P23-03: 生活の観察

2026-10-16 17:15:06 +09:00 · 2026-10-16 14:56:38 +09:00 · 2026-10-16 14:41:54 +09:00 · 2026-10-16 14:24:10 +09:00 · 2026-10-16 14:17:14 +09:00 · 2026-10-15 18:30:42 +09:00
22 changed files with 6139 additions and 114 deletions
@@ -3707,7 +3707,7 @@
      "template": "P23 life observation: <checks> checks, <failures> failures",
      "required": true,
      "fixed_check_count_used": false,
-      "source_sha256_at_schema_audit": "cc7049e1c3919585caeefd1747be12fbf3fa52141fcdb980929c8a16fd7e2ee1",
+      "source_sha256_at_schema_audit": "5c52db4dd70f9702c096e918d849703d14aec1acb88bb2458ca01049da71ce3c",
      "authority": {
        "path": "docs/spec/acceptance/P23-03.json",
        "scope": "component and read-only observation only; native unexecuted; whole unit blocked"
@@ -0,0 +1,7 @@
+{
+  "action": "request-repair",
+  "task_id": "P23-03",
+  "reason": "Same-task repair for P23-03. Keep submitted 336a8cd and all earlier work, and keep every failure record (R10 B: 484 checks / 11 failures; validate 4958d8f55b6b458694bf158769ce3bff). Root cause, confirmed in the code: in tests/fixtures/p23-03/observation.gd, prepare() at lines 48-77 adds up only the registered construction_cost of the builds, including B's EV, to set a common diagnostic budget. Lines 93-99 then hire 5 candidates (staff_akane/gina/chika/iris/juno) and buy 40 RAW_FOOD without including either cost in that budget. Construction uses up all of B's cash, so all 5 hires fail with VALIDATION_FAILED (available 0) and the purchase fails with INSUFFICIENT_FUNDS. A succeeds. This is an omission in the fixture's budget, not a gap in production values.\n\nRepair, inside allowed paths only (tests/fixtures/p23-03/**, tools/check_p23_life_observation_evidence.py, tools/test_p23_life_observation_evidence.py, docs/plans/.../execution/p23-03/** metadata):\n(1) In prepare(), raise the common A/B diagnostic `needed` by:\n- the actual signing_cost of each of the 5 candidates, read at runtime from the registered candidate definitions (the catalog's signing_cost field, scripts/data/mvp_content_catalog.gd:323);\n- the gold cost of buying 40 RAW_FOOD, taken from the existing quote path (SupplyCatalog unit_price via PurchaseQuoteV2.make / the supply_management_service quote). Use effective_total if the actual quote applies a modifier.\nRead these values from the runtime registry/quote at the time they are needed. Do not hard-code numbers, add a margin, change default resources, or change product/runtime data. Record each cost source (definition/candidate/item id with its value), the totals, and the resulting deltas in the budget record. A and B must get the same deltas. The improvement EV cost stays in both budgets, as it is now.\n(2) Have the fixture check the receipt of each setup command (build, set_facility_conditions, hire, assignment, purchase, kitchen production) and expect() it to succeed, so a failed setup step shows up as a named failure and is not silently skipped.\n(3) Keep everything else the same: the common A/B inputs, the declared demand/arrival cohort, the native comparison's 48-demand, the 7 conditions and 6 categories, staff/goods coverage, normal OS plus distinct-PID load, full177, the independent Opus review, integration/final, 30/88 with no partial credit, the OS child timeout of 3600, worker/native 900 and the null cap.\n(4) Add a cheap local Python check, in the checker or its test, that statically or via the recorded budget confirms every planned spend (construction, signing, purchase) is counted in needed/deltas. Include a negative test for a missing spend category, so this mistake is caught before an expensive OS replay.\n(5) Update the p23-03 execution metadata to describe the finding and the repair honestly.\n\nThe worker runs only Python syntax checks and the checker's local tests, then makes a genuine submit. No Godot, native, OS or full regression runs, and Root evidence is expected to be missing at submission. Afterwards Root runs the exact final proof on the frozen source, reusing only evidence whose applicability matches. No new UI, product/runtime changes, broad hardening, billing or permission changes.",
+  "codex_model": "gpt-6.1-sol",
+  "codex_model_reason": "The repair is small and well understood: one fixture function needs signing and purchase costs added from existing catalog fields and the quote API, plus receipt assertions and a local Python preflight with a negative test. The root cause is already confirmed and nothing here needs new design or deep cross-system reasoning, so the default Sol is enough on quality and cheaper and faster than Astra."
+}
@@ -0,0 +1,7 @@
+{
+  "action": "run",
+  "task_id": "P23-03",
+  "reason": "The Runner shows P23-03 in repair_ready with 19 repairs, and the owner has dispatched the recorded R10 budget repair decision. run is a valid action and it is the next step, so this is not a repeat of a completed transition. The run keeps the R10 direction exactly as narrow as recorded and starts from the clean submitted base 336a8cd0b30db1add11869ebceab847db693f531. The worker stays inside the task contract's allowed_paths. It may only run local Python checks: syntax, the checker, and the checker's negative tests. It does not run Godot, native, OS, or full regression; Root runs those only after a genuine submission. The worker must say that Root evidence is missing at submission rather than mark anything PASS. Root's capacity recovery runs separately and must finish before proof replay; Root does not edit source. This run does not change the spec, the acceptance denominator, the budget or caps, or approval, and it does not touch main integration. Independent Opus review and Root validation are still required before acceptance.",
+  "codex_model": "gpt-6-astra",
+  "codex_model_reason": "I chose gpt-6-astra over the default because the difficulty justifies it. (1) The frozen spec's worker_instructions name one Codex Astra worker, and the recorded commander direction (run 83e836e2...) set codex_model gpt-6-astra. Changing the model during a narrow repair would break the recorded direction. (2) There have been 19 repair cycles, which shows the remaining defects are subtle consistency problems across the observation report, the strict checker that recomputes fields, the evidence manifests, and the same-demand layout A/B budget and seed setup. These need careful reasoning across files more than speed. (3) The worker only runs local Python, so per-run usage is bounded. A stronger model that gets it right in one pass should cost less overall than more failed cycles on the default model."
+}
@@ -0,0 +1,45 @@
+# R10 diagnostic budget repair submission
+
+Status: submitted for Root verification; no acceptance credit claimed.
+Contract base: `860136f7099f5e3aee99da5798a627a77f2d8eea`.
+Starting submitted source: `336a8cd0b30db1add11869ebceab847db693f531`.
+
+The mirrored OPUS_R10_BUDGET_REPAIR_DIRECTION.json and
+ROOT_OS_R10_BUDGET_FINDING.json resolve the input gap recorded in
+R10_CONTEXT_GAP.md. That historical note and all copied evidence are retained.
+The failed R10 B run remains 484 checks / 11 failures, validate
+`4958d8f55b6b458694bf158769ce3bff`; it provides no acceptance credit.
+
+The observation fixture counted construction only. Its common A/B budget now
+also counts all five planned candidates' registered catalog signing_cost and
+the 40 RAW_FOOD purchase. The shared assignment list and purchase quantity
+feed both budgeting and actual setup commands. PurchaseQuoteV2.make reads the
+current SupplyCatalog unit_price; the service's eligibility branch selects
+base_total or effective_total. The live supply preview is recorded and checked
+against that budget immediately before purchase. No margin or production
+value is added. Both budgets still include B's improvement construction.
+The budget record retains construction rows and adds construction totals,
+candidate sources/costs and signing total, item/unit price/quote/purchase total,
+needed, initial resources and deltas.
+
+Every setup operation already passed through command(), which asserts its
+CommandResult.is_success() and records its receipt. Failure names now include
+the payload, distinguishing individual builds, hires and assignments. This
+continues to cover conditions, purchases and kitchen production without skips.
+
+A cheap checker preflight verifies all three spend categories enter needed
+before deltas and that the setup shares its assignment/quantity plan. Local
+negative tests remove each spend increment and delta coverage. These are
+static checks, not runtime evidence or acceptance assertions.
+
+Worker validation: 51 local Python checker tests passed; Python syntax checks
+passed. No Godot, native, OS, full regression, Runner operation, git write,
+production checkout access or additional agent was used.
+
+Root evidence for this repaired source is missing at submission, as expected.
+Root must freeze the source and perform the exact native and normal OS A/B
+proof, distinct-PID load, mandatory full177 regression, independent Opus review,
+integration/final and ledger. Reuse requires matching applicability. The native
+48-demand, observation cohort, seven conditions/six categories, staff/goods
+coverage, 30/88 with no partial credit, OS child timeout 3600, worker/native
+900 and null cap are unchanged.
@@ -0,0 +1,45 @@
+# R10 repair context gap
+
+Status: blocked on missing recorded R10 decision, not on missing Root proof.
+
+The read-only HEAD check returned the requested submitted starting point
+`336a8cd0b30db1add11869ebceab847db693f531`. The contract base remains
+`860136f7099f5e3aee99da5798a627a77f2d8eea`. The initial worktree contained an
+untracked `tools/__pycache__/`; it was preserved.
+
+The supplied run direction requires keeping the recorded R10 budget repair
+exactly narrow, but does not include that decision's concrete repair instructions
+or location. The available `OPUS_CONCRETE_IMPLEMENTATION_DIRECTION.json` is the
+initial recipe (run `83e836e232c74975aaf51f4fab5a1af9`), not the R10 decision.
+No R10 decision was found in this checkout's P23 execution materials or fixtures;
+`.review-material` is absent. A targeted JSON/Markdown search for R10 budget
+direction in the sibling P23 runtime directory found no match. Access to the
+normal production checkout's P23 execution directory and sibling validation
+checkouts returned `Operation not permitted`; no access workaround was attempted.
+
+Static context inspected before considering source edits:
+
+- `API_AND_STATE_MAP.md` describes deterministic initialization and public
+  arrival/build commands. Current `simulation_runtime.gd:203` still accepts the
+  explicit seed, and `game_state.gd:20` resets resources to defaults.
+- `layout_comparison.gd:21` declares the budget as `GameState.reset`; its run
+  resets state before initialization and records the initial snapshot before
+  issuing layout-specific build commands.
+- `observation.gd:49` onward derives a common diagnostic construction budget
+  from registered definition costs, including the declared B improvement, and
+  issues `change_resources` before building. This is a separate observation path.
+- `check_p23_life_observation_evidence.py:1212` requires equality of A/B initial
+  snapshots. These observations do not establish which precise repair R10
+  authorized, or justify inventing budget/seed policy.
+
+Required continuation input: the complete recorded R10 budget repair decision
+and its referenced failure evidence, mirrored into readable task materials (or
+provided directly), including any prescribed budget/seed and checker changes.
+No specification, acceptance, budget, production, fixture, checker, renderer,
+or policy change was inferred. Only this context-gap note was added.
+
+No tests were run: there is no source repair to validate. No Godot, native, OS,
+full regression, Runner operation, git write, integration, or additional agent
+was used. Root exact-candidate native/OS/resume/full-regression evidence and
+independent Opus review remain missing and required; no PASS or acceptance
+credit is claimed. Root capacity recovery remains separate.
@@ -0,0 +1,443 @@
+# P23-03 repair inspection
+
+## Repair 26 — submitted for Root validation; no acceptance credit
+
+This section supersedes the status below. Preserved the five premodified files,
+both R13 fixtures, all 86 existing tests, and every historical section. Historical
+49 and the twelve false R13 judgments are unchanged. Seven conditions, six
+categories and 30/88 are unchanged. No Runtime, gameplay, save, economy, mana,
+asset or UI-entry source was changed.
+
+Implemented the remaining local work:
+
+- Both independent packers now rebuild vertical components from fitted feet and
+  pose cores after every scale reduction. Scale ceilings decrease monotonically,
+  so termination has a finite measure and no iteration-limit exemption. Stable
+  components use the existing ordered forward/backward fitting and p=1 retained
+  member aggregation. A positive separates a previously intersecting component
+  at p=2; retaining the stale component would incorrectly reduce it to p=1.
+- Whole-state fixtures cover queue lanes plus work staff, queue precedence and
+  order, the actual R13 cafeteria/access-cell geometry, carried and treated
+  patients, occupied rest facilities, queued staff, passenger/floor coexistence
+  on FRONT/BACK, and independently calculated no-fit aggregates. Negatives
+  include stale groups, pose/feet, carrier-relative positions, category/region/
+  transit spoofing, missing identities and displaced aggregation badges.
+- Cabin candidates no longer silently truncate after two seat rows. Additional
+  identities reuse the existing two-row anchor pattern and participate in the
+  same independent fit/no-fit decision. The synthetic saturation configuration
+  retains every boarded identity in member lists; it changes no definition.
+- OS and offline callers use the shared hash-bound drawing_context loader.
+  Tests cover valid context, absent fields/objects, malformed raw/rest context,
+  report binding and object hash mismatches. Frame tests include valid transit
+  and occupied-rest authority, missing/malformed rest authority, and conservative
+  rejection of overlap without raw context. No flags grant exemptions.
+- GDScript component fixture now includes four occupants, scale-driven regroup,
+  and actual _fit_life_actors aggregation with nine drawn and ten retained IDs,
+  badge text and position. This fixture has NOT been executed by Godot.
+
+Source/consumer audit:
+
+- _build_life_observation calls fitting before publishing; _draw_life_overlay
+  consumes fitted floor/passenger rows and the aggregation badge at/text.
+  _draw_observed_actor consumes at/scale/mode/frame/flip. Logical cell, preferred
+  center and state remain separate. Existing click handling selects world
+  facilities/construction/stairs; there is no actor hit-test to relocate.
+- passive_observation.draw_rectangles and Python drawing_rectangle_rows consume
+  fitted actor extents and actual aggregation badge coordinates. frame_rows
+  carries all member identities and drawn_keys into review metadata. Whole typed
+  diagnostic captures preserve the same report; compare_actor_geometry rebuilds
+  classification, bounds, fit and aggregation independently. compare_rest invokes
+  complete geometry for occupied resting facilities. compare_motion binds the
+  actual published report/history, not unfitted preferred positions.
+- inspect_offline_mirror and inspect_os_requests both pass drawing_context's
+  hash-bound state/rest objects to frame_rows. inspect_stream separately runs
+  rest, cabin, motion and complete geometry checks. Missing objects raise and
+  cannot become a successful exemption. Unit tests exercise this shared loader;
+  no real OS/offline evidence was rerun or relabeled.
+- strict_test_markers.load_contracts rejects duplicate suite rows. The existing
+  single P23 row already declares the exact two source print sites' template
+  `P23 life observation: <checks> checks, <failures> failures`. parse_raw obtains
+  counts only from actual logs; source_sha256_at_schema_audit is historical
+  metadata, not a current-source acceptance bypass. No marker-contract append
+  is required, no suite-policy row was added, and no marker was fabricated.
+  Root's source map must bind the final modified fixture and all changed files.
+
+Local verification: 87, 91, 93, 95, 97, 98 and 99 tests passed as work progressed;
+the final rerun passed **99 tests in 0.213s**, exit 0. The original expected R13 row11/29 rejection
+is retained. Python compile checks for checker, tests and runner passed without
+bytecode output. git diff --check passed. No tests were deleted or weakened.
+Read-only discovery encountered inaccessible sibling worktrees and nonexistent
+BUILD_OUTPUT_DIR shorthand; the actual allocated build directory supplied both
+R13 direction files. A shell glob lookup had no match; no source was lost.
+
