feat(rows): the future folds — 7d tail token, hour cells, forecast guard, meter demo

CHANGELOG.md

@@ -2,6 +2,43 @@
 
 ## Unreleased
 
+## v0.30.0 — 2026-08-19 — the future folds
+
+**The 7d strip stopped drawing the future one cell at a time.** Thirty
+hollow cells after the now-marker carried one fact between them — "the
+week has room left" — and spent thirty columns saying it (thirty-one
+when the forecast painted them all `×`). The live row now keeps two
+cells after `▮` and folds the rest into `...▯5hx28`: the count of 5h
+slots to the reset, `×` red when the tail projects dry. History still
+draws in full — that is the information — and the fold only engages
+when it hides at least ten cells, so a closing week still draws to its
+edge. The `week` subcommand's wide ledger is untouched.
+
+**The 5h strip reads by the hour.** Ten half-hour cells were more
+resolution than a glance uses; five hour cells tell the same story in
+half the width, and the freed columns go to the notice beside them.
+
+**The live demo became the instrument it demos.** The GitHub Pages
+site dropped the fake macOS terminal window for a meter: a recessed
+register carrying the same frame-by-frame simulation (now with the
+folded week tail, hour cells, and a dry-forecast frame), a mono
+reading number the anatomy table cites (№ 0008), answering-pair
+anatomy rows — the left column states the reading, the right says
+what to do about it — and an odometer roll on quota digits that
+change. Narrow screens get a truthfully compacted line, the same way
+the script itself gives up the trace chip first. Stale claims fixed
+across the surfaces: 417 tests everywhere (README said 383, the site
+384, llms.txt 379), and llms.txt learned the new week-row grammar.
+
+**The forecast refuses a corrupt profile.** A weekday that claims to
+average more than the whole pool per day (`145%/day` of a 100-point
+week) can only come from a broken accountant — measured live when a
+pre-v0.29.0 build sharing the same home wrote `recent_24h: 343` and the
+walk called a 2%-used fresh window dry in 30 hours, red, on every
+render. Impossible input now earns silence, not a siren; `recent_24h`
+alone clamps to 100 (legitimately larger across a reset, but this
+window cannot lose more than everything in a day).
+
 ## v0.29.0 — 2026-08-19 — the store learns to count
 
 **Fixed: burn was counted wrong, by a factor of three.** The weekday

DESIGN.md

@@ -82,12 +82,15 @@ own signal.
 ## The ledgers (row 2)
 
 ```
-5h ▃▄▮▯▯▯▯▯▯▯ 0.6x @04:00  7d ▅▁▂ ▃▅ˍ▃▅ ▃▃▁▂▁ ▅ˍ▂▁▁ ˍˍˍ▃▅ ˍˍˍ▅ ▆▆ˍ▂▮▯▯ 0.7x @Wed 09:00
+5h ▃▄▮▯▯ 0.6x @04:00  7d ▅▁▂ ▃▅ˍ▃▅ ▃▃▁▂▁ ▮▯▯...▯5hx14 0.7x @Wed 09:00
 ```
 
-One grammar, two scales. `5h` = this window as 10 half hours; `7d` = the
-period as 34 five-hour windows, oldest left, a gap at each local midnight
-*in history only* — the run from `▮` on is contiguous. Each strip ends
+One grammar, two scales. `5h` = this window as 5 hours; `7d` = the
+period as 34 five-hour slots, oldest left, a gap at each local midnight
+*in history only* — the run from `▮` on is contiguous. History draws in
+full; the live row folds the 7d future after two kept cells into
+`...▯5hx14` — the slots left to the reset, all alike, `×` red when the
+tail projects dry (the `week` report still draws every slot). Each strip ends
 with its pace (used ÷ elapsed; dim <1x, pressure ≥1x, hidden under 15 min)
 and the reset its right edge is — axis labels, not restated badges.
 
