the one-Recharts-chunk guard remain unchanged. Raised to 3700 at v1.37.32 for the seventh interface language (Korean): the lazy message catalogs went 6->7 chunks (1026 KB gz measured), pushing the aggregate to 3631 against the old 3500. The raise is entirely the new catalog; no route budget moved, and the four eager route ceilings are unchanged and still passing. Most of the aggregate is the dynamic locale-switch and fallback path that a first paint never loads (see issue #253), so this ceiling guards bytes almost no visitor requests; 3700 leaves ~69 KB for ordinary drift until that double-count is cleaned up. v1.38.14 re-measures the four eager routes and changes what the number means. The ceilings had been re-stated but never re-based since v1.37.4, so /insights and /measurements sat within a kilobyte of theirs and / and /insights/mood within one of theirs; e2e run 34043772954 failed on `/insights/page: 445 KB gz exceeds the 445 KB budget` with the other two at 459 of 460, and none of it was a regression anyone had made. A ceiling a route is already touching does not measure the route, it measures whichever dependency bump landed that week. The file states the measured value per route now, plus one drift allowance over it: measured on a local arm64 build at d48bbff01 as / 447, /insights 434, /measurements 433, /insights/mood 447, with a 20 KB allowance, so the effective ceilings are 467 / 454 / 453 / 467 — a raise for all four. The allowance covers the runner delta (the CI runner measures each route a little heavier than a local arm64 build) and ordinary week-to-week drift, while the step change the gate exists for — a statically imported message catalog, a duplicated chart library — is several times larger and still fails. Every run now prints the delta against the baseline, so a route that has crept 15 KB is legible in the log long before it is red; the flat ceiling could only ever say pass or fail. Re-base by re-measuring on a build with nothing else in it and writing the new numbers here with the reason, exactly as a raised ceiling was written before. Raised to 3724 at v1.39.4 on a measured delta, not a feature. The release branch alone measured 3699 KB gz (3698.9) against the 3700 cap, so the ordinary drift room the v1.37.32 raise left had been used up by the releases since; the review fixes on top of it added 1.3 KB gz across the whole build (3700.2), almost all of it new copy in the seven lazy locale catalogs (the total and reused-table labels, the step and chip wording) plus the result-panel states. The four eager routes measured the same on both builds (/ 458, /insights/mood 456, /insights 445, /measurements 442), all inside their ceilings. The aggregate cap moves from 3700 to 3724, the same 24 KB headroom as the re-measurements before it. Raised to 3749 at v1.39.9 on a measured delta. Main measured 3718 KB gz after v1.39.7 and v1.39.8 (unit-aware copy, the phone layouts, the time-zone work and the account-settings restore labels, almost all of it new strings in the seven lazy locale catalogs), leaving 6 KB of the v1.39.4 headroom; the manual workout entry (#1085: the form, its sheet, the delete action and their copy in seven locales) adds about 7 KB and measured 3725. The four eager routes stay inside their drift allowances (/ +12.5, /insights +12.1, /insights/mood +10.2 KB of 20 on main). The aggregate cap moves from 3724 to 3749, the same 24 KB headroom as the re-measurements before it. Raised to 3790 at v1.40.0 on a measured delta. With the background document reads on main (#1103: the run polling hook, the still-reading states and their copy in seven locales), the key backup step and the plain-http sign-in notice (#1104) and the v1.40 fixes (#1106: the duration reader, the marker unit lock, the chart window) each branch measured 3752 KB gz on its own against the 3749 cap; together they add a few KB more. The four eager routes stay inside their drift allowances (/ +18.4, /insights +16.5, /measurements +14.9, /insights/mood +14.7 KB of 20). The aggregate cap moves from 3749 to 3790, about 24 KB above the combined build, the same headroom as the re-measurements before it. Re-based at v1.40.0 when the quick-entry forms began loading on demand. Custom symptoms added a fifth quick-entry form, and with all five forms bundled eagerly every one of the four eager routes measured over its ceiling (/ 472, /insights 457, /insights/mood 468, /measurements 454 KB gz, each 21 to 25 KB over the baseline). Loading the forms on demand brings them to / 429, /insights 368, /insights/mood 379, /measurements 411: 43 to 90 KB less on first load. The four route baselines move down to those measured values, so the gain is protected by the same 20 KB allowance rather than spent. The aggregate moves the other way, 3774 to 4077, for the reason described at v1.37.27: the aggregate counts each emitted chunk once per route group, and on-demand form chunks