# Are MISSION's gates MET and RECORDED, both halves? 2026-08-30.
#
# sylpheed-port found P0 complete-but-unindexed -- the work existed, the artifact
# existed, the gate record did not -- and named it as the argued-versus-indexed
# split one level up from the refutation register. This is the same audit on the
# Decoder's objective.
#
# MISSION's gate: "a written docs/re/ result with the evidence, and REFERENCE DATA
# COMMITTED ALONGSIDE IT". Two halves, and all ten questions read ✅ answered.
#
# HALF ONE -- every Q cites a docs/re/ result. All ten do:
#   Q1  ui-keyframe-time-unit.md + ui-keyframe-record-layout.md
#   Q2  ui-title-build-map.md
#   Q3  structures/ui-paint-order-key.md + ui-paint-order-derived-check.md
#   Q4  menu-navigation-semantics.md          Q5  menu-navigation-semantics.md
#   Q6  boot-config-and-gamepart-registry.md  Q7  screen-transitions.md
#   Q8  menu-audio-cues.md                    Q9  movie-binding.md
#   Q10 structures/bgm-two-stems.md + slb-bank-header-not-a-wave.md
#
# HALF TWO -- reference data committed alongside. Every data/ and captures/ path
# cited by those pages was resolved against the tree:
#
#   ui-keyframe-time-unit.md              3 cited, 0 missing
#   ui-title-build-map.md                11 cited, 0 missing
#   structures/ui-paint-order-key.md     11 cited, 0 missing
#   menu-navigation-semantics.md         11 cited, 0 missing
#   boot-config-and-gamepart-registry.md  3 cited, 0 missing
#   screen-transitions.md                 4 cited, 0 missing
#   menu-audio-cues.md                    2 cited, 0 missing
#   movie-binding.md                      2 cited, 0 missing
#   structures/bgm-two-stems.md           1 cited, 0 missing
#
# And spot-checked for substance rather than mere existence, since the gate's
# PURPOSE is that the port can work without a disc:
#   splash-ramp-check.txt          1 197 B   15 numeric lines
#   fade-envelope-menu-to-title    2 373 B   34
#   fade-four-transitions.txt     20 737 B  198
#   se-cue-runtime-offsets.txt     2 258 B   17
#   se-ui-cues.txt                 7 134 B  324
#   bgm-wave-census.txt            3 369 B   33
#
# ✅ CLEAN. Both halves present for all ten, and unlike the port's P0 they were
# also INDEXED -- HANDOFF's status table cites the page, the page cites the data.
#
# ⚠️ REACH, because a clean audit is only worth its checks. This tests that CITED
# files EXIST and are non-trivial. It does NOT test that the data supports the
# claim, and it cannot see data a page SHOULD have cited and did not. A page
# citing nothing would have passed as "0 missing"; none did, but the check would
# not have caught it. Existence and substance, not sufficiency.

################################################################################
# THE ABSENCE CHECK -- the half the audit above could not see. 2026-08-30.
#
# sylpheed-port's sufficiency audit found the thing that "passes every check by
# being absent": an authored value carrying no `why` at all. The analogue here is
# a page that cites NO reference data, which my gate audit above would score as
# "0 missing" and pass. So: of every page with a measured/decoded/CONFIRMED
# status, which cites no data/ or captures/ path?
#
#   42 pages with such a status
#    3 citing no data/ or captures/ path
#
# ⚠️ INSPECTED BEFORE PUBLISHING, per the port's rule that a first count from a new
# detector is a measurement of the detector. All three are FALSE POSITIVES, and
# each was verified rather than waved through:
#
#   structures/slb-bank-header-not-a-wave.md -- cites tests/slb_leading_segment_
#     disc.rs, and that file EXISTS in crates/sylpheed-formats/tests/. Its evidence
#     is a disc-wide check over 9 519 sound.pak entries plus regression tests.
#   structures/ui-screen-runtime.md -- 26 rows of inline evidence tables, live
#     guest-memory reads matched field by field against the file.
#   five-screens-acceptance.md -- a consolidation page; its evidence is the six
#     pages it links and the numbers it tabulates from them.
#
# ✅ 3 -> 0 real.
#
# 📌 AND THE REAL FINDING IS ABOUT THE EARLIER AUDIT. This corpus carries evidence
# in at least THREE forms -- committed data files, inline tables, and committed
# disc tests -- and both checks look for exactly one of them. "48 citations, 0
# missing" above is therefore a statement about the data-file form, not about
# whether the gates are evidenced. The gates ARE evidenced; the audit was narrower
# than its wording suggested.