+Root native/OS evidence, the 177-test full regression and independent Opus
+all-condition review remain missing, as expected for worker submission.
+Submission is not acceptance and earns no unit credit. No Godot, native/OS,
+full regression, Runner, provider, git write or agent was run.
+
+## Repair 25 continuation — current status: blocked, implementation incomplete
+
+This section supersedes the status of the historical sections below. There is
+no specification, permission, or floor-interpretation gap. The remaining blocker
+is unfinished implementation and verification, not missing Root evidence alone.
+All 809 pre-existing added lines were preserved and extended. No candidate is
+claimed implemented or ready for acceptance. Seven conditions, six categories,
+and 30/88 remain exact; historical 49 is preserved without relabeling.
+
+Read before editing: AGENTS.md, Phase 0 API_AND_STATE_MAP, recorded concrete
+direction, both authoritative R13 geometry/region directions, renderer,
+checker/tests, diagnostic/passive collectors, Runtime initialize/spawn/save/load
+entry points, enums, marker parser and existing contract. No new assets or
+Runtime/gameplay/save/UI-entry changes.
+
+Changes in this continuation:
+
+- Drawing-row validation accepts the existing raw-state/rest-audit context and
+  invokes complete geometry classification and fitting. OS frames pass the
+  hash-bound stream objects; offline image review loads the matching writer
+  capture and checks its report binding before passing the same objects.
+- Track validation checks the raw actor endpoint, source-valid adjacent origin,
+  finite nonfuture times and boolean pause. A drawing's region/in_transit flags
+  cannot grant an exemption. Without raw context the old conservative check
+  remains, and frame output explicitly says `absent: fail closed, no transit
+  exemptions`. Historical images have not been rerun or relabeled.
+- Rest selection still independently validates routes, costs, locks and
+  eligibility. For actual occupied rest facilities its geometry now runs through
+  complete cross-kind fitting, including queue precedence and independently
+  reconstructed aggregation identities/no-fit, instead of requiring every
+  rest-slot identity to have an unshifted drawing row.
+- Added the actual +m badge to Python and GDScript image-audit rectangle lists;
+  frame metadata retains all aggregate members and drawn keys. This additional
+  diagnostic row type does not change the six acceptance categories.
+- Added full state-to-geometry positives for actual track interpolation,
+  staff/guest mixed packing on FRONT/BACK, fractional low-world-pixel scale and
+  saturated aggregation with all identities retained. Added track/time/pause,
+  forged-region/out-of-cell, aggregate identity loss and badge omission checks.
+  R13 row11/29 remains an expected FAIL. Existing tests were not removed.
+
+Precise outstanding TODOs before an implemented submission:
+
+1. Complete full state-to-geometry positives for queue lane + work staff,
+   passenger + floor, carried/resting/treated staff and cabin aggregation.
+   Exercise compare_rest with an actual occupied facility, legitimate no-fit
+   aggregate and queued staff; its new integration lacks those full fixtures.
+2. Complete full semantic negatives for queue precedence/order and forcing the
+   receiving R13 guest back into its access cell. Existing primitive receiving
+   fixture tests and overlay negatives are retained but are not substitutes for
+   every required state-to-geometry case.
+3. Audit and implement regrouping after scale changes in BOTH packers. They
+   currently form vertical groups before fitting; fitted feet/cores can change
+   vertical intersections. Do not claim the required regrouping positive passes.
+   Confirm carrier-relative pose/feet, cabin member coverage, and badge placement
+   through the required semantic cases before declaring the renderer finished.
+4. Add full frame/offline/OS-context unit cases (including rest authority and
+   malformed/missing stream context). Standalone/full transit paths have tests;
+   production caller wiring was read but no OS/offline evidence run occurred.
+5. Finish the GDScript UI fixture and source-authority integration audit. Existing
+   four-occupant UI preparation is retained and unexecuted. Marker contracts and
+   suite policies are unchanged; do not append a duplicate suite row or fabricate
+   an acceptance marker. If an authority append is necessary, only the single
+   strict contract-permitted entry is authorized.
+
+Local checks actually executed:
+
+- Checker unit command `PYTHONDONTWRITEBYTECODE=1 python3
+  tools/test_p23_life_observation_evidence.py`: 78 tests OK (0.104s), 80 OK
+  (0.112s), 83 OK (0.134s), 85 OK (0.112s), then **86 OK (0.124s)**.
+  No checker unit failure occurred in this continuation.
+- No-bytecode `compile(Path(p).read_bytes(), p, "exec")` syntax check for
+  checker, checker tests and Root runner: all three OK, exit 0.
+- Read-only `git diff --check`: exit 0, no output.
+- One shell heredoc edit attempt failed before execution (`can't create temp
+  file for here document: operation not permitted`); applied the intended edit
+  with apply_patch afterward. One read-only search named a nonexistent
+  `tools/regression/suite_policy.py`; read the actual JSON policy and runner
+  directly. These were tool/read failures, not test passes.
+
+No Godot/native/OS, full regression, Runner, provider, git write, integration or
+other agent was run. Root's **new native/OS evidence**, the **177-test full
+regression**, and the **independent Opus all-condition review** are still missing.
+Those expected later proofs alone are not the reason for this blocked status.
+
+## Repair 24 continuation — current status: blocked, implementation incomplete
+
+The floor/anchored-region interpretation is RESOLVED. No clarification or
+permission is requested. This status is an incomplete implementation, not an
+external specification blocker and not a submitted candidate. Preserve this
+worktree and every earlier uncommitted change. Historical sections below are
+retained as history, not current interpretations.
+
+Read the current renderer, RoomPresentation, the recorded concrete direction,
+both R13 direction files, Phase 0 API map, Runtime initialization/spawn/formal
+save/load sources, and cleaning pose dependencies. No domain inputs or assets
+were changed. Seven conditions, six categories and 30/88 remain exact.
+
+Implemented preparation in this continuation:
+
+- Added per-actor cell/track, lane, facility, work-column and carried-patient
+  bounds, inset by world pixel; added original logical cell to drawing rows.
+  Staff queue placement now precedes the special staff anchors.
+- Added a cross-kind vertical grouping and forward/backward interval packer,
+  integer scale trials and deterministic retained-member aggregation with a
+  readable +m badge. Cabin rows participate. Preferred centers and pose fields
+  remain separate from fitted drawing coordinates. This GDScript has NOT been
+  executed and is not established as native-correct.
+- Added independent Python interval fitting and replaced original-scale
+  assertions with reconstructed centers, bounds, poses, fitted rows and
+  aggregation comparison. Queue and room reports are cross-checked before
+  geometry classification; boarded keys are rebuilt from EV reports.
+- Added tests for the 56px four-mover case (all p=2), mixed actor kinds,
+  saturation with all identities retained, individual intervals, and R13's
+  receiving cafeteria guest outside its unchanged access cell. Added a full
+  synthetic state-to-geometry four-mover case and mutations for cell, state,
+  scale, position, missing/duplicate actors and unnecessary aggregation.
+- Added the four-occupant GDScript component fixture. It is NOT native evidence.
+
+Known remaining implementation work (must be resolved before submission):
+
+1. The full geometry entry point passes independently derived transit keys to
+   pairwise checking, but standalone drawing-row/offline checks still use the
+   conservative default with no transit exemptions. Reconcile those paths
+   using authoritative inputs; never trust an in_transit/region flag.
+2. Existing compare_rest still expects each visible slot in actor_drawings and
+   its original rest center. Reconcile legitimate aggregation and the new
+   staff queue precedence without bypassing the independent no-fit check.
+3. Finish full classification tests for queue/work mixed groups, queue
+   precedence/order, spoofed category, forced access-cell receiving position,
+   carried/resting/treated staff, passengers and FRONT/BACK. Existing primitive
+   packing positives are not a substitute for these full semantic cases.
+4. Audit pose/feet and regrouping after scale changes, low-world-pixel views,
+   carried patients and cabin aggregation; validate the actual GDScript and
+   report consumers through Root only after worker readiness. No new native
+   correctness or complete collision-removal claim is made here.
+
+Checks actually executed in this continuation, all local:
+
+- `PYTHONDONTWRITEBYTECODE=1 python3 tools/test_p23_life_observation_evidence.py`:
+  71 tests OK (0.093s), 76 tests OK (0.102s), then **78 tests OK (0.111s)**.
+  Existing R13 row11/29 remains an expected rejection. No unit test failure
+  occurred during this continuation; earlier failures remain recorded below.
+- No-bytecode `compile(Path(p).read_bytes(), p, "exec")` for checker,
+  checker tests and Root runner: all three syntax checks OK, exit 0.
+- One dependency lookup used the wrong cleaning-bridge directory and returned
+  an rg missing-file error; located and read the actual source under
+  scripts/environment. This was a read failure, not a passing test.
+
+No Godot/native/OS, full regression, Runner, provider, git write or agent was
+run. Root's **new native/OS evidence**, the **177-test full regression**, and
+the **independent Opus all-condition review** are still missing. Historical
+49 remains preserved and is not relabeled. No partial acceptance is claimed.
+
+The inspection below is retained history. The repair 24 section above records
+the current status; the former full-box impossibility argument is superseded.
+
+Scope: the seven immutable acceptance conditions and existing allowed paths.
+Validation: local Python syntax and checker unit tests only; no Godot execution.
+
+Read the immutable acceptance spec, recorded concrete implementation direction,
+existing Phase 0 API map, renderer, actor art, and independent checker.
+The R13 provider stdout, command, and prompt SHA256 values match 01-actual.json.
+The actual structured review contains 8 frames and 132 rows. Its summary's
+168-row claim is not used to replace the structured rows. False judgments remain
+false. The reader-request-0003 whole screenshot confirms overlapping actors and
+obscured traffic presentation. Historical runtime evidence is unchanged.
+
+Root native/OS evidence for a new candidate, the 177-test full regression, and
+independent Opus all-condition review remain required and are not worker tests.
+
+## Status: blocked; partial independent checker preparation only
+
+The renderer repair and traffic geometry/layer audit are **not implemented**.
+No claim of submission readiness or acceptance is made. No renderer, Runtime,
+policy, immutable spec, source authority pin, or historical evidence was changed.
+
+### Placement specification gap and concrete failing evidence
+
+In the original `os-pair/A/reader-request-0003/request.json`, SHA256
+`7dc3ee0bcc428e9d103726ee2448c767f1dd4107f9490833dc860c0b5140068c`,
+the decoded report contains guests `os-p23-012` through `os-p23-015`, all
+`MOVING`, logical cell `[1,1,0]`, with no service or queue request. Their
+tracks' `from` and `cell` are both `[1,1,0]`. The current cell is 64x96 pixels
+(`traffic` row 40's inset rect is `[709,390,60,92]`). Each actor's current
+draw/report footprint is 48x54 pixels: the 16x18 PixelPeople grid at scale 3.
+The four rectangles require area 10368, while the cell has area 6144.
+Consequently offsets alone, preserving those sizes and containing all four
+rectangles in that cell, cannot satisfy the requested nonoverlap invariant.
+This is already present in the supplied evidence, not a hypothetical load.
+
+Source references: `world_preview.gd::_cell_dimensions`, `_actor_drawing`,
+`_tracked_actor_position`; checker `drawing_rectangle_rows` and
+`compare_actor_geometry`; `pixel_people.gd`'s 16x18 drawing grid.
+
+The direction specifies offsets inside room/cell art, containment consistent
+with logical cell/state, retained motion, and no visual redesign. It does not
+resolve what presentation change is permitted when existing footprints cannot
+fit. Shrinking actors or expanding the permitted projection area could solve
+this capacity issue, but neither is silently treated here as authorized by an
+offset repair. This is a question about the admissible presentation, not a
+claim that every possible rendering solution is impossible.
+
+Minimal proposal for the owner: specify whether deterministic scale reduction
+within the original logical cell is permitted (including the required readable
+minimum and behavior when it still cannot fit), or define an allowed room/lane
+projection for these non-queued moving actors. No Runtime capacity change,
+actor omission, or acceptance waiver is proposed. Retain all seven conditions.
+
+### Independent work preserved
+
+Added `validate_actor_nonoverlap` to both the semantic geometry path and offline
+drawing-row gate. It recomputes full footprints from raw `at`/`scale`, includes
+cabin passengers, rejects duplicate identities and invalid numeric geometry,
+and does not use eligibility labels or asserted booleans. Edge contact is
+allowed; positive-area overlap is rejected with actor identities.
+
+Five added unit tests cover edge contact, overlap despite forged verdicts and
+off-screen labels, cabin/floor overlap, invalid footprints, and the actual R13
+false row 11 against staff row 29. The small historical-negative fixture copies
+only the needed raw drawing inputs and preserves `[11,false,false,false]` with
+request/manifest hashes. Its inputs were checked against the original typed
+report. The original review files and all their rows remain untouched.
+
+The original manifest hash matches
+`53aa8e2f12e5d0a089529457fcd55ee1ba0600e8dffff2225bb8b057d007fe1a`;
+all crops in that manifest pass SHA verification. Opened the whole decoded image
+and crops 11, 29, 40 directly; crops 11 and 29 show the same covered figure.
+
+Still missing: collision-free renderer placement, draw-consumed band rectangle,
+actual layer/geometry audits, and the requested additional cell/state, band,
+layer, and stale-frame negatives. Passing these local tests does not resolve
+those missing deliverables.
+
+### Local checks actually run
+
+Before edits:
+`PYTHONDONTWRITEBYTECODE=1 python3 tools/test_p23_life_observation_evidence.py`
+returned exit 0, `Ran 51 tests in 0.142s`, `OK`.
+
+After edits the same command returned exit 0,
+`Ran 56 tests in 0.093s`, `OK`.
+
+Syntax command (no bytecode output):
+```
+python3 -c 'from pathlib import Path; files=["tools/check_p23_life_observation_evidence.py","tools/test_p23_life_observation_evidence.py","tools/run_p23_life_observation.py"]; [(compile(Path(p).read_bytes(),p,"exec"),print("syntax OK:",p)) for p in files]'
+```
+Exit 0; printed `syntax OK:` for each of those three files.
+`git diff --check` returned exit 0 with no output (read-only check).
+No Godot, native/OS, full regression, provider, Runner, git write, or agent
+execution occurred. The initial shell heredoc attempt failed before execution
+because the shell could not create its temporary file; subsequent inspection
+used `python3 -c`.
+
+## Recorded geometry direction continuation — blocked pending placement clarification
+
+Read the complete `BUILD_OUTPUT_DIR/negative-image-review-r13/
+OPUS_GEOMETRY_RESOLUTION_DIRECTION.json`, the owner run direction, the current
+PixelPeople and WorldPreview sources, RoomPresentation slots, the existing
+Phase 0 audit and the actual initialize/spawn/save/load API sources. The spec
+still matches its contracted **stripped-text** SHA256. No spec, Runtime, save,
+asset, UI entry, gameplay value, authority entry or suite policy was changed.
+
+### Implemented independent preparation
+
+- Replaced the full-box overlap invariant with identity cores. The table is
+  pinned to PixelPeople SHA256
+  `a171703083aa2d00f8be2b63e1daef9c41717c8fe59d8056878bc4b7df1e0b0c`.
+  Upright walking/carry-walking phases 2 and 6 shift the core up one grid unit.
+  Standard-room sleep phases 2, 3 and 4 extend its top from y=11 to y=10,
+  keeping the bottom at y=16, as required by the actual breathing marks.
+  Flips use the actual 16-grid mirror. Missing pose inputs now fail closed.
+- Preserved the edge, forged-verdict, cabin, invalid-geometry and R13 tests.