@@ -97,6 +100,7 @@ and the reset its right edge is — axis labels, not restated badges.
 ░          unknown — no sample; never drawn as idle
 ▮          now
 ▯          ahead — the hollow of ▮
+...▯5hx14  the folded future: 14 more 5h slots to the reset (live 7d row)
 ×          pace won't cover it (7d: learned forecast, linear when cold; 5h: linear)
 ```
 

README.md

@@ -91,7 +91,7 @@ Every component earns its place:
 | User tier | Neutral white-weight (MAX bold, PRO normal, dim otherwise) — identity, never a status color. Truncated display name. |
 | Quota | Integer percentages. The 5h badge always carries its reset time while a window is live — `5h[42%@14:30]` reads "42% used, resets at 14:30" — because on a 5h horizon the reset is the number you plan the current sitting around. Wall-clock, not a countdown, on purpose: Claude Code only re-renders the statusline on activity, so a relative "@1h38m" silently decays into a lie during idle gaps, while "@14:30" stays true in a frozen frame. (The 7d badge is hybrid: day-relative `@5d` while the reset is >= 24h out — decays one day per day, mild and narrow — switching to the same wall-clock `@04:00` inside the last day, where an `@6h`/`@<1h` countdown decayed by the hour exactly when pressure keeps the suffix visible.) When a window's utilization climbs between renders, a reverse-video `+N` token appears right after the badge for ~60s: `5h[44%@14:32]+2` means "you just burned 2%". A drop (window reset) stays quiet — the fresh low number is its own signal. **7d is forecast, not leveled**: a learned per-weekday burn profile (EWMA over your own usage history) plus your recent 24h burn project whether the quota outlasts the window — your heavy Tuesday counts more than a generic average. The verdict is color alone; under pressure the badge shows when relief arrives: `7d[44%@5d]` red means "at your pace, dry days before the reset 5 days from now"; inside the last day it reads `7d[92%@04:00]` — resets at 04:00. Cold start (<14 days history) falls back to window-average pacing. Recovery color when reset is imminent. **Model-scoped weekly quota**: when the usage API carries a per-model weekly limit (`limits[]`, `kind=weekly_scoped`) for the model your session is running, it renders right after the model+context block — `fabl5[1m][12%] fb[67%]` on a Fable 5 session, `op[33%]` on Opus — because the quota is a property of the model you're running, not of the account-wide 5h/7d cluster. It's a weekly number (same reset as the 7d badge), scoped to one model. Other models' scoped quotas stay hidden: only the limit constraining *this* session is signal. Supersedes the legacy `seven_day_opus`/`seven_day_sonnet` fields, which the API now sends as null. |
 | Extra usage | Monthly spend, limit, prepaid balance. `--extra auto` shows when quota runs out. |
-| Week row | **A row of its own, under the badges**: `5h ▂▅█▃▮▯▯▯▯▯ 0.9x @23:00  7d ▅▁▂ ▃▅ˍ▃▅ …▮▯▯ 0.7x @Wed 09:00` — this sitting as ten half hours, the week as its 5h windows (day-gapped), height = what each cell burned, `▮` now, `×` where the pool runs dry, each strip ending with its pace and reset. Reconstructed from your own usage log; `auto` shows it once there is history to show. See [Week row](#week-row). |
+| Week row | **A row of its own, under the badges**: `5h ▅█▃▮▯ 0.9x @23:00  7d ▅▁▂ ▃▅ˍ▃▅ …▮▯▯...▯5hx19 0.7x @Wed 09:00` — this sitting by the hour, the week as its 5h windows (day-gapped, the far future folded to a counted `...▯5hx19`), height = what each cell burned, `▮` now, `×` where the pool runs dry, each strip ending with its pace and reset. Reconstructed from your own usage log; `auto` shows it once there is history to show. See [Week row](#week-row). |
 | Deadman | **Invisible until a switch is armed.** Surfaces [deadman](https://github.com/thevibeworks/deadman) — a dead man's switch that hands the session off when you stop responding. `[☠ armed 42m]` (dim) counts down to the auto-handoff; `[☠ warned 3m]` (yellow) means the phone warning went out; `[☠ due]` means the handoff fires imminently. Sits on the left lane next to the path — it describes this session's lifecycle, not a quota. One `command -v` when the tool is absent, one fast file read when present; nothing armed renders nothing. `--deadman off` disables it. |
 | Cache health | **Quiet until it bites.** Claude Code never re-renders an idle session, and while you work the prompt cache is always freshly ~1 TTL from expiry — so a proactive "expiring soon" isn't honestly observable, and `auto` spends no width on it. It speaks only when a rewrite actually happens: `≡!419k` the instant you resume onto a dead cache (idle longer than the TTL) or a mid-session prefix collapse — a 419k-token re-cache at ~20x the read rate (and the same burn on your 5h/7d quota on subscriptions). **Bold red past 200k** — the premium-band miss. `≡~` while a large prefix rebuilds. `--cache always` additionally keeps the freeze-safe deadline `≡@15:20` (last request + TTL; a past time in a frozen frame reads "expired at 15:20"). TTL defaults to 1h (claude.ai subscriber sessions) or 5m (API-key auth); an observed usage breakdown overrides it. The `≡` glyph (U+2261) reads as stacked cache layers — one terminal column, quiet and distinct. |
 
@@ -118,7 +118,7 @@ red, matching its Claude Code TUI color) = model family; everything else is
 | OAuth + macOS Keychain | -- | Yes |
 | Change flash on every refresh (`+.37` cost, `+9`/`-27` context, `+N` quota) | -- | Yes |
 | Model-scoped weekly quota (`fb`/`op`/`sn`) | -- | Yes |
-| Week row: the 5h window as half hours, the 7d period as 5h windows, what each cost | -- | Yes |
+| Week row: the 5h window by the hour, the 7d period as 5h windows, what each cost | -- | Yes |
 | Works behind trusted mitm proxies (NODE_EXTRA_CA_CERTS) | -- | Yes |
 | 383 bats tests + CI | -- | Yes |
 
@@ -255,7 +255,7 @@ right edge shared with line 1:
 