land in every route group that can open a quick-entry sheet. Bundling the five forms into one shared on-demand module measured the same (4078), so the split is not the cause. The aggregate cap moves from 3790 to 4100. Bytes on disk rose; bytes a reader downloads fell on every eager route. Raised to 4155 at v1.40.2 on a measured delta. The Coach redesign (the right conversation panel, its row menu and the quick-settings overlay, all loaded on demand), the answer action row, the dashboard's Today overview, the document procedure filter and their copy in seven locales took the whole-build sum from 4080 to 4129 KB gz; most of it is new strings in the seven lazy locale catalogs and the on-demand Coach chunks, which a first paint does not load. The four eager routes stay well inside their drift allowances (/ +1.9, /measurements +1.9, /insights/mood +1.4, /insights +1.0 KB of 20). The aggregate cap moves from 4100 to 4155, about 26 KB above the build, the same headroom as before; per-route baselines are unchanged. Raised to 4182 at v1.42.0 on a measured delta. Main measured 4146 KB gz (lazy catalogs 1177, other chunks 2969); the release branch's shared contract change measured 4158 (lazy catalogs 1188, other chunks 2969): the whole 12 KB is new strings in the seven lazy locale catalogs (the Health Connect import card, air quality and pollen, the managed-profile handover and claim, own vaccines, the lab conversion note, the per-type Apple Health list), and no eager route moved (/ +3.6, /measurements +1.5, /insights/mood +1.1, /insights +0.7 KB of 20). The aggregate cap moves from 4155 to 4182, about 24 KB above the build, the same headroom as before; per-route baselines are unchanged. Re-based at v1.42.0 when the settings sections and Zod left the shared bundles. The settings section route imported all twenty-one sections statically; each now loads on its own, and /settings/[section] measured 611 KB gz before and 403 after. The account-access parse that runs on every boot was the shell's only Zod import, about 63 KB gz on every route; it is plain TypeScript now, held to the published schema by an equivalence test, and the layout entry fell from 199 to 136 KB gz. The canonical unit table moved out of the measurement validation module, which took Zod off the routes whose only reason for it was the display transform: /insights 369 to 304, /medications 428 to 365, /labs 400 to 337. Across the 627 client routes 62 fell, none grew by more than 2 KB, and the per-route sum fell by 1844 KB gz. The /insights baseline moves down to 304 and /settings/[section] joins the watched routes at 403, so both gains are held by the same 20 KB allowance rather than spent. The aggregate moves the other way, 4167 to 4311, for the reason described at v1.37.27 and v1.40.0: the aggregate counts each emitted chunk once, and modules that used to sit in one shared chunk are now emitted into each route group that still needs them (the measurement validation module and the unit table into about twenty-five each, the settings sections into their own chunks). Bytes on disk rose; bytes a reader downloads fell on every route that moved. The aggregate cap moves from 4182 to 4335, about 24 KB above the build, the same headroom as before. Raised to 4400 at v1.42.0 on a measured delta. The release branch after its tickets merge measured 4365 KB gz (lazy catalogs 1206, other chunks 3159), already 30 KB over the 4335 cap; the day view and timeline contract adds 10 more (catalogs 1215, other 3160), almost all of it the base copy for the day, the timeline and life events in seven locales. The aggregate cap moves from 4335 to 4400, about 25 KB above the build, the same headroom as before; per-route baselines are unchanged. Raised to 4470 at v1.42.0 on a measured delta. The release branch with the day view, the timeline and their server measured 4386 KB gz (lazy catalogs 1228, other chunks 3158); giving every chart drawn in days the same doors to its day (one shared hook for the click, the touch tooltip's button, the open day's line, the row of day dots and the caption, a shared calendar adapter, a data table under each newly linked chart) measured 4446 (catalogs 1228, other chunks 3218). The 60 KB is the aggregate's double count again: the hook, the data table and the day link are emitted into each chart's own chunk group rather than once. The eager routes moved by a kilobyte or a few (/ +1.6, /measurements +1.7, /settings/[section] +0.8, /insights/mood +4.6, /insights +1.5 KB) and stay inside their allowances. The aggregate cap moves from 4400 to 4470, about 24 KB above the build, the same headroom as before; per-route baselines are unchanged.",
   "routeBaselineKbGz": {
     "/page": 429,
     "/insights/page": 304,
@@ -8,6 +8,6 @@
     "/settings/[section]/page": 403
   },
   "routeDriftAllowanceKbGz": 20,
-  "totalClientKbGz": 4400,
+  "totalClientKbGz": 4470,
   "maxRechartsChunks": 1
 }