+  Edge coordinates now use 6x12 core extents instead of the rejected 16x18
+  full-box extents; the deliberate overlap moves from [15,17] to [5,11].
+  Added a passing full-box-overlap/disjoint-core case and bob, breathing,
+  flip, missing-pose, duplicate, forged-transit and source-pin tests.
+  R13 row11/29 still fails: both recomputed cores are [669,520,18,36].
+  The fixture's newly required mode/frame/flip values were copied from the
+  original request and checked against it. False judgments remain false.
+- Added the LifeOverlay Node2D between InactiveLayerRooms and LayerLens,
+  connected its draw signal, and queued its redraw from the parent draw.
+  The overlay draws traffic first, then cabin passengers and floor actors.
+  Existing inactive-art alpha remains 0.22. Actor rows and animation inputs
+  are unchanged. The report records observation/consumed serials, actual
+  child indices and the passes executed in the callback.
+- Added the prescribed band rectangle at cell_bottom - 2*world_pixel,
+  with existing thickness clamp, colour thresholds and 0.90 opacity.
+  The blocked X is contained in the band pass. `compare_traffic` independently
+  recomputes both cell and band rectangles from typed state. Overlay checks
+  reject stale serial, reversed passes, child-order errors, a core covered
+  by the band, and forged band geometry. Source wiring/order is also checked,
+  so valid report flags do not excuse the tested reversed source callback.
+  FRONT/BACK local synthetic checks and GDScript serial/order assertions were
+  added. GDScript was **not executed**.
+
+### Exact outstanding interpretation
+
+Placement rule 1 gives service/rest/treatment anchors a facility region, while
+rule 2 says a floor actor must *also* remain in its own logical cell. It is
+unclear whether “floor actor” excludes these facility-anchored actors or means
+every non-passenger. This is consequential in the actual R13 request, not only
+in a fabricated layout:
+
+- `guest-auto-00000044` receives `service_00000009` at cafeteria
+  `facility_00000003`. Its unchanged logical cell is [5,0,0]. The recorded
+  traffic inset [965,486,60,92] reconstructs cell [963,484,64,96].
+  The facility rect is [1027,388,256,192], and its four-pixel inset starts at
+  x=1031. There is no common x interval with the logical cell ending x=1027.
+- `guest-auto-00000019` receives `service_00000002` at standard room
+  `facility_00000007`, rect [131,292,128,96]. Its logical access cell [-6,2,0]
+  reconstructs to [259,292,64,96]. The facility inset ends x=255; again there
+  is no common x interval. Its room pose is standard_room_sleep.
+
+`r13_service_region_gap.json` preserves the needed original report fields,
+with source request SHA256
+`7dc3ee0bcc428e9d103726ee2448c767f1dd4107f9490833dc860c0b5140068c`.
+Direct comparison verified every copied guest, traffic and room rectangle.
+A diagnostic unit test demonstrates the empty intersection for both examples.
+These extracted fields are not substituted for whole typed acceptance evidence.
+
+This does **not** establish that the new specification is impossible: rule 7
+allows aggregation when p=1 cannot fit. If floor means all non-passengers, these
+otherwise roomy facilities must aggregate their service guests with zero drawn
+members. If floor means unanchored actors, those guests can be packed and drawn
+in their facility region. The initial commentary's “constraint conflict” is
+therefore narrowed to this interpretation question. No cell restriction was
+silently removed and no zero-drawn service aggregation was silently adopted.
+An asynchronous question asks the owner to select the intended interpretation.
+
+### Remaining implementation and evidence
+
+Ordered placement, scale fitting, no-fit aggregation, queue-order preservation,
+carrier-relative fitted placement, transit recomputation/exemptions and their
+checker integration remain unimplemented. The checker remains conservative:
+an asserted in_transit flag never bypasses overlap, but the required genuine
+transit exception is not implemented. The existing geometry checker still
+expects original actor positions/scales. The four-occupant GDScript fixture,
+positive packing cases (staff/guest, passenger/floor, queue, saturation and
+carried patient), and the new containment/aggregation/queue-order negatives
+remain outstanding. The partial renderer changes are not a submitted candidate
+and are not claimed to solve R13 actor collisions. All existing tests remain.
+
+Still missing: Root **new native/OS evidence**, the **177-test full regression**,
+and **independent Opus all-condition review** for the final candidate.
+Historical 49 is preserved and is not relabeled or credited to these edits.
+Seven immutable conditions and all6/30/88 remain exact, without partial credit.
+
+### Checks actually run in this continuation
+
+- `PYTHONDONTWRITEBYTECODE=1 python3 tools/test_p23_life_observation_evidence.py`:
+  first core run: 63 tests, 7 failures and 1 error because older synthetic
+  reports omitted pose fields. Updated those fixtures with explicit poses;
+  63 tests then passed (0.089s), followed by 64 (0.094s) after the bounds
+  diagnostic. First overlay run: 71 tests, 1 error because the older spatial
+  fixture lacked the new band rectangle. Updated that fixture with explicit
+  band geometry and consumed-frame fields. Final run: **71 tests, OK, 0.081s**.
+- The existing no-bytecode `compile(Path(p).read_bytes(), p, "exec")` syntax
+  check passed for checker, checker tests and Root runner script (exit 0).
+- Direct Python read-only comparisons verified both extracted R13 fixtures
+  against the original SHA-bound request and verified the immutable spec pin.
+- A shell heredoc read attempt failed before execution with “can't create
+  temp file”; inspection continued with `python3 -c`.
+
+No Godot, native/OS, full regression, Runner, nested provider, git write or
+other-agent execution occurred. Only allowed files were edited. The worktree's
+previous uncommitted work was retained and extended.
@@ -0,0 +1,86 @@
+# Repair28 continuation / repair29 source handback
+
+Record: SOURCE-READY, source-only submission for P23-03. This is not acceptance.
+The preserved repair28 implementation is unchanged by this continuation. Only the
+existing P23 marker source audit SHA and this new note are changed in repair29.
+Older notes, raw receipts and failure history remain unchanged.
+
+## Input and preserved-state verification
+
+Read all six Root inputs in full from the explicitly supplied
+`.runtime/P23-03/build/audit-support/root-repair29-input/` directory outside W.
+Each byte count and SHA256 matches MANIFEST.json. Its authoritative patch is
+64,737 bytes (SHA256 6b0a2a236a08f1c473850c06f7a761d1f0f8f4749db271771550d409d46e37b0),
+despite the older instruction's 384318-byte description. No input is missing.
+Before editing, all eight W file lengths and SHA256 values matched
+PRESERVED_MANIFEST.json; `git diff --binary` equalled PRESERVED_DIFF.patch byte
+for byte. The untracked r15 fixture was separately verified by its manifest SHA.
+
+## A–E disposition and source map
+
+| Item | Existing implementation / disposition |
+| --- | --- |
+| A1–A2 | scenes/main/world_preview.gd:1057 expands stops by reported shaft width; :1076 stops outward lanes at the footprint; :1094 retains member/visible/overflow accounting and badge offset. tools/check_p23_life_observation_evidence.py:965 independently reconstructs queues. |
+| A3 | scripts/ui/room_presentation.gd:77 supplies count, operation, doors and phase text; scenes/main/world_preview.gd:1224 draws existing PixelArt cabin/gate and badge on LifeOverlay; :811 draws cabins above inactive-layer context. |
+| A4 | scenes/main/world_preview.gd:640 collects opaque obstacles, :674 subtracts rectangles, :831 draws every traffic row including fully covered rows beneath opaque later cabins/badges. No density row is discarded. |
+| A5 | scenes/main/world_preview.gd:564 deep-copies the consumed model, including shaft_footprint (:1097), badge_rect (:1286), traffic_obstacles (:649), segment_rects (:665). |
+| B(a–d) | tests/ui/test_p23_life_observation.gd:32 covers width-2, overflow, empty cabin, both layers, operation/doors/phase and traffic geometry. tools/test_p23_life_observation_evidence.py:40 adds meaningful R15 intrusion/traffic and typed-state negatives. tests/fixtures/p23-03/passive_observation.gd:194 uses the reproducible badge rectangle. |
+| C1–C2 | tools/check_p23_life_observation_evidence.py:1840 validates external S/M/T/Q, allowed changes, deterministic merge tree, S-only/M-only blobs and complete T source/settings/input/QA bindings; :1799 and :1821 enforce the structural marker join retaining M/P18. tools/run_p23_life_observation.py:245 consumes external roles; :334 records v2 evidence. tools/test_p23_life_observation_evidence.py:135 covers composition negatives. |
+| C2(vi), C3 | tools/check_p23_life_observation_evidence.py:2108 refuses v1 history as current evidence; :1986 allows one screenshot only for screenshot requests, retaining real-input pairs. Existing staff, OS receipt, distinct-PID resume and image gates remain. Root owns final full regression and independent review. |
+| D | The earlier assertion that kitchen full177 was consistent is withdrawn: p04-final-product-full-r01 failed the relationships-v2 kitchen envelope (legacy/version/original_relationships added; threshold_candidates/next_candidate_sequence missing). Old 177 gives no T credit. No kitchen source or overlay was changed. ACTUAL_SUPPORT_COMPLETED.json records approved QA3a9c support integration 5041b88bdf25d0deaf3cb335be981896522c4880 (reported main tree df9b2851edaa7d02a55b834328cf9e9a9bb7203a). Root must bind the actual later M and its kitchen QA blob to T; neither 9b nor any old full result is fixed as authority. The four D05 choices are not worker blockers. |
+| E; repair29(b–c) | This note supplies the continuation handback and structured SOURCE-READY record. |
+| repair29(a), step 4 | completion_marker_contracts.json:3710 now pins the actual final GDScript test SHA. Only that existing row's audit SHA changes; template, required flag, authority scope and every other row remain intact. |
+
+No concrete remaining A–C source omission was identified in this bounded check;
+therefore no code or test was changed and no new negative test was needed.
+
+## R15 negative fixture provenance
+
+tests/fixtures/p23-03/repair28_r15_negative.json is an extraction of historical
+local-coordinate fields, not a raw receipt. It identifies source 9a26f9a,
+manifest a25e15f00d09acb16b05e180a6263c3771c84449bd6e0c99b6968fd739727bd2,
+request 7099ca03da3cef5819e7965f6fdea8777b9995a9e12806e870e844ec2b02b9a5,
+R15 os-pair/A/reader-request-0003/request.json and negative rows 6/17.
+Tests calculate counterfactual geometry only in memory. This extraction and all
+synthetic projections grant no S/T execution or OS credit and do not relabel raw history.
+
+## Frozen changed-source SHA256
+
+| Path | SHA256 |
+| --- | --- |
+| scenes/main/world_preview.gd | 6c43e19fbce8ff5c442017ab22443f39f65f356c308ee895355cd7678d307f33 |
+| scripts/ui/room_presentation.gd | 2eda851534777575298ea005e2e6d0ee90785a81357d5b9c34cfb95f615a8181 |
+| tests/fixtures/p23-03/passive_observation.gd | 812c2a3b399b079333e4eae0fbf78946f1c39041e6859e65f8bc7c44aaad6bfc |
+| tests/fixtures/p23-03/repair28_r15_negative.json | 8525cdfbc504583778624ab103efb4642374715be73187e97bdbf18d180a1616 |
+| tests/ui/test_p23_life_observation.gd | 610ff2f33ec0e42bdca55d692e640c9bd8fe9244b9b49b0263419b1998ff9e9e |
+| tools/check_p23_life_observation_evidence.py | 767f3452cd9f0671e25bfda113153b0b1471fcbfdfa7aba6133f991a61cf306d |
+| tools/run_p23_life_observation.py | 1fa85aa53886887f9611997194240bf6e112f831e8f36ea3aff982d12538bdec |
+| tools/test_p23_life_observation_evidence.py | 4a52215b125f3445abde9992ee69ad9b09fc114977e359b52f190c966e6d3c39 |
+
+## Verification and pending Root evidence
+
+ROOT_LOCAL_TEST_RESULT.json is authoritative: `python3 -B
+tools/test_p23_life_observation_evidence.py`, exit 0, **123 tests, 0 failures**,
+0.267 seconds. The earlier worker count of 121 is superseded. These unchanged
+tests were not rerun in repair29; no Python source changed, so no syntax rerun
+was needed. Manifest/hash/diff comparisons via `python3 -B -c` exited 0.
+`git diff --check` exited 0. Final `python3 -B -c` hash/byte comparison exited 0:
+all eight preserved sources remain unchanged, and the marker file differs from
+HEAD by exactly the one authorized SHA replacement. Marker file SHA256 is
+4dbba39cad454d02a004ee3d000457c3e13f7fc017dcfdb847aca7b192a8e30b.
+The P23 marker entry SHA256 is
+f83f022c44db6f89fc18d634f7ec9ce54b56450299c0e30fb1275cd2b88b6b39
+(UTF-8 JSON, sorted keys, compact separators, ensure_ascii=False).
+Initial shell attempts failed before validation: heredoc temporary-file creation
+was denied (exit 1); `python` was unavailable (exit 127). The successful checks
+used `python3 -B -c` without temporary files.
+
+Still missing and explicitly Root's work: exact-source native/component and
+impact evidence, same-demand A/B, normal OS on both layouts across all six
+categories and FRONT/BACK/pan/zoom, formal save and distinct-PID normal load,
+actual CUA receipts and independent bound-frame image judgments, one mandatory
+full regression on T, and independent Opus all-condition review. Root binds
+S/M/T/Q after submission and performs normal integration/final gates. No worker
+engine, OS, Runner, Git write, main operation or subagent was used. Seven
+conditions, six categories, 30/88 denominator, formal13, max repairs null and
+all counters/caps remain unchanged. No acceptance verdict is issued.
@@ -0,0 +1,26 @@
+# Repair 28: traffic projection checker
+
+Scope: checker, its Python tests, one minimal decoded fixture and this note. Renderer, Runtime, save paths and controller unchanged. No git writes, Godot/native/OS, Runner or full regression operations.
+
+## Separate routing and defect records
+
+- Normal negative validate 97b6cdf9 failed because the current index was absent. Historical 49 was preserved by canonical rename retaining index SHA c0cc9e…. That missing-current-proof result is not evidence of this static defect.
+- Independent readiness 887f9684 (READINESS_19FC3F5_ACTUAL.json) identifies the actual checker overconstraint. ROOT_19FC3F5_BAND_MISMATCH_REPRO.json and OPUS_19FC3F5_BAND_SCOPE_DIRECTION.json describe its repro and authorized scope.
+
+All seven repair27 input files were read/decoded directly and their byte counts and SHA256 checked against MANIFEST.json (original paths retained in fixture provenance). The original frame-180 landing densities remain zero. The new 45 KB fixture contains only the unchanged decoded fields used by overlay validation: report frame/source metadata, actor drawings, cabins, traffic and camera/elevator track fields; raw world bounds, elevators and staff tasks. It is an extraction, not a raw receipt. No unused megabytes of state, rest or motion were copied. The positive test is explicitly COUNTERFACTUAL: only its in-memory report densities at (0,1,0) and (1,1,0) become 1, covering guests 018/019. It grants no engine/OS acceptance.
+
+## Correction and caller audit
+
+compare_traffic passes its actual snapshot into validate_life_overlay(report, root=ROOT, snapshot=None). inspect_stream already binds that state to its raw object and invokes compare_traffic; native layout and OS writer/reader stream validation therefore receive raw authority. The controller run_p23_life_observation.py has no direct validate_life_overlay/compare_traffic caller. Offline/OS drawing paths already use drawing_context to bind report/state/rest and retain compare_rest, compare_cabin_geometry and compare_actor_geometry. No new drawing gate was added there. Existing direct unit callers without state remain fail closed.
+
+Cabin exemptions apply only to passenger entries belonging to that same raw shaft; cabin geometry reuses compare_cabin_geometry. A floor entry cannot inherit a passenger exemption by copying its key. Carried exemptions require actual active rescue transport binding and reuse compare_actor_geometry for original binding, track and raised-region fitting, excluding boarded/queue placements. Existing rest-authority checks remain unchanged. Missing-carrier KeyError remains deferred and fails closed.
+
+Serial, consumed serial, reported passes, source child/pass order, source connection/parent restrictions, band rectangle and same-report checks remain unchanged. Floor cores keep the same strict intersection predicate. Every intersecting actor contributes a clipped horizontal interval. Sorted interval union must leave a nonzero uncovered span; no ratio or threshold was introduced.
+
+## Local verification
+
+- Python 3 -B compile() of both changed Python files: passed without writing bytecode.