 ```text
 proj (main*)       +84/-14 8m $6.72 fabl5[1m][██░░42%] fb[66%] [MAX|@work] 5h[38%@23:00] 7d[39%]
-- 3x5h left · 20%/win   5h ▂▅█▃▮▯▯▯▯▯ 0.9x  7d ▅▁▂ ▃▅ˍ▃▅ ▃▃▁▂▁ ▅ˍ▂▁▁ ˍˍˍ▃▅ ˍˍˍ▅ ▆▆ˍ▂▮▯▯ 0.7x @Wed 09:00
+- 3x5h left · 20%/win   5h ▅█▃▮▯ 0.9x  7d ▅▁▂ ▃▅ˍ▃▅ ▃▃▁▂▁ ▅ˍ▂▁▁ ˍˍˍ▃▅ ˍˍˍ▅ ▆▆ˍ▂▮▯▯ 0.7x @Wed 09:00
 ```
 
 The 5h strip prints no reset: line 1's `5h[38%@23:00]` already carries
@@ -272,7 +272,7 @@ edge and the full sentence sits flush-left beneath them:
 
 ```text
 proj (main*)   fabl5[1m][██░░42%] fb[66%] [MAX|@work] 5h[38%@23:00] 7d[39%]
-   5h ▂▅█▃▮▯▯▯▯▯ 0.9x @23:00  7d ▅▁▂ ▃▅ˍ▃▅ ▃▃▁▂▁ ▅ˍ▂▁▁ ˍˍˍ▃▅ ˍˍˍ▅ ▆▆ˍ▂▮▯▯ 0.7x @Wed 09:00
+   5h ▅█▃▮▯ 0.9x @23:00  7d ▅▁▂ ▃▅ˍ▃▅ ▃▃▁▂▁ ▅ˍ▂▁▁ ˍˍˍ▃▅ ˍˍˍ▅ ▆▆ˍ▂▮▯▯ 0.7x @Wed 09:00
    - budget ~3x5h left · even 20%/win · heading ~52%
 ```
 
@@ -285,13 +285,16 @@ budget line shows with it (windows left, what even looks like, where
 you land); pressure and surplus clauses still take its place when they
 fire.
 
-- **`5h`** — this sitting: the current 5h window as 10 half-hour cells,
-  height = the 5h points that half hour added (each positive step between
-  consecutive samples credited to the half hour the later sample fell in).
-- **`7d`** — the week: the 7d period as its 5h windows (34 cells, oldest
+- **`5h`** — this sitting: the current 5h window as 5 hour cells,
+  height = the 5h points that hour added (each positive step between
+  consecutive samples credited to the hour the later sample fell in).
+- **`7d`** — the week: the 7d period as its 5h windows (34 slots, oldest
   left, the last a 3h stub), height = the 7d points that window burned,
   with a thin gap at each local midnight so days read as clusters — and a
-  day that held five windows shows it — without a ruler.
+  day that held five windows shows it — without a ruler. History draws in
+  full; the future folds: two hollow cells after `▮`, then `...▯5hx28` —
+  28 more 5h slots to the reset, all alike (`×` red when the tail projects
+  dry). The `week` subcommand's wide ledger still draws every slot.
 
 | Cell | Meaning |
 |------|---------|
@@ -301,6 +304,7 @@ fire.
 | `▮` | the cell you are in now |
 | `▯` | a cell still ahead of you — the hollow of `▮`, an empty slot waiting |
 | `×` | a cell the pool will not cover at the current pace (7d: the learned forecast's dry point, linear when untrained; 5h: linear, the same projection as the badge) |
+| `...▯5hx28` | the folded 7d future: 28 more 5h slots to the reset, one token instead of 28 hollow cells (`×` red when the tail ends dry) |
 
 Burn cells take their badge's pressure color; everything else is neutral,
 so the row never adds an alarm channel of its own. Both strips are
@@ -327,7 +331,7 @@ long one:
 
 ```text
 proj (main*)      +84/-14 8m $6.72 fabl5[1m][██░░42%] fb[91%] [MAX|@work] 5h[38%@23:00] 7d[55%]
-+ fb 91% vs 7d 55% · go op   5h ▂▅█▃▮▯▯▯▯▯ 0.9x  7d ▅▁▂ ▃▅ˍ▃▅ … ▆▆ˍ▂▮▯▯ 0.6x @Wed 09:00
++ fb 91% vs 7d 55% · go op   5h ▅█▃▮▯ 0.9x  7d ▅▁▂ ▃▅ˍ▃▅ … ▆▆ˍ▂▮▯▯ 0.6x @Wed 09:00
 + fb weekly 91% against 7d 55% · the model caps first, not the account; op sits at 33%, so run it for the bulk
 ```
 
@@ -504,7 +508,7 @@ run. Setting only `CLAUDE_CACHE_DIR` keeps the legacy single-dir behavior.
 npm exec --yes bats -- t/
 ```
 
-383 tests across `t/statusline.bats` (371 statusline + integration) and
+417 tests across `t/statusline.bats` (405 statusline + integration) and
 `t/install.bats` (12 installer). CI runs on push and PR to `main`.
 
 ## Project Structure