re: first audit of INDEX against the pages it links -- 3 of 8 rows were stale

INDEX is read every iteration and is the first thing a new reader meets, and I had
admitted three times to never auditing it. Found by accident: I was about to spend
a boot measuring whether A skips a movie because INDEX said 🟡, when movie-binding.md
had it settled since 2026-08-28 with a three-boot baseline and a delivery counter,
and HANDOFF carried it correctly. The staleness was in the index alone.

Three stale rows fixed: movie skippability, SE audio extractability, and B leaving
the main menu. Five other hits read and left alone -- index and page were talking
about different clauses.

One of the three should have been caught by check_refuted.py: REFUTED holds the
same dead claim with a different second clause, and the register matches exact
wording. Its docstring documents that weakness; this is the first live instance.

Also flags a within-page contradiction not fixed here: menu-audio-cues.md line 81
still heads a section saying SE audio is not extractable, which its own line 189
refutes.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Wuu56cE8vJGTBtn1ppsk8v

docs/re/INDEX.md

@@ -153,11 +153,11 @@ files, which is how the same ground got covered twice.
 | [`title-crash-stl-tree.md`](title-crash-stl-tree.md) | The title-screen crash is an STL `map`/`set` erase on a bad iterator |  ✅ CONFIRMED — the guest throws std::out_of_range from an STL |
 | [`ui-paint-order-third-permutation.md`](ui-paint-order-third-permutation.md) | A third measured paint order — tool built and validated, screen not reached |  ✅ the reader works and is CONFIRMED against both previously |
 | [`ui-quad-class-foothold.md`](ui-quad-class-foothold.md) | The guest's UI quad class — a foothold found from the capture's vertex layout |  🟡 PROBABLE for the identification below (it is a static read, but |
-| [`menu-navigation-semantics.md`](menu-navigation-semantics.md) | The title menu — how it moves, and where each button goes |  ✅ measured: wraps both ends, Ⓑ restores focus, ⬅➡ inert; all 5 destinations driven. 🟡 GamePart id is a name match. 🟡 **Ⓑ leaving the MAIN menu downgraded 2026-08-29** — uncited, and the main menu is the only screen whose footer omits Ⓑ (0 glyph px in frame vs 514/518 elsewhere). ✅ **MISSION SELECT's stuck cursor was a LOCKED stage list** — labels have three brightnesses, locked 104 / unfocused 183 / focused 254 |
+| [`menu-navigation-semantics.md`](menu-navigation-semantics.md) | The title menu — how it moves, and where each button goes |  ✅ measured: wraps both ends, Ⓑ restores focus, ⬅➡ inert; all 5 destinations driven. 🟡 GamePart id is a name match. ✅ **Ⓑ leaving the MAIN menu re-measured 2026-08-30** — goes to the title, delivery-confirmed, ≤ 0.4 s, no loading screen, and the plate is re-drawn ~7 s later; ~~downgraded 2026-08-29 as uncited~~, and the main menu is the only screen whose footer omits Ⓑ (0 glyph px in frame vs 514/518 elsewhere). ✅ **MISSION SELECT's stuck cursor was a LOCKED stage list** — labels have three brightnesses, locked 104 / unfocused 183 / focused 254 |
 | [`screen-transitions.md`](screen-transitions.md) | Between two screens — a fade through black, and where its timing lives |  ✅ the fade quad's keyframe group is decoded (disc-wide: per-pak all-or-nothing; `GP_TITLE` = the 6 screens, not the 6 overlays); the ~0.4 s fade-OUT is measured, not on the disc |
-| [`menu-audio-cues.md`](menu-audio-cues.md) | Menu audio — the event vocabulary is on the disc, the binding is not |  ✅ `SE_UI_*` cue names/ids decoded and `BANK_SE`→`Static.slb` (0/322 in FILES); 🟡 event binding is a name match; ❔ `Static.slb` has no wave boundaries, so SE audio is not extractable |
+| [`menu-audio-cues.md`](menu-audio-cues.md) | Menu audio — the event vocabulary is on the disc, the binding is not |  ✅ `SE_UI_*` cue names/ids decoded and `BANK_SE`→`Static.slb` (0/322 in FILES); 🟡 event binding is a name match; ✅ **SE audio IS extractable** — the waves are located in `Static.slb` by playing them (move `0x1ec0`, confirm `0x5d6c0`, back `0x0ec0`); ~~❔ `Static.slb` has no wave boundaries, so SE audio is not extractable~~ is **refuted** and was still asserted here |
 | [`boot-config-and-gamepart-registry.md`](boot-config-and-gamepart-registry.md) | What the game reads at boot — `config.ini`, and which GameParts exist |  ✅ `config.ini` selects the language (the disc's only config); ❔ its `[SYSTEM]` is empty so the boot order is not in config; 🟡 24/29 ids bind to a class, `GP_ADVERTISE_DEMO` is never registered |
-| [`movie-binding.md`](movie-binding.md) | Which movie plays where — boot intro, attract loop, new-game intro |  ✅ decoded from the movie manifest (`ADVERTISE_MOVIE`→`ADV.wmv`, `MS00A`→`S00A.wmv`); attract identity confirmed independently by frame matching; 🟡 skippability unsettled |
+| [`movie-binding.md`](movie-binding.md) | Which movie plays where — boot intro, attract loop, new-game intro |  ✅ decoded from the movie manifest (`ADVERTISE_MOVIE`→`ADV.wmv`, `MS00A`→`S00A.wmv`); attract identity confirmed independently by frame matching; ✅ **skippability settled 2026-08-28: one Ⓐ skips a movie** — title at 57 s against a 193/196/193 s three-boot baseline, the press proved singular by Canary's own delivery counter (3→4), and the skipped-to title fully functional. ~~🟡 skippability unsettled~~ was stale here while the page and `HANDOFF.md` both had it answered |
 | [`ready-room-probe.md`](ready-room-probe.md) | S1 — the Ready Room probe: no-go, and not for the reason expected |  ✅ it is 2D and enumerates (60 builds), but the pak is briefing/tactical-map content; and `kind == 0x3002` finds 0 buttons there |
 | [`ui-title-build-map.md`](ui-title-build-map.md) | Which `GP_TITLE` build is which screen state |  ✅ CONFIRMED for title / `PRESS Ⓐ` / main menu / `EXTRAS` against live captures; the archive is 8 screens × EN/JP, and "6/8/9 are submenus" is withdrawn  ✅ **2026-08-29: the two "unidentified `DELTASABER` plates" are the LOADING screen** — builds 0/1 the plain variant, 10/11 the dressed one, decoded from their `pgloading_*` element names, and the executable (`sub_821C4EB0`, bytes checked in the image) names exactly five title-side screens: `TITLE_SCREEN`, `BUTTON`, `TITLE_MENU`, `LOADING`, `LOADING2`. 🟡 which loading bundle takes which of the two names is undecided. 🟡 the English member of a pair is the one in the first half of `GP_TITLE.p00` — 8/8 structurally, 3/3 where a capture can check it. |
 | [`ui-title-paint-order-capture.md`](ui-title-paint-order-capture.md) | The title screen's paint order, measured from the guest's draw submissions |  ✅ CONFIRMED — the order in which the running game paints the title |