+- Entire checker suite: 108 tests passed (all 99 existing plus 9 new); original R13 row11/29 negative retained. New tests cover counterfactual cabin success, occupied and blocked full union occlusion, spoofed cabin rectangle/membership, absent raw state, floor copy of cabin key, missing carrier binding, and carried binding/track/position mutations. Existing reversed report/source pass tests remain.
+- git diff --check: passed.
+
+Root native/component, A/B on both layouts, normal OS, one full177 and independent Opus all-condition pixel review are expected after submission and do not block worker submission. No such unexecuted checks are claimed passed. Registered timeout 900 s, seven conditions, six categories, 30/88 and formal credit 0 remain unchanged.
@@ -0,0 +1,90 @@
+# Repair 35 — source submission
+
+Scope: P23-03 cabin-intrusion presentation repair only. Source input is
+`93248f4d79a3be71d7a1ac0592e9219d405cd6b7`; recorded tested Q is
+`428f653cdac375c1d4589e6c2d6138e914acf9da`. No Runtime, demand, assets,
+acceptance, policy, process ownership or timeout changes.
+
+## Exact inputs and Phase 0
+
+First read the repair35 `EXACT_INPUTS.json` under
+`/Users/korizatou114/Projects/TheTree-managed-worktrees/p23-03-code/.runtime/P23-03/build/audit-support/cabin-repair35-exact-inputs/`.
+Read its complete supplied `original_full_request.txt`, `actual_Opus_repair.json`
+and `actual_Opus_run.json`; all three supplied SHA256 pins match. Provenance-only
+source locations were not needed. Read the sibling actual-worker preflight.
+
+Read the SHA-verified preserved all87 probe
+`audit-support/cabin-repair34-exact-inputs/CURRENT932_ALL_CABIN_INTRUSION_PROBE.json`
+under the same BUILD (SHA256
+`da395d697e0ec1bd2fb805757aa320314bbc840260f6062d12ecf5d859ceebd0`).
+Read all 87 complete collision rows and all their whole SHA-verified typed
+state/report objects: 19 distinct stream/state/report pairs, 31 distinct objects.
+Also decoded all referenced objects in all 161 current stream captures and read
+every file in the supplied r16 stage (1962 files, including the actual Q sources).
+No required evidence was truncated, compressed, relabelled or edited.
+
+Used that preserved evidence to apply the new candidate-region rule: 70 rows
+select a corridor landing and 17 require identity aggregation because no lane is
+available. All 87 candidate regions intersect a cabin; outside-class rows: zero.
+This is static application to old negative evidence, not new rendered proof.
+
+The existing `API_AND_STATE_MAP.md` remains the deterministic-demand/save-load
+audit. Directly checked current GuestRuntime enum (`MOVING` = 3, lines 6–19),
+WorldService `get_occupant`/`get_entity_cells`/`is_walkable` (194/204/228), and
+ElevatorService's copied public report (345). These read-only APIs supply the
+same queue lane topology and full-width shaft geometry; no new API is needed.
+Current complete dependency SHA256:
+
+| Source | SHA256 |
+| --- | --- |
+| scripts/world/world_service.gd | add67a0799f961a7dce1d382ea6dac67022fbd013f8c4c6b11321df96f37944a |
+| scripts/transport/elevator_service.gd | cea09dd9204551c26b31b30cd3aa847cc77e3b75fdac31662b9694a95fb3580c |
+| scripts/guest/guest_runtime.gd | 57ba99e5d977b5302987f4359a19056f1ab09c8fb7e4762ee8d79cbb9316741a |
+
+The immutable spec matches `8568457d…a9c89f` using the existing checker's
+`read_text().strip()` hash convention. Original acceptance and recorded complete
+Opus repair/run directions remain authoritative.
+
+## Six pose classes
+
+| Case | Product, world_preview.gd | Independent checker |
+| --- | --- | --- |
+| (a) MOVING/idle footprint actor | 1608 candidate region, 1651 cabin rule, 1743 center before fitting | 1532 candidate region, 1577 cabin rule, 1584 center before jitter |
+| (b) Lift-to-corridor walking transit | 1555 merged track bounds, 1565 landing becomes non-transit | 1532 merged track bounds, 1458 landing becomes non-transit |
+| (c) Carrier and raised patient | 1624 recursively relocated carrier region, then minus 12p | 1556 recursively relocated carrier region, then minus 12p; 1612 center |
+| (d) Fractional cabin crossing adjacent floor | 1565 positive-area intersection with actual cabin rect | 1458 intersection after independent cabin/badge validation at 1505 |
+| (e) No corridor lane | 1590 landing badge anchor, 1759 aggregation with every identity | 1482 landing badge anchor, 1703 identity aggregation |
+| (f) Queue plus relocated actor | 1602 prefer unreserved lane, otherwise existing packing | 1487 reservation choice, 1712 existing packing |
+
+Room slots, queue placement/overflow, cabin/passenger geometry and overlay order
+are retained. The new landing centers preserve original jitter; carried patients
+use the relocated carrier before the 12p raise. No-lane actors share a counted
+aggregation at the floor's landing badge anchor, retaining key/id/state/cell/mode.
+Report-only region/boarded assertions cannot authorize a position.
+
+`validate_cabin_intrusion`, `validate_actor_nonoverlap`, `actor_track_transit`
+are byte-identical to the supplied Q checker; `PIXEL_PEOPLE_CORE_SHA256` is
+unchanged. The existing P23-03 marker source-audit SHA alone was updated; all
+other marker rows and suite-policy are preserved.
+
+`run_p23_life_observation.py:315` inspects both completed headless layout streams
+inside the existing exception/receipt/cleanup path, before OS collection.
+Failure therefore prevents OS collection. This is staging, not OS evidence.
+
+## Verification and remaining Root work
+
+- `py_compile` passes for the runner, checker and checker tests.
+- The full Python checker test file passes: 142 tests, including the seven new
+  tests covering (a)–(f), old in-cabin positions, tampered flags, missing actors,
+  missing aggregation identities, combined no-lane identities and unwarranted
+  relocation. Existing 135 tests are retained.
+- Native assertions were extended at `tests/ui/test_p23_life_observation.gd:56`
+  for same-floor and fractional moving cabins on both layers; not executed here.
+- No Godot, native/OS collection, full regression, Git writes, Runner operations
+  or subagents were run. Root native/OS evidence is missing.
+
+Root staging remains: affected native/UI test; typed headless A/B with early
+strict inspection and new-output intrusion probe requiring zero; then current
+S/M/T/Q OS r17; then one final full 179 and independent review. Keep all r16
+negative inputs and the all87 probe versioned. Seven conditions,
+30/88/formal-13 and no partial credit remain unchanged.
@@ -0,0 +1,335 @@
+{
+  "candidate": "336a8cd0b30db1add11869ebceab847db693f531",
+  "r10_path": "/Users/korizatou114/Projects/とりにく工房ゲーム制作部/artifacts/original-completion-2026-10-02/p23-03-native-os-atomic-336a8cd-r10",
+  "terminal_error": "AssertionError: native OS collection failed: AssertionError: ordinary native OS exit0",
+  "cleanup_errors": [],
+  "A_writer_pid": 56892,
+  "A_reader_pid": 60176,
+  "B_writer_pid": 61335,
+  "B_reader": "not run; B writer failed",
+  "all_conditions_accepted": false,
+  "formal_credit": 0,
+  "confirmed_failure": "B construction-only diagnostic budget leaves zero cash; 5 hires and raw-food purchase rejected. Do not change production values or weaken conditions.",
+  "actual_command_objects": {
+    "A": [
+      {
+        "object_sha256": "0d360907ad9289b7db5bc759a220cf274184f18cff3fe98c912941fed22fb1a9",
+        "decoded": {
+          "command": "hire_staff_candidate",
+          "payload": {
+            "candidate_id": "staff_akane"
+          },
+          "result": {
+            "code": "OK",
+            "command_id": 24,
+            "target_id": "",
+            "message": "",
+            "resolution_hints": [],
+            "details": {
+              "candidate_id": "staff_akane",
+              "staff_id": "staff_akane",
+              "role_id": "receptionist",
+              "daily_wage": 120,
+              "employment_kind": "FREE_HIRE",
+              "signing_cost": 180,
+              "shift_id": "shift_00000001_d00000",
+              "player_cash": 1320
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      },
+      {
+        "object_sha256": "6dfd11522dee4b5ac9e1eefad4c2a27f01265cd46f1bb3e364665a85cd78769a",
+        "decoded": {
+          "command": "hire_staff_candidate",
+          "payload": {
+            "candidate_id": "staff_gina"
+          },
+          "result": {
+            "code": "OK",
+            "command_id": 26,
+            "target_id": "",
+            "message": "",
+            "resolution_hints": [],
+            "details": {
+              "candidate_id": "staff_gina",
+              "staff_id": "staff_gina",
+              "role_id": "porter",
+              "daily_wage": 110,
+              "employment_kind": "FREE_HIRE",
+              "signing_cost": 150,
+              "shift_id": "shift_00000003_d00000",
+              "player_cash": 1170
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      },
+      {
+        "object_sha256": "e87798f0030e97a630c76ed3ad9ecdc4e032a090efc5fbb4a39ac606857828dd",
+        "decoded": {
+          "command": "hire_staff_candidate",
+          "payload": {
+            "candidate_id": "staff_chika"
+          },
+          "result": {
+            "code": "OK",
+            "command_id": 28,
+            "target_id": "",
+            "message": "",
+            "resolution_hints": [],
+            "details": {
+              "candidate_id": "staff_chika",
+              "staff_id": "staff_chika",
+              "role_id": "cook",
+              "daily_wage": 145,
+              "employment_kind": "FREE_HIRE",
+              "signing_cost": 250,
+              "shift_id": "shift_00000005_d00000",
+              "player_cash": 920
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      },
+      {
+        "object_sha256": "00c167c28d271fbe133642650ab17b1240e73abd1d5fbb1bcea683f62e603c26",
+        "decoded": {
+          "command": "hire_staff_candidate",
+          "payload": {
+            "candidate_id": "staff_iris"
+          },
+          "result": {
+            "code": "OK",
+            "command_id": 30,
+            "target_id": "",
+            "message": "",
+            "resolution_hints": [],
+            "details": {
+              "candidate_id": "staff_iris",
+              "staff_id": "staff_iris",
+              "role_id": "attendant",
+              "daily_wage": 130,
+              "employment_kind": "FREE_HIRE",
+              "signing_cost": 210,
+              "shift_id": "shift_00000007_d00000",
+              "player_cash": 710
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      },
+      {
+        "object_sha256": "969d0116e1bcc57e28019273e401ff1bbed54ca5ab22455d677889e2375f45f5",
+        "decoded": {
+          "command": "hire_staff_candidate",
+          "payload": {
+            "candidate_id": "staff_juno"
+          },
+          "result": {
+            "code": "OK",
+            "command_id": 32,
+            "target_id": "",
+            "message": "",
+            "resolution_hints": [],
+            "details": {
+              "candidate_id": "staff_juno",
+              "staff_id": "staff_juno",
+              "role_id": "attendant",
+              "daily_wage": 140,
+              "employment_kind": "FREE_HIRE",
+              "signing_cost": 240,
+              "shift_id": "shift_00000009_d00000",
+              "player_cash": 470
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      },
+      {
+        "object_sha256": "092af99c4a2f84786641430e6c1cebce62f813b69dbac5d5ac397a3017fe6a84",
+        "decoded": {
+          "command": "purchase_supplies",
+          "payload": {
+            "warehouse_id": "facility_00000004",
+            "item_id": "ingredient/raw_food",
+            "quantity": 40
+          },
+          "result": {
+            "code": "OK",
+            "command_id": 34,
+            "target_id": "",
+            "message": "",
+            "resolution_hints": [],
+            "details": {
+              "ok": true,
+              "reason": "",
+              "total_cost": 120,
+              "cash_after": 350,
+              "quantity_after": 40,
+              "max_quantity": 100,
+              "warehouse_id": "facility_00000004",
+              "item_id": "ingredient/raw_food",
+              "quantity": 40,
+              "transaction_id": "ledger_00000060"
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      }
+    ],
+    "B": [
+      {
+        "object_sha256": "bcb27db495ae9e94a20578e1aadcb64e0fa316bb1d4a6e954e581f4bdb34b296",
+        "decoded": {
+          "command": "hire_staff_candidate",
+          "payload": {
+            "candidate_id": "staff_akane"
+          },
+          "result": {
+            "code": "VALIDATION_FAILED",
+            "command_id": 25,
+            "target_id": "",
+            "message": "An account does not permit a negative balance.",
+            "resolution_hints": [],
+            "details": {
+              "account_id": "PLAYER_CASH",
+              "available": 0,
+              "resulting_balance": -180
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      },
+      {
+        "object_sha256": "81f9484c3fcce64e084a7e2ead8bf84dd8c56e144860624845eccb8c4218417a",
+        "decoded": {
+          "command": "hire_staff_candidate",
+          "payload": {
+            "candidate_id": "staff_gina"
+          },
+          "result": {
+            "code": "VALIDATION_FAILED",
+            "command_id": 27,
+            "target_id": "",
+            "message": "An account does not permit a negative balance.",
+            "resolution_hints": [],
+            "details": {
+              "account_id": "PLAYER_CASH",
+              "available": 0,
+              "resulting_balance": -150
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      },
+      {
+        "object_sha256": "16f4a094baa4204d4e254b6d52d824f32aa33e914e6bc1698aa933a8fa51207a",
+        "decoded": {
+          "command": "hire_staff_candidate",
+          "payload": {
+            "candidate_id": "staff_chika"
+          },
+          "result": {
+            "code": "VALIDATION_FAILED",
+            "command_id": 29,
+            "target_id": "",
+            "message": "An account does not permit a negative balance.",
+            "resolution_hints": [],
+            "details": {
+              "account_id": "PLAYER_CASH",
+              "available": 0,
+              "resulting_balance": -250
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      },
+      {
+        "object_sha256": "303e545649bf1c84c736293a6e1b54491befbc376b9a051d3f3c75659e2b147e",
+        "decoded": {
+          "command": "hire_staff_candidate",
+          "payload": {
+            "candidate_id": "staff_iris"
+          },
+          "result": {
+            "code": "VALIDATION_FAILED",
+            "command_id": 31,
+            "target_id": "",
+            "message": "An account does not permit a negative balance.",
+            "resolution_hints": [],
+            "details": {
+              "account_id": "PLAYER_CASH",
+              "available": 0,
+              "resulting_balance": -210
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      },
+      {
+        "object_sha256": "de1717d64250c9ae721fc38230d185b0621a066e01890f9fb4020b4c7e36e7c0",
+        "decoded": {
+          "command": "hire_staff_candidate",
+          "payload": {
+            "candidate_id": "staff_juno"
+          },
+          "result": {
+            "code": "VALIDATION_FAILED",
+            "command_id": 33,
+            "target_id": "",
+            "message": "An account does not permit a negative balance.",
+            "resolution_hints": [],
+            "details": {
+              "account_id": "PLAYER_CASH",
+              "available": 0,
+              "resulting_balance": -240
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      },
+      {
+        "object_sha256": "c257c14418959e558b88d4f47591c14726fff4763a30177275d8725cb6639b9f",
+        "decoded": {
+          "command": "purchase_supplies",
+          "payload": {
+            "warehouse_id": "facility_00000004",
+            "item_id": "ingredient/raw_food",
+            "quantity": 40
+          },
+          "result": {
+            "code": "VALIDATION_FAILED",
+            "command_id": 35,
+            "target_id": "",
+            "message": "Supply command rejected: INSUFFICIENT_FUNDS",
+            "resolution_hints": [],
+            "details": {
+              "ok": false,
+              "reason": "INSUFFICIENT_FUNDS",
+              "total_cost": 120,
+              "cash_after": -120,
+              "quantity_after": 40,
+              "max_quantity": 0
+            },
+            "is_replay": false
+          },
+          "minute": 910
+        }
+      }
+    ]
+  },
+  "next": "Actual Opus diagnosis and narrow same-task repair before full regression"
+}
@@ -79,6 +79,8 @@ var _queue_hidden: Dictionary = {}
 var _landing_badges: Array[Dictionary] = []
 var _transport_report: Dictionary = {}
 var _life_observation: Dictionary = {}
+var _life_overlay: Node2D
+var _life_observation_serial := 0
 var _dragging := false
 var _drag_button := MOUSE_BUTTON_NONE

@@ -104,6 +106,10 @@ func _ready() -> void:
 	_inactive_layer_art = InactiveLayerRooms.new()
 	_inactive_layer_art.name = "InactiveLayerRooms"
 	add_child(_inactive_layer_art)
+	_life_overlay = Node2D.new()
+	_life_overlay.name = "LifeOverlay"
+	_life_overlay.draw.connect(_draw_life_overlay)
+	add_child(_life_overlay)
 	_layer_lens = ColorRect.new()
 	_layer_lens.name = "LayerLens"
 	_layer_lens.mouse_filter = Control.MOUSE_FILTER_IGNORE
@@ -513,6 +519,8 @@ func _draw() -> void:

 func _draw_contents() -> void:
 	_life_observation = {}
+	if _life_overlay != null:
+		_life_overlay.queue_redraw()
 	if _inactive_layer_art != null:
 		_inactive_layer_art.visible = false
 	_draw_background()
@@ -524,6 +532,9 @@ func _draw_contents() -> void:
 	if map_rect.size == Vector2.ZERO:
 		return
 	_life_observation = _build_life_observation()
+	_life_observation_serial += 1
+	_life_observation[&"observation_serial"] = _life_observation_serial
+	_life_observation[&"overlay_consumed_serial"] = -1
 	_room_presentations = _life_observation[&"rooms"]
 	_carried_patients = _life_observation[&"carried_patients"]
 	_treated_patients = _life_observation[&"treated_patients"]
@@ -538,13 +549,9 @@ func _draw_contents() -> void:
 	_draw_world_cells(bounds, map_rect)
 	_draw_world_entity_icons()
 	_draw_legacy_stairs()
-	_draw_elevators()
 	_draw_other_layer()
-	_draw_congestion_overlay(bounds)
-	_draw_life_badges()
 	_draw_construction_labels()
 	_draw_facility_labels()
-	_draw_actors()
 	_draw_excavation_jobs()
 	_draw_build_preview()
 	_draw_excavation_preview()
@@ -599,7 +606,8 @@ func _build_life_observation() -> Dictionary:
 	model[&"guests"] = guests
 	model[&"staff"] = staff
 	model[&"tasks"] = tasks
-	var bridge: Variant = _service(&"supply_logistics_bridge")
+	_fit_life_actors(model)
+	var bridge: Variant = _service(&"supply_logistics")
 	var goods: Dictionary = bridge.get_report() if bridge != null else {&"requests": [], &"deliveries": []}
 	model[&"goods"] = goods
 	var badges: Array[Dictionary] = []
@@ -629,6 +637,14 @@ func _build_life_observation() -> Dictionary:
 					&"at": rect.position + Vector2(2, 16), &"requests": requests,
 					&"text": RoomPresentation.supply_label(requests)})
 	model[&"badges"] = badges
+	var obstacles: Array[Rect2] = []
+	for cabin: Dictionary in model[&"cabins"]:
+		obstacles.append(cabin[&"rect"])
+		obstacles.append(cabin[&"badge_rect"])
+	for collection: Array in [badges, model[&"landing_badges"], model.get(&"actor_aggregations", [])]:
+		for badge: Dictionary in collection:
+			obstacles.append(_readable_badge_rect(badge[&"at"], badge[&"text"]))
+	model[&"traffic_obstacles"] = obstacles
 	var traffic: Array[Dictionary] = []
 	var path: Variant = _service(&"path_service")
 	var bounds := _world_bounds()
@@ -638,14 +654,46 @@ func _build_life_observation() -> Dictionary:
 				var cell := Vector3i(x, y, layer_mode)
 				var row: Dictionary = path.get_cell_traffic(cell)
 				if not row.is_empty():
-					traffic.append({&"cell": cell, &"traffic": row, &"rect": get_cell_rect(cell).grow(-2.0)})
+					var cell_rect := get_cell_rect(cell)
+					var rect := cell_rect.grow(-2.0)
+					var world_pixel := _cell_dimensions().x / 16.0
+					var thickness := clampf(rect.size.y * 0.16, 2.0, 5.0)
+					var band := Rect2(Vector2(rect.position.x + rect.size.x * 0.12,
+						cell_rect.end.y - 2 * world_pixel - thickness * 0.5),
+						Vector2(rect.size.x * 0.76, thickness))
+					traffic.append({&"cell": cell, &"traffic": row, &"rect": rect, &"band_rect": band,
+						&"segment_rects": _traffic_segments(band, rect, obstacles)})
 	model[&"traffic"] = traffic
 	return model


-func _draw_life_badges() -> void:
+## Subtract opaque observation regions, retaining a visible density mark even
+## when the ordinary floor band is completely behind a moving cabin.
+func _traffic_segments(band: Rect2, cell: Rect2, obstacles: Array[Rect2]) -> Array[Rect2]:
+	for y: float in [band.position.y, cell.position.y, cell.end.y - band.size.y]:
+		var pieces: Array[Rect2] = [Rect2(Vector2(band.position.x, y), band.size)]
+		for obstacle: Rect2 in obstacles:
+			var remaining: Array[Rect2] = []
+			for piece: Rect2 in pieces:
+				var cut := piece.intersection(obstacle)
+				if not cut.has_area():
+					remaining.append(piece)
+					continue
+				for part: Rect2 in [Rect2(piece.position, Vector2(piece.size.x, cut.position.y-piece.position.y)),
+					Rect2(Vector2(piece.position.x, cut.end.y), Vector2(piece.size.x, piece.end.y-cut.end.y)),
+					Rect2(Vector2(piece.position.x, cut.position.y), Vector2(cut.position.x-piece.position.x, cut.size.y)),
+					Rect2(Vector2(cut.end.x, cut.position.y), Vector2(piece.end.x-cut.end.x, cut.size.y))]:
+					if part.has_area():
+						remaining.append(part)
+			pieces = remaining
+		if not pieces.is_empty():
+			return pieces
+	return []
+
+
+func _draw_life_badges(canvas: CanvasItem) -> void:
 	for badge: Dictionary in _life_observation.get(&"badges", []):
-		_draw_readable_badge(badge[&"at"], badge[&"text"])
+		_draw_readable_badge(badge[&"at"], badge[&"text"], canvas)


 func _draw_background() -> void:
@@ -758,9 +806,29 @@ func _draw_leaf_cluster(center: Vector2, p: float, variation: int) -> void:
 		draw_rect(Rect2(c+offset,Vector2(3+i%2,2)*p),Color("95ad69") if i%3==0 else Color("648957"))


-func _draw_congestion_overlay(_bounds: Dictionary) -> void:
+func _draw_life_overlay() -> void:
+	if _life_observation.is_empty():
+		return
+	# These are actual child indices and the passes executed by this callback.
+	_life_observation[&"overlay_child_indices"] = [_inactive_layer_art.get_index(), _life_overlay.get_index(), _layer_lens.get_index()]
+	_life_observation[&"overlay_passes"] = []
+	_life_observation[&"overlay_passes"].append(&"traffic")
+	_draw_congestion_overlay(_life_overlay)
+	_life_observation[&"overlay_passes"].append(&"cabins")
+	_draw_elevators(_life_overlay)
+	_draw_life_badges(_life_overlay)
+	_life_observation[&"overlay_passes"].append(&"actors")
+	for cabin: Dictionary in _life_observation.get(&"cabins", []):
+		for actor: Dictionary in cabin[&"passengers"]:
+			_draw_observed_actor(actor, _life_overlay)
+	_draw_actors(_life_overlay)
+	for badge: Dictionary in _life_observation.get(&"actor_aggregations", []):
+		_draw_readable_badge(badge[&"at"], badge[&"text"], _life_overlay)
+	_life_observation[&"overlay_consumed_serial"] = _life_observation[&"observation_serial"]
+
+
+func _draw_congestion_overlay(canvas: CanvasItem) -> void:
 	for row: Dictionary in _life_observation.get(&"traffic", []):
-		var rect: Rect2 = row[&"rect"]
 		var traffic: Dictionary = row[&"traffic"]
 		var density := int(traffic[&"current_density"])
 		var blocked := bool(traffic[&"temporary_block"])
@@ -768,13 +836,13 @@ func _draw_congestion_overlay(_bounds: Dictionary) -> void:
 			continue
 		var ratio := float(density) / maxi(int(traffic[&"base_capacity"]), 1)
 		var color := Color("e34332") if blocked or ratio > 1.0 else Color("e3a62f") if ratio >= 0.65 else Color("55c96b")
-		var center := rect.get_center()
-		draw_line(Vector2(rect.position.x + rect.size.x * 0.12, center.y),
-			Vector2(rect.end.x - rect.size.x * 0.12, center.y),
-			Color(color, 0.90 if show_grid else 0.28), clampf(rect.size.y * 0.16, 2.0, 5.0))
-		if blocked:
-			draw_line(rect.position, rect.end, Color(color, 0.90 if show_grid else 0.28), 2.0)
-			draw_line(Vector2(rect.end.x, rect.position.y), Vector2(rect.position.x, rect.end.y), Color(color, 0.90 if show_grid else 0.28), 2.0)
+		# At overview zoom a whole cell can be covered by opaque badges. Still
+		# submit its density band, beneath those later draws, and report no visible
+		# fragments. No traffic cell or raw density row is dropped.
+		if row[&"segment_rects"].is_empty():
+			canvas.draw_rect(row[&"band_rect"], Color(color, 0.90))
+		for segment: Rect2 in row[&"segment_rects"]:
+			canvas.draw_rect(segment, Color(color, 0.90))


 ## Visual-only level comes from the existing live FacilityInstance, not the
@@ -887,13 +955,20 @@ func _draw_world_entity_icons() -> void:
 			draw_rect(rect.grow(-p),Color("f5d582"),false,2*p)


-func _draw_readable_badge(at: Vector2, text: String) -> void:
-	var width := ThemeDB.fallback_font.get_string_size(text, HORIZONTAL_ALIGNMENT_LEFT, -1, 10).x+6
-	var rect := Rect2(at, Vector2(width, 14))
+func _readable_badge_rect(at: Vector2, text: String) -> Rect2:
+	# A conservative em-width box makes geometry independently reproducible.
+	# Text still uses the existing fallback font and size.
+	return Rect2(at, Vector2(text.length() * 10 + 6, 14))
+
+
+func _draw_readable_badge(at: Vector2, text: String, canvas: CanvasItem = null) -> void:
+	if canvas == null:
+		canvas = self
+	var rect := _readable_badge_rect(at, text)
 	if not rect.intersects(Rect2(Vector2.ZERO, size)):
 		return
-	draw_rect(rect, Color("34413b", 0.94))
-	draw_string(ThemeDB.fallback_font, at+Vector2(3, 11), text, HORIZONTAL_ALIGNMENT_LEFT, -1, 10, Color("fff2ce"))
+	canvas.draw_rect(rect, Color("34413b"))
+	canvas.draw_string(ThemeDB.fallback_font, at+Vector2(3, 11), text, HORIZONTAL_ALIGNMENT_LEFT, -1, 10, Color("fff2ce"))


 func _prepare_room_presentations() -> void:
@@ -982,7 +1057,8 @@ func _observe_queues() -> Dictionary:
 	var lift_cells: Dictionary = {}
 	for shaft: Dictionary in _transport_report.get(&"shafts", []):
 		for stop: Vector3i in _cells_from_serialized(shaft.get(&"stops", [])):
-			lift_cells[stop] = true
+			for offset: int in range(maxi(int(shaft.get(&"width", 2)), 1)):
+				lift_cells[stop + Vector3i(offset, 0, 0)] = true
 	var reserved: Dictionary = {}
 	var cells: Array = groups.keys()
 	cells.sort_custom(func(a: Vector3i, b: Vector3i) -> bool: return a.y < b.y or (a.y == b.y and a.x < b.x))
@@ -997,6 +1073,8 @@ func _observe_queues() -> Dictionary:
 				var next := cell + Vector3i(direction*distance, 0, 0)
 				if world == null or not world.is_walkable(next):
 					break
+				if lift_cells.has(next):
+					break
 				var occupant: StringName = world.get_occupant(next)
 				var facility: Variant = facilities.get_instance(occupant) if facilities != null else null
 				if facility != null and facility.definition_id not in [&"guest_corridor", &"service_corridor"]:
@@ -1016,7 +1094,7 @@ func _observe_queues() -> Dictionary:
 		var text := RoomPresentation.queue_label(members.size(), members.size()-index)
 		_landing_badges.append({&"at": get_cell_rect(cell).position + Vector2(-80 if lift_cells.has(cell) else 0, -16), &"text": text, &"cell": cell, &"members": members.duplicate(), &"visible_count": index, &"overflow": members.size()-index})

-	return {&"queue_positions": _queue_positions, &"queue_hidden": _queue_hidden, &"landing_badges": _landing_badges, &"elevators": _transport_report}
+	return {&"queue_positions": _queue_positions, &"queue_hidden": _queue_hidden, &"landing_badges": _landing_badges, &"elevators": _transport_report, &"shaft_footprint": lift_cells.keys()}


 func _add_queue_member(groups: Dictionary, key: String, cell: Vector3i) -> void:
@@ -1143,15 +1221,13 @@ func _draw_facility_labels() -> void:
 		draw_string(ThemeDB.fallback_font,at+Vector2(4,12),label,HORIZONTAL_ALIGNMENT_CENTER,label_width-8,font_size,Color("f0e2b9"))


-func _draw_elevators() -> void:
+func _draw_elevators(canvas: CanvasItem) -> void:
 	for row: Dictionary in _life_observation.get(&"cabins", []):
-		PixelArt.elevator_cabin(self, row[&"rect"], row[&"pixel"], layer_mode == LayerMode.BACK, row[&"operational"])
-		for actor: Dictionary in row[&"passengers"]:
-			_draw_observed_actor(actor)
-		PixelArt.elevator_gate(self, row[&"rect"], row[&"pixel"], row[&"doors_open"])
-		_draw_readable_badge(row[&"badge_at"], row[&"badge"])
+		PixelArt.elevator_cabin(canvas, row[&"rect"], row[&"pixel"], layer_mode == LayerMode.BACK, row[&"operational"])
+		PixelArt.elevator_gate(canvas, row[&"rect"], row[&"pixel"], row[&"doors_open"])
+		_draw_readable_badge(row[&"badge_at"], row[&"badge"], canvas)
 	for badge: Dictionary in _life_observation.get(&"landing_badges", []):
-		_draw_readable_badge(badge[&"at"], badge[&"text"])
+		_draw_readable_badge(badge[&"at"], badge[&"text"], canvas)


 func _observe_elevators(observation: Dictionary) -> Array[Dictionary]:
@@ -1188,8 +1264,7 @@ func _observe_elevators(observation: Dictionary) -> Array[Dictionary]:
 		var doors_open := operational and phase in [&"IDLE", &"BOARDING", &"ALIGHTING"]
 		var passengers_drawn: Array[Dictionary] = []
 		var seat_columns := maxi(int((cabin.size.x / p - 4.0) / 8.0), 1)
-		var visible_count := mini(passengers.size(), seat_columns*2)
-		for index: int in range(visible_count):
+		for index: int in range(passengers.size()):
 			var actor_key := _elevator_actor_key(String(passengers[index]))
 			var is_staff := actor_key.begins_with("staff/")
 			var person: Variant = null
@@ -1200,13 +1275,15 @@ func _observe_elevators(observation: Dictionary) -> Array[Dictionary]:
 			var carrying := is_staff and person != null and not String(person.carried_inventory_id).is_empty()
 			var state := int(person.state) if person != null else 0
 			var x := cabin.position.x + (2.0 + (index % seat_columns + 0.5)*8.0)*p
-			var feet := cabin.end.y - (2.0 + int(index / seat_columns)*9.0)*p
+			var feet := cabin.end.y - (2.0 + (int(index / seat_columns) % 2)*9.0)*p
 			var center := Vector2(x, feet+3*p-_cell_dimensions().y*0.5)
 			passengers_drawn.append(_actor_drawing(center,actor_key,is_staff,state,false,carrying,&"",true,observation))
 		cabins.append({&"shaft_id": shaft_id, &"rect": cabin, &"pixel": p,
 			&"operational": operational, &"doors_open": doors_open,
+			&"phase": phase, &"passenger_count": passengers.size(),
 			&"passengers": passengers_drawn,
 			&"badge_at": Vector2(cabin.position.x, cabin.position.y-14),
+			&"badge_rect": _readable_badge_rect(Vector2(cabin.position.x, cabin.position.y-14), RoomPresentation.elevator_label(shaft)),
 			&"badge": RoomPresentation.elevator_label(shaft)})

 	for shaft_id: Variant in _elevator_tracks.keys():
@@ -1240,9 +1317,9 @@ func _tracked_elevator_floor(shaft_id: StringName, current_floor: float, observa
 	return lerpf(float(track[&"from"]),current_floor,t)