docs/re/data/index-vs-pages-audit.txt

@@ -0,0 +1,58 @@
+# Does INDEX.md agree with the pages it links? -- 2026-08-30. FIRST AUDIT.
+#
+# INDEX.md is read every iteration by both agents and is the first thing a new
+# reader meets. sylpheed-port's phrase is the right one: an index is an
+# AMPLIFIER -- a status wrong there is wrong everywhere it is quoted from. It is
+# also a file I have admitted three times I had never audited.
+#
+# FOUND BY ACCIDENT, which is why the audit happened: I was about to spend a boot
+# measuring whether Ⓐ skips a movie, because the port's BLOCKED.md lists it 🟡 and
+# so did INDEX. `movie-binding.md` has had it ✅ SETTLED since 2026-08-28 -- with a
+# three-boot baseline and a delivery counter -- and HANDOFF carries it correctly.
+# The staleness was in the index alone, and it nearly cost a run.
+#
+# TOOL: tools/re-capture/index_vs_pages.py. It prints a LISTING, NOT A VERDICT:
+# 8 rows where the index is 🟡/❔ while the linked page has a "✅ settled/decoded"
+# heading. A row can legitimately be unsure about one clause while its page is
+# sure about another, so every hit is a prompt to read. I read all 8.
+#
+# ✅ 5 ARE LEGITIMATE -- index and page are talking about different clauses:
+#   77  flight-speed-law        "🟡 for the unit scale"           -- still true
+#   127 isl-builtins            "🟡 28 ..."                        -- still true
+#   132 mcol-collision          "🟡 OPENED ..."                    -- a status word
+#   159 boot-config-registry    "❔ [SYSTEM] is empty ..."          -- still true
+#   162 ui-title-build-map      "🟡 which loading bundle takes which name" -- open
+#
+# 🔴 3 WERE STALE, all three contradicted by the page they link:
+#
+#   160 movie-binding      "🟡 skippability unsettled"
+#       -> page: ✅ settled 2026-08-28. One Ⓐ skips: title at 57 s against a
+#          193/196/193 s three-boot baseline, press proved singular by Canary's
+#          own delivery counter going 3→4, skipped-to title fully functional.
+#
+#   158 menu-audio-cues    "❔ Static.slb has no wave boundaries, so SE audio is
+#                           not extractable"
+#       -> page line 189: "both waves are located in Static.slb. The port can
+#          have the audio." The waves are at move 0x1ec0, confirm 0x5d6c0,
+#          back 0x0ec0.
+#       ⚠️ AND THE PAGE CONTRADICTS ITSELF: its line 81 still heads a section
+#          "❔ And a new negative: an individual SE's audio is not extractable
+#          yet", which its own line 189 refutes. Not fixed here -- flagged.
+#
+#   156 menu-navigation     "🟡 Ⓑ leaving the MAIN menu downgraded 2026-08-29 --
+#                            uncited"
+#       -> page line 27: ✅ measured 2026-08-30, delivery-confirmed, ≤0.4 s, no
+#          loading screen, plate re-drawn ~7 s later, three captures cited.
+#
+# 🔴 AND ONE OF THE THREE SHOULD HAVE BEEN CAUGHT BY MY OWN REGISTER.
+# REFUTED.md holds "`Static.slb` has no wave boundaries, so its layout is
+# unknown". INDEX said "...so SE audio is not extractable". Same dead claim,
+# different second clause -- and check_refuted.py matches EXACT wording, so it
+# saw nothing. That weakness is documented in the tool's own docstring ("a clean
+# run means no VERBATIM revival"); this is the first LIVE instance of it, and it
+# survived in the index for days.
+#
+# ⚠️ REACH: this compares an index row's emoji against the presence of a ✅
+# heading on the linked page. It cannot see a row that is confidently WRONG
+# (no 🟡 to trip on), a row whose page has no status heading, or anything in
+# HANDOFF.md or BLOCKED.md. 3 of 8 hits were real; the other 5 cost one read each.