-func _draw_actors() -> void:
+func _draw_actors(canvas: CanvasItem = null) -> void:
 	for row: Dictionary in _life_observation.get(&"actor_drawings", []):
-		_draw_observed_actor(row)
+		_draw_observed_actor(row, canvas)


 func _observe_actors(observation: Dictionary) -> Array[Dictionary]:
@@ -1291,6 +1368,9 @@ func _observe_actors(observation: Dictionary) -> Array[Dictionary]:
 				continue
 			var position := _tracked_actor_position(key,staff.current_cell,observation)
 			if staff.current_cell.z==layer_mode:
+				if _queue_positions.has(key):
+					rows.append(_actor_drawing(_queue_positions[key],key,true,int(staff.state),false,not String(staff.carried_inventory_id).is_empty(),_staff_work_animation(staff),false,observation))
+					continue
 				if _carried_patients.has(staff.staff_id):
 					var carrier: Variant = staff_service.get_staff(_carried_patients[staff.staff_id])
 					if carrier != null:
@@ -1412,11 +1492,13 @@ func _draw_pixel_actor(center: Vector2, key: String, staff: bool, state: int, re
 	_draw_observed_actor(_actor_drawing(center, key, staff, state, receiving, carrying, activity, riding))


-func _draw_observed_actor(row: Dictionary) -> void:
+func _draw_observed_actor(row: Dictionary, canvas: CanvasItem = null) -> void:
+	if canvas == null:
+		canvas = self
 	var at: Vector2 = row[&"at"]
 	var p: float = row[&"scale"]
-	draw_rect(Rect2(at + Vector2(5, 17) * p, Vector2(8, 1) * p), Color("303c40", 0.2))
-	People.draw(self, at, p, row[&"mode"], row[&"frame"], row[&"variant"], row[&"staff"], row[&"flip"])
+	canvas.draw_rect(Rect2(at + Vector2(5, 17) * p, Vector2(8, 1) * p), Color("303c40", 0.2))
+	People.draw(canvas, at, p, row[&"mode"], row[&"frame"], row[&"variant"], row[&"staff"], row[&"flip"])


 func _actor_drawing(center: Vector2, key: String, staff: bool, state: int, receiving: bool, carrying: bool, activity: StringName = &"", riding := false, observation: Dictionary = {}) -> Dictionary:
@@ -1454,6 +1536,270 @@ func _actor_drawing(center: Vector2, key: String, staff: bool, state: int, recei
 		&"riding": riding, &"flip": float(track.get(&"direction", 1.0)) < 0.0}


+## Identity cores are tied to the unchanged PixelPeople poses, not sprite boxes.
+func _life_core(row: Dictionary) -> Rect2:
+	var phase := int(row[&"frame"]) % 8
+	var core := Rect2(6, 1, 6, 12)
+	if row[&"mode"] in [&"sleep", &"collapsed", &"standard_room_sleep"]:
+		core = Rect2(2, 11, 13, 5)
+		if row[&"mode"] == &"standard_room_sleep" and phase in [2, 3, 4]:
+			core.position.y -= 1
+			core.size.y += 1
+	elif row[&"mode"] in [&"walk", &"carry_walk"] and phase in [2, 6]:
+		core.position.y -= 1
+	if row[&"flip"]:
+		core.position.x = 16-core.end.x
+	return Rect2(row[&"at"]+core.position*row[&"scale"], core.size*row[&"scale"])
+
+
+func _life_track_region(key: String, cell: Vector3i, model: Dictionary) -> Dictionary:
+	var track: Dictionary = model[&"actor_tracks"].get(key, {})
+	var transit: bool = not model[&"paused"] and track.get(&"from", cell) != cell and _animation_time-float(track.get(&"at", 0.0)) < 0.42
+	var rect := get_cell_rect(cell)
+	if transit:
+		rect = rect.merge(get_cell_rect(track[&"from"]))
+	return {&"bounds": rect, &"transit": transit}
+
+
+## Keep non-passengers on a real corridor landing when a cabin crosses their pose.
+func _life_landing_region(region: Dictionary, cell: Vector3i, model: Dictionary) -> Dictionary:
+	var blocked := false
+	for cabin: Dictionary in model[&"cabins"]:
+		if (region[&"bounds"] as Rect2).intersects(cabin[&"rect"]):
+			blocked = true
+			break
+	if not blocked:
+		return region
+	var world: Variant = _service(&"world_service")
+	var facilities: Variant = _service(&"facility_service")
+	var footprint: Array = model[&"shaft_footprint"]
+	var origins: Array = footprint.filter(func(c: Vector3i) -> bool: return c.y == cell.y and c.z == cell.z)
+	origins.sort_custom(func(a: Vector3i, b: Vector3i) -> bool: return absi(a.x-cell.x) < absi(b.x-cell.x) or (absi(a.x-cell.x) == absi(b.x-cell.x) and a.x < b.x))
+	var lane: Array[Vector3i] = []
+	for origin: Vector3i in origins:
+		for direction: int in [-1, 1]:
+			for distance: int in range(1, 9):
+				var target := origin + Vector3i(direction*distance, 0, 0)
+				if world == null or footprint.has(target) or not world.is_walkable(target):
+					break
+				var facility: Variant = facilities.get_instance(world.get_occupant(target)) if facilities != null else null
+				if facility != null and facility.definition_id not in [&"guest_corridor", &"service_corridor"]:
+					break
+				if not lane.has(target):
+					lane.append(target)
+	lane.sort_custom(func(a: Vector3i, b: Vector3i) -> bool: return absi(a.x-cell.x) < absi(b.x-cell.x) or (absi(a.x-cell.x) == absi(b.x-cell.x) and a.x < b.x))
+	if lane.is_empty():
+		var anchor := get_cell_rect(cell).position+Vector2(-80, -16)
+		var nearest := INF
+		for badge: Dictionary in model[&"landing_badges"]:
+			var landing: Vector3i = badge[&"cell"]
+			if landing.y == cell.y and landing.z == cell.z and absf(landing.x-cell.x) < nearest:
+				nearest = absf(landing.x-cell.x)
+				anchor = badge[&"at"]
+		return {&"bounds": region[&"bounds"], &"transit": false, &"no_lane": true, &"badge_at": anchor}
+	var selected: Vector3i = lane[0]
+	for target: Vector3i in lane:
+		if not model[&"queue_positions"].values().has(cell_to_canvas(target)):
+			selected = target
+			break
+	return {&"bounds": get_cell_rect(selected), &"transit": false, &"center": cell_to_canvas(selected)}
+
+
+func _life_actor_region(row: Dictionary, person: Dictionary, model: Dictionary) -> Dictionary:
+	var key: String = row[&"key"]
+	var cell: Vector3i = person[&"cell"]
+	var result := _life_track_region(key, cell, model)
+	var p := _cell_dimensions().x/16.0
+	var world: Variant = _service(&"world_service")
+	var facility_id: StringName = &""
+	if model[&"queue_positions"].has(key):
+		result = {&"bounds": Rect2(model[&"queue_positions"][key]-_cell_dimensions()*0.5, _cell_dimensions()), &"transit": false}
+	elif not row[&"staff"]:
+		if not String(person[&"service"]).is_empty():
+			var room: Dictionary = model[&"rooms"].get(person[&"target"], {})
+			if (room.get(&"people", []) as Array).has(person[&"id"]):
+				facility_id = person[&"target"]
+	else:
+		var id: StringName = person[&"id"]
+		if model[&"carried_patients"].has(id):
+			for carrier: Dictionary in model[&"staff"]:
+				if carrier[&"id"] == model[&"carried_patients"][id]:
+					result = _life_actor_region({&"key": "staff/"+String(carrier[&"id"]), &"staff": true}, carrier, model)
+					var rect: Rect2 = result[&"bounds"]
+					rect = rect.grow(p)
+					rect.position.y -= 12*p
+					result[&"bounds"] = rect
+					if result.has(&"center"):
+						result[&"center"] -= Vector2(0, 12*p)
+		elif model[&"resting_staff"].has(id):
+			facility_id = model[&"resting_staff"][id]
+		elif model[&"treated_patients"].has(id):
+			facility_id = model[&"treated_patients"][id]
+		elif person[&"state"] in ["WORKING", "AVAILABLE"] and world != null:
+			var cells: Array[Vector3i] = world.get_entity_cells(person[&"facility"])
+			var room: Dictionary = model[&"rooms"].get(person[&"facility"], {})
+			var slot := (room.get(&"staff", []) as Array).find(id)
+			var rect := _cell_bounds_rect(cells)
+			var columns := maxi(int(rect.size.x/(16*p)), 1)
+			if cells.has(cell) and slot >= 0 and slot < columns and rect.size.y >= 40*p:
+				result = {&"bounds": Rect2(rect.position+Vector2(slot*rect.size.x/columns, 0), Vector2(rect.size.x/columns, rect.size.y)), &"transit": false}
+	if not facility_id.is_empty() and world != null:
+		var cells: Array[Vector3i] = world.get_entity_cells(facility_id)
+		if not cells.is_empty():
+			result = {&"bounds": _cell_bounds_rect(cells), &"transit": false}
+	if not model[&"queue_positions"].has(key) and not result.has(&"center") and not result.get(&"no_lane", false):
+		result = _life_landing_region(result, cell, model)
+	result[&"bounds"] = (result[&"bounds"] as Rect2).grow(-p)
+	return result
+
+
+func _life_scaled(row: Dictionary, scale: float) -> Dictionary:
+	var result := row.duplicate(true)
+	result[&"scale"] = scale
+	result[&"at"] = (row[&"center"]+Vector2(-8*scale, _cell_dimensions().y*0.5-3*_cell_dimensions().x/16.0-17*scale)).round()
+	return result
+
+
+## Forward/backward packing keeps each member's own interval. No relocation.
+func _life_pack(group: Array, scale: float) -> Array:
+	var trial: Array = []
+	var right := -INF
+	for member: Dictionary in group:
+		var row := _life_scaled(member[&"row"], minf(scale, member[&"row"][&"scale"]))
+		var core := _life_core(row)
+		var bounds: Rect2 = member[&"bounds"]
+		var x := maxf(maxf(core.position.x, bounds.position.x), right)
+		row[&"at"].x += x-core.position.x
+		right = x+core.size.x
+		trial.append(row)
+	var left := INF
+	for i: int in range(trial.size()-1, -1, -1):
+		var row: Dictionary = trial[i]
+		var core := _life_core(row)
+		var bounds: Rect2 = group[i][&"bounds"]
+		var x := minf(core.position.x, minf(bounds.end.x, left)-core.size.x)
+		row[&"at"].x += x-core.position.x
+		left = x
+		if x < bounds.position.x or core.position.y < bounds.position.y or core.end.y > bounds.end.y:
+			return []
+	return trial
+
+
+## Rebuild vertical components at every fitted scale. Ceilings only decrease;
+## once none decreases, the feet/cores and therefore the partition are fixed.
+func _life_fitted_groups(members: Array) -> Array:
+	var groups: Array = []
+	while true:
+		for member: Dictionary in members:
+			member[&"row"].merge(_life_scaled(member[&"row"], member[&"row"][&"scale"]), true)
+		members.sort_custom(func(a: Dictionary, b: Dictionary) -> bool:
+			var ay := _life_core(a[&"row"]).position.y
+			var by := _life_core(b[&"row"]).position.y
+			return ay < by or (ay == by and a[&"row"][&"key"] < b[&"row"][&"key"]))
+		groups = []
+		var bottom := -INF
+		for member: Dictionary in members:
+			var core := _life_core(member[&"row"])
+			if groups.is_empty() or core.position.y >= bottom:
+				groups.append([])
+				bottom = core.end.y
+			else:
+				bottom = maxf(bottom, core.end.y)
+			groups.back().append(member)
+		var reduced := false
+		for group: Array in groups:
+			group.sort_custom(func(a: Dictionary, b: Dictionary) -> bool: return a[&"row"][&"center"].x < b[&"row"][&"center"].x or (a[&"row"][&"center"].x == b[&"row"][&"center"].x and a[&"row"][&"key"] < b[&"row"][&"key"]))
+			var ceiling := 1.0
+			for member: Dictionary in group:
+				ceiling = maxf(ceiling, member[&"row"][&"scale"])
+			if _life_pack(group, ceiling).is_empty() and ceiling > 1:
+				for member: Dictionary in group:
+					var scale := minf(member[&"row"][&"scale"], maxf(1, int(ceiling)-1))
+					member[&"row"].merge(_life_scaled(member[&"row"], scale), true)
+				reduced = true
+		if not reduced:
+			return groups
+	return groups
+
+
+func _fit_life_actors(model: Dictionary) -> void:
+	var people: Dictionary = {}
+	for person: Dictionary in model[&"guests"]:
+		people["guest/"+String(person[&"id"])] = person
+	for person: Dictionary in model[&"staff"]:
+		people["staff/"+String(person[&"id"])] = person
+	var members: Array = []
+	var stranded: Array = []
+	var all_rows: Array = model[&"actor_drawings"].duplicate()
+	var passenger_bounds: Dictionary = {}
+	for cabin: Dictionary in model[&"cabins"]:
+		for row: Dictionary in cabin[&"passengers"]:
+			all_rows.append(row)
+			passenger_bounds[row[&"key"]] = (cabin[&"rect"] as Rect2).grow(-_cell_dimensions().x/16.0)
+	for row: Dictionary in all_rows:
+		var person: Dictionary = people[row[&"key"]]
+		row[&"cell"] = person[&"cell"]
+		var region := _life_actor_region(row, person, model) if not row[&"riding"] else {&"bounds": passenger_bounds[row[&"key"]], &"transit": false}
+		if region.has(&"center"):
+			var jitter := Vector2.ZERO
+			if not model[&"carried_patients"].has(person[&"id"]):
+				var track: Dictionary = model[&"actor_tracks"][row[&"key"]]
+				var t := 1.0 if model[&"paused"] else clampf((_animation_time-float(track[&"at"]))/0.42, 0, 1)
+				jitter = row[&"center"]-cell_to_canvas(track[&"from"]).lerp(cell_to_canvas(person[&"cell"]), t)
+			row[&"center"] = region[&"center"]+jitter
+			row.merge(_life_scaled(row, row[&"scale"]), true)
+		region[&"row"] = row
+		if region.get(&"no_lane", false):
+			stranded.append(region)
+		elif not region[&"transit"]:
+			members.append(region)
+	var groups := _life_fitted_groups(members)
+	model[&"actor_aggregations"] = []
+	var hidden: Dictionary = {}
+	for member: Dictionary in stranded:
+		var row: Dictionary = member[&"row"]
+		hidden[row[&"key"]] = true
+		var aggregate: Dictionary = {}
+		for existing: Dictionary in model[&"actor_aggregations"]:
+			if existing[&"at"] == member[&"badge_at"]:
+				aggregate = existing
+				break
+		if aggregate.is_empty():
+			aggregate = {&"members": [], &"drawn_keys": [], &"at": member[&"badge_at"], &"text": ""}
+			model[&"actor_aggregations"].append(aggregate)
+		aggregate[&"members"].append({&"key": row[&"key"], &"id": people[row[&"key"]][&"id"], &"state": row[&"state"], &"cell": row[&"cell"], &"mode": row[&"mode"]})
+		aggregate[&"text"] = "+%d" % aggregate[&"members"].size()
+	for group: Array in groups:
+		group.sort_custom(func(a: Dictionary, b: Dictionary) -> bool: return a[&"row"][&"center"].x < b[&"row"][&"center"].x or (a[&"row"][&"center"].x == b[&"row"][&"center"].x and a[&"row"][&"key"] < b[&"row"][&"key"]))
+		var base := 1.0
+		for member: Dictionary in group:
+			base = maxf(base, member[&"row"][&"scale"])
+		var fitted := _life_pack(group, base)
+		var retained: Array = group
+		if fitted.is_empty():
+			retained = []
+			var identities: Array = []
+			var drawn: Array = []
+			for member: Dictionary in group:
+				var row: Dictionary = member[&"row"]
+				identities.append({&"key": row[&"key"], &"id": people[row[&"key"]][&"id"], &"state": row[&"state"], &"cell": row[&"cell"], &"mode": row[&"mode"]})
+				var attempt := retained.duplicate()
+				attempt.append(member)
+				var packed := _life_pack(attempt, 1)
+				if not packed.is_empty():
+					retained = attempt
+					fitted = packed
+					drawn.append(row[&"key"])
+				else:
+					hidden[row[&"key"]] = true
+			model[&"actor_aggregations"].append({&"members": identities, &"drawn_keys": drawn, &"at": (group[0][&"bounds"] as Rect2).position, &"text": "+%d" % (group.size()-retained.size())})
+		for i: int in fitted.size():
+			retained[i][&"row"].merge(fitted[i], true)
+	model[&"actor_drawings"] = model[&"actor_drawings"].filter(func(row: Dictionary) -> bool: return not hidden.has(row[&"key"]))
+	for cabin: Dictionary in model[&"cabins"]:
+		cabin[&"passengers"] = cabin[&"passengers"].filter(func(row: Dictionary) -> bool: return not hidden.has(row[&"key"]))
+
+
 func _draw_excavation_jobs() -> void:
 	var excavation: Variant = _service(&"excavation_service")
 	if excavation == null or not excavation.has_method(&"get_jobs"):
@@ -82,7 +82,11 @@ static func elevator_label(shaft: Dictionary) -> String:
 		value = "満員 " + value
 	if not bool(shaft[&"operational"]):
 		value = "停止 " + value
-	return value
+	else:
+		value = "運転 " + value
+	var phase := String(shaft.get(&"phase", &"IDLE"))
+	var doors_open := bool(shaft[&"operational"]) and phase in ["IDLE", "BOARDING", "ALIGHTING"]
+	return value + (" 開 " if doors_open else " 閉 ") + phase


 static func supply_label(requests: Array[Dictionary]) -> String:
@@ -2,6 +2,7 @@ extends "res://tests/fixtures/p08-03/placement.gd"
 ## Real Runtime commands; inherited formal codec/save-operation helpers unchanged.
 const Stream := preload("res://tests/fixtures/p23-03/diagnostic_stream.gd")
 const Comparison := preload("res://tests/fixtures/p23-03/layout_comparison.gd")
+const PurchaseQuoteV2 := preload("res://scripts/supply/supply_purchase_quote_v2.gd")
 var stream: RefCounted
 var layout := "A"
 var facility_ids: Dictionary = {}
@@ -28,7 +29,7 @@ func setup() -> void:

 func command(id: StringName, payload: Dictionary) -> bool:
 	var result: CommandResult = runtime.submit_command(id, payload)
-	expect(result.is_success(), String(id) + ": " + str(result.code))
+	expect(result.is_success(), String(id) + " " + str(payload) + ": " + str(result.code))
 	expect(stream.record("command", {&"command": id, &"payload": payload,
 		&"result": result.to_dictionary(), &"minute": clock.absolute_minute}), "record public command")
 	return result.is_success()
@@ -46,7 +47,7 @@ func observe_progress() -> void:
 	pass

 func prepare() -> void:
-	# Both layouts receive the same definition-derived diagnostic construction budget.
+	# Both layouts receive the same runtime-derived budget for all setup spends.
 	# Cafeteria/kitchen move together to x=6, leaving x=4..5 for the declared B EV.
 	var builds: Array = [[&"reception", Vector3i(-8, 0, 0)],
 		[&"cafeteria", Vector3i(6, 0, 0)], [&"warehouse", Vector3i(-8, 0, 1)],
@@ -59,6 +60,9 @@ func prepare() -> void:
 	var improvement: Dictionary = Comparison.new().declaration()[&"improvement"]
 	var budget_builds := builds.duplicate(true)
 	budget_builds.append([improvement[&"definition_id"], improvement[&"origin"]])
+	var assignments: Array = [[&"staff_akane", &"reception"], [&"staff_gina", &"warehouse"],
+		[&"staff_chika", &"kitchen"], [&"staff_iris", &"cafeteria"], [&"staff_juno", &"cafeteria"]]
+	var purchase_quantity := 40
 	var needed := {&"gold": 0, &"materials": 0}
 	var costs: Array = []
 	for row: Array in budget_builds:
@@ -66,13 +70,36 @@ func prepare() -> void:
 		costs.append({&"definition": row[0], &"cost": definition.construction_cost.duplicate(true)})
 		for key: StringName in needed:
 			needed[key] += int(definition.construction_cost.get(key, 0))
+	var construction_total := needed.duplicate(true)
+	var signing_costs: Array = []
+	var signing_total := 0
+	for assignment: Array in assignments:
+		var candidate: Dictionary = runtime.mvp_content_catalog.get_recruitment_entry(assignment[0])
+		expect(candidate.has(&"signing_cost"), "registered signing cost " + String(assignment[0]))
+		var signing_cost := int(candidate[&"signing_cost"])
+		signing_costs.append({&"candidate_id": assignment[0], &"signing_cost": signing_cost,
+			&"source": "MvpContentCatalog.get_recruitment_entry.signing_cost"})
+		signing_total += signing_cost
+	needed[&"gold"] += signing_total
+	# Before warehouse construction, use the purchase service's catalog/quote
+	# branch. Check this total against its live preview immediately before buying.
+	var item := SupplyCatalog.get_item(SupplyCatalog.RAW_FOOD)
+	var purchase_quote := PurchaseQuoteV2.make(SupplyCatalog.RAW_FOOD, purchase_quantity, int(item[&"unit_price"]))
+	expect(not purchase_quote.is_empty(), "registered raw-food purchase quote")
+	var purchase_total := int(purchase_quote[&"effective_total"] if PurchaseQuoteV2.eligible(runtime) else purchase_quote[&"base_total"])
+	needed[&"gold"] += purchase_total
+	var purchase_cost := {&"item_id": SupplyCatalog.RAW_FOOD, &"quantity": purchase_quantity,
+		&"unit_price": item[&"unit_price"], &"quote": purchase_quote, &"total_cost": purchase_total,
+		&"source": "SupplyCatalog.unit_price / PurchaseQuoteV2.make; preview_purchase eligibility"}
 	var state: Variant = tree.root.get_node("GameState")
 	var deltas := {}
 	for key: StringName in needed:
 		if needed[key] > state.get_resource(key):
 			deltas[key] = needed[key] - state.get_resource(key)
 	budget = {"setup": "diagnostic", "costs": costs, "needed": needed,
-		"initial": state.resources.duplicate(true), "deltas": deltas}
+		"initial": state.resources.duplicate(true), "deltas": deltas,
+		"construction_total": construction_total, "signing_costs": signing_costs,
+		"signing_total": signing_total, "purchase_cost": purchase_cost}
 	if not deltas.is_empty(): command(&"change_resources", {&"deltas": deltas})
 	expect(stream.record("budget", budget), "common diagnostic budget")
 	if layout == "B": builds.append([improvement[&"definition_id"], improvement[&"origin"]])
@@ -90,13 +117,15 @@ func prepare() -> void:
 		facility_ids[facility.definition_id] = facility.instance_id
 	command(&"set_facility_conditions", {&"facility_id": facility_ids[&"cafeteria"],
 		&"conditions": {&"related_facility_id": facility_ids[&"kitchen"]}})
-	for assignment: Array in [[&"staff_akane", &"reception"], [&"staff_gina", &"warehouse"],
-		[&"staff_chika", &"kitchen"], [&"staff_iris", &"cafeteria"], [&"staff_juno", &"cafeteria"]]:
+	for assignment: Array in assignments:
 		command(&"hire_staff_candidate", {&"candidate_id": assignment[0]})
 		command(&"set_staff_work_assignment", {&"staff_id": assignment[0],
 			&"facility_id": facility_ids[assignment[1]], &"template_id": &"DAY"})
+	var actual_quote: Dictionary = runtime.preview_supply_purchase(facility_ids[&"warehouse"], SupplyCatalog.RAW_FOOD, purchase_quantity)
+	expect(int(actual_quote[&"total_cost"]) == purchase_total, "budget matches actual purchase quote")
+	expect(stream.record("setup-purchase-quote", actual_quote), "record actual setup purchase quote")
 	command(&"purchase_supplies", {&"warehouse_id": facility_ids[&"warehouse"],
-		&"item_id": SupplyCatalog.RAW_FOOD, &"quantity": 40})
+		&"item_id": SupplyCatalog.RAW_FOOD, &"quantity": purchase_quantity})
 	command(&"set_kitchen_production", {&"facility_id": facility_ids[&"kitchen"],
 		&"enabled": true, &"target_stock": 8})
 	# The OS observation uses the already declared transport cohort, plus
@@ -166,6 +166,9 @@ func draw_rectangles(report: Dictionary, view: WorldPreview) -> Dictionary:
 			var rest := actor_rectangle(actor)
 			rest["category"] = "rest"
 			rows.append(rest)
+	for index: int in range(report.get(&"actor_aggregations", []).size()):
+		var badge: Dictionary = report[&"actor_aggregations"][index]
+		rows.append(badge_rectangle("aggregation", str(index), badge[&"at"], badge[&"text"]))
 	for index: int in range(report.get(&"badges", []).size()):
 		var badge: Dictionary = report[&"badges"][index]
 		rows.append(badge_rectangle("goods" if badge[&"kind"] == &"supply" else "business", str(index), badge[&"at"], badge[&"text"]))
@@ -190,7 +193,7 @@ func draw_rectangles(report: Dictionary, view: WorldPreview) -> Dictionary:

 func badge_rectangle(category: String, id: String, at: Vector2, text: String) -> Dictionary:
 	var width := ThemeDB.fallback_font.get_string_size(text, HORIZONTAL_ALIGNMENT_LEFT, -1, 10).x + 6
-	return {"category": category, "id": id, "rect": Rect2(at, Vector2(width, 14)),
+	return {"category": category, "id": id, "rect": Rect2(at, Vector2(text.length() * 10 + 6, 14)),
 		"text": text, "font_width": width - 6, "style_source": "scenes/main/world_preview.gd"}

 func actor_rectangle(actor: Dictionary) -> Dictionary:
@@ -0,0 +1,15 @@
+{
+  "purpose": "Historical negative only; not fresh acceptance evidence",
+  "source_request": "os-pair/A/reader-request-0003/request.json",
+  "source_request_sha256": "7dc3ee0bcc428e9d103726ee2448c767f1dd4107f9490833dc860c0b5140068c",
+  "manifest_sha256": "53aa8e2f12e5d0a089529457fcd55ee1ba0600e8dffff2225bb8b057d007fe1a",
+  "independent_judgment": [11, false, false, false],
+  "source_rows": [11, 29],
+  "report_inputs": {
+    "actor_drawings": [
+      {"key": "guest/guest-auto-00000041", "at": [651.0, 517.0], "scale": 3.0, "mode": "wait", "frame": 12, "flip": false},
+      {"key": "staff/staff_gina", "at": [651.0, 517.0], "scale": 3.0, "mode": "idle", "frame": 3, "flip": false}
+    ],
+    "cabins": []
+  }
+}
@@ -0,0 +1,17 @@
+{
+  "purpose": "Historical diagnostic input for placement-rule clarification; not fresh acceptance evidence",
+  "source_request": "os-pair/A/reader-request-0003/request.json",
+  "source_request_sha256": "7dc3ee0bcc428e9d103726ee2448c767f1dd4107f9490833dc860c0b5140068c",
+  "guests": [
+    {"id": "guest-auto-00000044", "state": "RECEIVING_SERVICE", "cell": [5, 0, 0], "target": "facility_00000003", "service": "service_00000009", "queue": ""},
+    {"id": "guest-auto-00000019", "state": "RECEIVING_SERVICE", "cell": [-6, 2, 0], "target": "facility_00000007", "service": "service_00000002", "queue": ""}
+  ],
+  "room_rects": {
+    "facility_00000003": [1027.0, 388.0, 256.0, 192.0],
+    "facility_00000007": [131.0, 292.0, 128.0, 96.0]
+  },
+  "traffic": [
+    {"cell": [5, 0, 0], "traffic": {"base_capacity": 4, "current_density": 6, "dynamic_cost": 0, "temporary_block": false}, "rect": [965.0, 486.0, 60.0, 92.0]},
+    {"cell": [-6, 2, 0], "traffic": {"base_capacity": 4, "current_density": 3, "dynamic_cost": 0, "temporary_block": false}, "rect": [261.0, 294.0, 60.0, 92.0]}
+  ]
+}
@@ -0,0 +1,14 @@
+{
+  "applicability": "Historical negative only. Distilled local-coordinate fields; never S/T execution proof.",
+  "source_commit": "9a26f9ab8c6691ba3cbe32ac4bc4525f4a9c7106",
+  "manifest_sha256": "a25e15f00d09acb16b05e180a6263c3771c84449bd6e0c99b6968fd739727bd2",
+  "request_sha256": "7099ca03da3cef5819e7965f6fdea8777b9995a9e12806e870e844ec2b02b9a5",
+  "request_relative_to_R15": "os-pair/A/reader-request-0003/request.json",
+  "actual_negative_rows": [6, 17],
+  "cabin_rect": [659, 463, 96, 76],
+  "passenger_count": 0,
+  "actor": {"key": "guest/os-p23-024", "center": [739, 532], "at": [715, 517], "scale": 3, "mode": "wait", "frame": 6, "flip": false},
+  "traffic": {"cell": [0, 1, 0], "rect": [645, 390, 60, 92], "band_rect": [652.2, 473.5, 45.6, 5], "traffic": {"base_capacity": 4, "current_density": 16, "dynamic_cost": 0, "temporary_block": false}},
+  "queue_badge": {"at": [563, 468], "text": "待ち27人（他18人）"},
+  "note": "Raw R15 maps/receipts remain unchanged. New geometry tests calculate a counterfactual projection in memory only."
+}
@@ -5,6 +5,98 @@ const Stream := preload("res://tests/fixtures/p23-03/diagnostic_stream.gd")
 var checks := 0
 var failures := 0

+class QueueWorld extends RefCounted:
+	var revision := 0
+	func get_bounds() -> Dictionary:
+		return {&"min_x": -1, &"max_x": 2, &"min_y": 0, &"max_y": 1}
+	func is_walkable(cell: Vector3i) -> bool:
+		return cell.y >= 0 and cell.y <= 1 and cell.x >= -1 and cell.x <= 2
+	func get_occupant(_cell: Vector3i) -> StringName:
+		return &""
+
+class QueueLift extends RefCounted:
+	var layer := 0
+	var visual_floor := 0.3
+	func get_report() -> Dictionary:
+		var trips: Array = []
+		for i: int in 3:
+			trips.append({&"state": PathService.ELEVATOR_REQUEST_QUEUED, &"from": Vector3i(0,0,layer),
+				&"actor_id": "guest:q%d" % i, &"enqueued_at": i})
+		return {&"shafts": [{&"shaft_id": &"wide", &"stops": [Vector3i(0,0,layer), Vector3i(0,1,layer)],
+			&"width": 2, &"capacity": 4, &"boarded": [], &"visual_floor": visual_floor,
+			&"operational": true, &"phase": &"PICKUP"}], &"trips": trips}
+
+class QueueRuntime extends RefCounted:
+	var world_service := QueueWorld.new()
+	var elevators := QueueLift.new()
+
+func repair28_geometry() -> void:
+	var view := WorldPreview.new()
+	view.size = Vector2(408,332)
+	var fixture := QueueRuntime.new()
+	view._runtime = fixture
+	root.add_child(view)
+	for layer: int in [0,1]:
+		view.layer_mode = layer as WorldPreview.LayerMode
+		fixture.elevators.layer = layer
+		var model: Dictionary = view._observe_queues()
+		expect(model[&"shaft_footprint"].has(Vector3i(1,0,layer)), "width-2 interior belongs to shaft")
+		expect(model[&"queue_positions"].size() == 1 and model[&"queue_hidden"].size() == 2, "shaft stops outward lane and retains overflow")
+		var badge: Dictionary = model[&"landing_badges"][0]
+		expect(badge[&"members"].size() == badge[&"visible_count"] + badge[&"overflow"], "no queue member lost")
+		model.merge({&"boarded": {}, &"elevator_tracks": {}, &"paused": true})
+		var cabins: Array[Dictionary] = view._observe_elevators(model)
+		expect(cabins.size() == 1, "active-layer cabin remains visible")
+		for at: Vector2 in model[&"queue_positions"].values():
+			for cell: Vector3i in model[&"shaft_footprint"]:
+				expect(not view.get_cell_rect(cell).has_point(at), "queue anchor outside full footprint")
+			expect(not cabins[0][&"rect"].has_point(at), "queue anchor outside empty cabin")
+		expect(cabins[0][&"passenger_count"] == 0 and cabins[0][&"passengers"].is_empty(), "empty cabin has no passenger")
+		expect(cabins[0][&"operational"] and not cabins[0][&"doors_open"] and cabins[0][&"phase"] == &"PICKUP", "typed moving cabin state")
+		for floor: float in [0.0, 0.8]:
+			fixture.elevators.visual_floor = floor
+			var pose_model: Dictionary = view._observe_queues()
+			var cell := Vector3i(0, 0 if floor == 0 else 1, layer)
+			var key := "guest/landing"
+			var person := {&"id": &"landing", &"cell": cell, &"service": &"", &"target": &""}
+			pose_model.merge({&"boarded": {}, &"elevator_tracks": {}, &"paused": true,
+				&"actor_tracks": {key: {&"cell": cell, &"from": cell, &"at": 0.0, &"direction": 1.0}},
+				&"rooms": {}, &"guests": [person], &"staff": [], &"carried_patients": {},
+				&"resting_staff": {}, &"treated_patients": {}})
+			pose_model[&"cabins"] = view._observe_elevators(pose_model)
+			var row: Dictionary = view._actor_drawing(view.cell_to_canvas(cell), key, false,
+				GuestRuntime.State.MOVING, false, false, &"", false, pose_model)
+			var region: Dictionary = view._life_actor_region(row, person, pose_model)
+			expect(region.has(&"center") and not region[&"transit"], "same-floor and fractional cabin use landing region")
+			expect(not (region[&"bounds"] as Rect2).intersects(pose_model[&"cabins"][0][&"rect"]), "landing bounds outside cabin")
+			pose_model[&"actor_drawings"] = [row]
+			view._fit_life_actors(pose_model)
+			expect(pose_model[&"actor_drawings"].size() == 1 and pose_model[&"actor_aggregations"].is_empty(), "landing actor retained")
+			expect(not view._life_core(row).intersects(pose_model[&"cabins"][0][&"rect"]), "fitted non-passenger core outside cabin")
+			expect(row[&"cell"] == cell and not row[&"riding"] and row[&"state"] == GuestRuntime.State.MOVING, "landing placement preserves actual state")
+		fixture.elevators.visual_floor = 0.3
+		for operational: bool in [true,false]:
+			for phase: StringName in [&"IDLE", &"BOARDING", &"PICKUP"]:
+				var shaft: Dictionary = fixture.elevators.get_report()[&"shafts"][0]
+				shaft[&"operational"] = operational
+				shaft[&"phase"] = phase
+				var text := Presentation.elevator_label(shaft)
+				expect(text.contains("0/4人") and text.contains(String(phase)), "count and phase visible")
+				expect(text.begins_with("運転" if operational else "停止"), "operation visible beyond lamp")
+				expect(text.contains(" 開 " if operational and phase != &"PICKUP" else " 閉 "), "doors visible beyond lamp")
+	var obstacles: Array[Rect2] = [Rect2(20,0,20,90), Rect2(45,80,20,14)]
+	var segments := view._traffic_segments(Rect2(10,85,60,5), Rect2(0,0,80,96), obstacles)
+	var area := 0.0
+	for segment: Rect2 in segments:
+		area += segment.get_area()
+		for obstacle: Rect2 in obstacles:
+			expect(not segment.intersects(obstacle), "traffic never crosses cabin or badge")
+	expect(is_equal_approx(area, 100.0), "traffic preserves all uncovered area")
+	var covering: Array[Rect2] = [Rect2(0,80,80,16)]
+	var fallback := view._traffic_segments(Rect2(10,85,60,5), Rect2(0,0,80,96), covering)
+	expect(fallback.size() == 1 and fallback[0] == Rect2(10,0,60,5), "shaft-column density remains visible")
+	view.free()
+
 func _initialize() -> void:
 	call_deferred("run")

@@ -25,6 +117,7 @@ func run() -> void:
 		return
 	var clock: Variant = root.get_node("GameClock")
 	clock.set_paused(true)
+	repair28_geometry()
 	# Isolated real typed components test state labels, without claiming gameplay.
 	var labels: Dictionary = {}
 	for state: int in FacilityInstance.State.values():
@@ -56,6 +149,61 @@ func run() -> void:
 	view.size = Vector2(1440, 810)
 	view.set_runtime(runtime)
 	root.add_child(view)
+	# R13's four same-cell MOVING guests. Component fixture only, no Runtime edit.
+	var crowd: Array = []
+	var offsets := [0, 16, -16, 8]
+	for i: int in 4:
+		crowd.append({&"row": {&"key": "guest/os-p23-%03d" % (12+i),
+			&"center": Vector2(108+offsets[i], 118-4*(i%2)),
+			&"at": Vector2(84+offsets[i], 103-4*(i%2)), &"scale": 3.0,
+			&"mode": &"idle", &"frame": 0, &"flip": false,
+			&"state": GuestRuntime.State.MOVING, &"cell": Vector3i(1,1,0)},
+			&"bounds": Rect2(80,74,56,88), &"transit": false})
+	crowd.sort_custom(func(a: Dictionary, b: Dictionary) -> bool: return a[&"row"][&"center"].x < b[&"row"][&"center"].x)
+	# _life_pack uses the view's cell dimensions for the unchanged feet formula.
+	# The x-capacity assertion is independent of the current default world size.
+	expect(view._life_pack(crowd, 3).is_empty(), "four cores cannot fit 72px in 56px")
+	var saved_size := view.size
+	var saved_zoom := view.zoom_level
+	var dim: Vector2 = view._cell_dimensions()
+	view.zoom_level *= 64.0/dim.x
+	var fitted: Array = view._life_pack(crowd, 2)
+	expect(fitted.size() == 4, "all four R13 occupants fit at p=2")
+	for row: Dictionary in fitted:
+		expect(row[&"scale"] == 2.0 and row[&"cell"] == Vector3i(1,1,0), "fitting preserves cell and uses p=2")
+	var regroup: Array = crowd.slice(0, 2).duplicate(true)
+	for i: int in 2:
+		regroup[i][&"row"][&"center"] = Vector2(108, 118+26*i)
+		regroup[i][&"bounds"] = Rect2(80,74,20,130)
+	var regrouped: Array = view._life_fitted_groups(regroup)
+	expect(regrouped.size() == 2, "scale change splits vertical component")
+	for group: Array in regrouped:
+		expect(group.size() == 1 and group[0][&"row"][&"scale"] == 2.0, "regroup avoids stale p=1 fit")
+	var saturated: Array = []
+	for i: int in 10:
+		var member: Dictionary = crowd[0].duplicate(true)
+		member[&"row"][&"key"] = "guest/saturation-%02d" % i
+		member[&"row"][&"center"] = Vector2(108,118)
+		saturated.append(member)
+	var saturated_groups: Array = view._life_fitted_groups(saturated)
+	expect(saturated_groups.size() == 1, "saturation keeps one vertical component")
+	expect(view._life_pack(saturated_groups[0], 1).is_empty(), "ten cores cannot fit 60px in 56px")
+	expect(view._life_pack(saturated_groups[0].slice(0,9), 1).size() == 9, "nine cores fit without losing identities")
+	var aggregate_model := {&"guests": [], &"staff": [], &"actor_drawings": [], &"cabins": []}
+	var aggregate_rows: Array = []
+	for member: Dictionary in saturated:
+		var row: Dictionary = member[&"row"].duplicate(true)
+		row[&"riding"] = true
+		aggregate_rows.append(row)
+		aggregate_model[&"guests"].append({&"id": String(row[&"key"]).trim_prefix("guest/"), &"cell": Vector3i(1,1,0)})
+	aggregate_model[&"cabins"].append({&"rect": Rect2(76,70,64,96), &"passengers": aggregate_rows})
+	view._fit_life_actors(aggregate_model)
+	expect(aggregate_model[&"cabins"][0][&"passengers"].size() == 9, "aggregation draws all nine fitting members")
+	var aggregate: Dictionary = aggregate_model[&"actor_aggregations"][0]
+	expect(aggregate[&"members"].size() == 10 and aggregate[&"drawn_keys"].size() == 9, "aggregation retains every identity")
+	expect(aggregate[&"text"] == "+1" and aggregate[&"at"] == Vector2(80,74), "actual aggregation badge position and text")
+	view.zoom_level = saved_zoom
+	view.size = saved_size
 	var before: PackedByteArray = var_to_bytes(runtime.create_snapshot())
 	var tracks: PackedByteArray = var_to_bytes([view._actor_tracks, view._elevator_tracks])
 	var model := view._build_life_observation()
@@ -67,6 +215,10 @@ func run() -> void:
 	await process_frame
 	var report := view.get_life_observation_report()
 	expect(not report.is_empty(), "actual drawing published an observation")
+	expect(report.get(&"observation_serial", -1) == report.get(&"overlay_consumed_serial", -2), "LifeOverlay consumed this observation")
+	expect(report.get(&"overlay_passes", []) == [&"traffic", &"cabins", &"actors"], "traffic drawn before cabins and actors")
+	var indices: Array = report.get(&"overlay_child_indices", [])
+	expect(indices.size() == 3 and indices[0] < indices[1] and indices[1] < indices[2], "LifeOverlay above inactive art and below lens")
 	# The getter exposes the last actual draw and must not create another model.
 	var original: PackedByteArray = var_to_bytes(report)
 	if report.has(&"traffic"):
@@ -20,7 +20,7 @@ CONDITIONS = ('room_occupancy_and_operation', 'queues_and_overflow',
              'elevator_state_and_passengers', 'staff_state_and_jobs',
              'supplies_and_carried_goods', 'same_demand_route_improvement',
              'normal_os_and_distinct_pid_resume')
-OS_CHILD_TIMEOUT_SECONDS = 900
+OS_CHILD_TIMEOUT_SECONDS = 3600
 RECORDER = 'tests/fixtures/p23-03/passive_os_recorder.gd'

 def collect_screenshots(out, pair, role):
@@ -235,12 +235,15 @@ def collect(args):
        raise ValueError('evidence output must be outside the frozen source checkout')
    r = module('p23_standard_runner', root/'tools/run_all_tests.py')
    commit = subprocess.check_output(['git', '-C', str(root), 'rev-parse', 'HEAD'], text=True).strip()
-    r.require(commit == args.commit, 'actual final submitted commit required')
+    r.require(commit == args.commit, 'actual tested commit Q required')
    r.require(not subprocess.check_output(['git','-C',str(root),'status','--porcelain'], text=True), 'frozen clean worktree required')
    r.require(not out.exists(), 'fresh output; never overwrite earlier proof')
    refs = r.source_refs(root)
    checker = module('p23_initial_inputs', root/'tools/check_p23_life_observation_evidence.py')
    inputs_before = checker.input_map(root)
+    root_roles = checker.strict_json(args.root_composition.read_bytes())
+    composition = checker.verify_composition(root, root_roles,
+        {name: pin['sha256'] for name, pin in refs.items()}, inputs_before)
    outer = r.module('p23_owned_outer', r.SUPPORT/'owned_outer.py')
    r.require(not outer.native.helper['live_engines'](), 'one native engine')
    engine = shutil.which(args.godot)
@@ -308,6 +311,8 @@ def collect(args):
            os.environ.update(P23_LAYOUT=layout, P23_OUTPUT=str(directory))
            phase('layout-'+layout.lower(), 'tests/ui/test_p23_life_observation.gd')
            cases.append({'directory':directory.name, 'phase':phases[-1], 'files':r.inventory(directory)})
+        for layout in ('A', 'B'):
+            checker.inspect_stream(out/('layout-'+layout), root)
        if args.input_mode == 'native_os':
            os_phases = collect_os(r, root, out, stage, engine, commit, refs, protected, outer)
            checker = module('p23_os_input_map', root/'tools/check_p23_life_observation_evidence.py')
@@ -328,7 +333,8 @@ def collect(args):
    r.write(out/'runner-ended.json', {'error':error, 'cleanup':cleanup, 'cleanup_errors':cleanup_errors,
        'phases':phases, 'formal_acceptance':False, 'native_Windows':'pending'})
    checker = module('p23_observation_checker', root/'tools/check_p23_life_observation_evidence.py')
-    r.write(out/'index.json', {'schema':'p23-life-observation-v1', 'commit':commit,
+    r.write(out/'index.json', {'schema':'p23-life-observation-v2', 'commit':commit,
+        'composition':composition,
        'source_map':selected['map'], 'input_map':inputs_before,
        'provenance':checker.input_provenance(root),
        'marker_contract_sha256':r.sha(root/CONTRACT),
@@ -345,6 +351,8 @@ def main(argv=None):
    parser.add_argument('--project-root', type=absolute, required=True)
    parser.add_argument('--output', type=absolute, required=True)
    parser.add_argument('--commit', required=True)
+    parser.add_argument('--root-composition', type=absolute, required=True,
+                        help='Root external S/M/T/Q and complete tested-tree blob bindings')
    parser.add_argument('--godot', default='godot')
    parser.add_argument('--input-mode', choices=('handler', 'native_os'), default='handler')
    parser.add_argument('--collect', action='store_true', required=True)
Author	SHA1	Message	Date
Agent Control Runner	26f60095bf	P23-03: 生活の観察	2026-10-16 17:15:06 +09:00
Agent Control Runner	93248f4d79	P23-03: 生活の観察	2026-10-16 14:56:38 +09:00
Agent Control Runner	7f1920b329	P23-03: 生活の観察	2026-10-16 14:41:54 +09:00
Agent Control Runner	447f77a13b	P23-03: 生活の観察	2026-10-16 14:24:10 +09:00
Agent Control Runner	6c13940cf4	P23-03: 生活の観察	2026-10-16 14:17:14 +09:00
Agent Control Runner	9a26f9ab8c	P23-03: 生活の観察	2026-10-15 18:30:42 +09:00
Agent Control Runner	19fc3f50c7	P23-03: 生活の観察	2026-10-15 18:01:00 +09:00
Agent Control Runner	49445d4a4d	P23-03: 生活の観察	2026-10-15 12:08:24 +09:00
Agent Control Runner	336a8cd0b3	P23-03: 生活の観察	2026-10-15 10:57:46 +09:00
Agent Control Runner	56bf5a3e86	P23-03: 生活の観察	2026-10-15 10:53:39 +09:00