Bound donor capacity in plan_zone_ratio_retry by min(static_slack,
max(0, clientHeight-scrollHeight)) when both Step 14 measured fields
are present; fall back to static contract slack when absent. Prevents
the donor from being over-allocated when full-but-not-overflowing,
avoiding a wasted Selenium rerender before cascade falls to
cross_zone_redistribute.
- src/phase_z2_retry.py: planner block L122-157 only; donor filter
(L107-112), slack<=0 gate, base_plan, greedy aggregation untouched.
Adds measured_empty_px + slack_bound_source telemetry to
donor_candidates_considered (additive only).
- tests/phase_z2/test_phase_z2_retry_measured_bound.py: 5-axis
regression (static_fallback / measured<static / measured>=static /
measured==0 excludes / filter+bool guard).
Guardrails honored: V4 rank-1 frame lock preserved, no frame_swap,
no spacing/padding/gap/line-height/font shrink, no content drop,
no MDX 03/04/05 branching, no Step 14 schema mutation. Static
fallback idempotent when measured fields absent.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
BACKLOG line 93 + INSIGHT-MAP line 150 (verbatim mirror per anchor sync
rule). Trigger axis now requires both Phase R' archive trigger AND
§2.1/§2.2 SoT signature unification to keep guardrail = code-removal-only.
No source files touched.
Bug discovered during #24 IMP-24 K6 Stage 2 (2026-05-20):
- Codex r1, r2, r3 started with '=== IMPLEMENTATION_UNITS ===' on first line
(not '[Codex #N] ...'), so detect_agent (P0-1 strict, first-line only)
returned None.
- For non-audit issues, the P5 supplement guard was audit-only gated → silent
loop until Codex r4 happened to use correct format. 4 rounds wasted.
Verified that #21 Stage 4 had the same latent silent loop pattern
('## [Codex #1]' first line) — orchestrator looped through ~10 Claude rounds
before random recovery. P5b fix addresses this long-standing bug.
Patch (defensive parser-contract hardening; does not assume single root cause):
1. RULES global gets explicit "FIRST non-empty line MUST be [Claude #N] /
[Codex #N]" rule that OVERRIDES any stage-specific "body MUST contain"
constraint.
2. COMPACT_PLAN_RULE wording clarified: "body" begins AFTER the first-line
agent header. The 'body MUST contain ONLY' set no longer accidentally
permits '=== IMPLEMENTATION_UNITS ===' on line 1.
3. is_codex None supplement guard:
- audit-only gate REMOVED → fires for all issues (#24 latent loop fixed)
- Throttle: max 2 supplements per stage; on 3rd violation, orchestrator
hard-stops the issue with explicit "user action required" message
and exits run_stage cleanly
- Supplement message names both Claude AND Codex (Claude's first-line
violation also breaks downstream via Codex mimicry)
- Body-head 80 chars logged on detection failure (debugging aid)
4. Regression tests (+5 cases in test_orchestrator_core.py):
- TestDetectAgent: '=== IMPLEMENTATION_UNITS ===' first line → None
- TestDetectAgent: [Codex #N] first line + units after → 'codex' OK
- TestDetectAgent: '## ', '📌 **', '**' prefix all → None
- TestRulesAndCompactPlanFirstLineContract: RULES wording has FIRST/OVERRIDES
- TestRulesAndCompactPlanFirstLineContract: COMPACT_PLAN_RULE has carve-out
Cosmetic side effect (accepted): Claude's '📌 **[Claude #N] ...**' or
'## [Codex #N] ...' decoration prefixes will fail detect_agent. Agents
will drop decorations from line 1; line 2+ can still use them.
Out of scope (NOT included to keep regression risk low):
- detect_agent function logic UNCHANGED (P0-1 strict preserved)
- consensus parser UNCHANGED
- stage loop structure UNCHANGED
- git/Gitea retrieval logic UNCHANGED
- audit-only mode P4/P4a guards UNCHANGED
- pre-post comment validation (future axis, larger refactor)
Total: 131/131 pytest pass (126 prior + 5 new).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Bug discovered during #56 INTEGRATION-AUDIT-02 execution (2026-05-20):
- Both Claude and Codex put "Audit anchor: ..." as the FIRST line of every
Gitea comment per the #56 issue body instruction "cite anchor at start
of every stage".
- detect_agent (P0-1 strict, first-line only) then returns None for these
comments because the first line is "Audit anchor:..." not "[Codex #N]"
or "[Claude #N]".
- Result: orchestrator's "is_codex" check (line ~1288) flips false →
"Codex 응답 미감지 — continuing" → infinite Stage 4 loop. #56 reached
Round #14 (>300 comments, ~2 hours wasted token).
Fix path (NOT relaxing detect_agent — that would revive the original #45
pre-P0-1 bug where [Claude #N] citations inside Codex bodies caused
mis-detection):
1. AUDIT_ONLY_NOTE updated to enforce comment format:
- FIRST non-empty line MUST be `[Claude #N] <stage>` or `[Codex #N] <stage>`
- Audit anchor / banners / prefaces MUST appear line 2 or later
- Concrete CORRECT example included
- Explicit warning that violation breaks stage advance
2. is_codex None guard auto-supplements:
- When _audit_mode(title) AND detect_agent returns None, orchestrator
posts a Gitea supplement comment requesting the correct format
- Next round's Claude/Codex see the supplement and correct
- Breaks the infinite loop automatically (no manual ctrl-C needed)
3. Regression tests in TestDetectAgent (test_orchestrator_core.py):
- test_audit_anchor_preface_breaks_detection: confirms P0-1 strict
correctly returns None when anchor is first line
- test_audit_anchor_after_header_works: correct format passes
Total: 96/96 pytest pass (94 prior + 2 P5 regression).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
프로젝트의 왜 / 무엇을 위해 / 어떻게 라는 질문에 대한 master 답.
이 문서가 있으면 매번 처음부터 framing 설명할 필요 없음.
구조:
1. Destination — Phase Z 22-step + AI zone-fit frame generation
2. Q~Y 검토 = 이미 완료 (과거형). 결과 = INSIGHT-MAP + 28 초기 이슈.
3. INSIGHT-MAP catalog 구조 (§0~§5)
4. IMP 이슈 좌표 체계 (관련 step + source + priority + scope + guardrails)
5. orchestrator 의 disciplined executor 역할 (Claude + Codex 합의)
6. Audit cycle (meta-governance) — 발견은 follow-up 이슈로 분리
7. 도착점 도달 기준 5 항목
8. 자주 헷갈리는 anti-patterns (heritage 보존 X, MDX 최적화 X 등)
9. 핵심 참조 문서 인덱스
10. 한 줄 요약
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Audit follow-up F-2 (INTEGRATION-AUDIT-01 §10.2). Phase Z families surface
showed 11 tracked / 11 contracted / 13 on disk. The 2 untracked WIP files
(app_sw_package_vs_solution.html, pre_construction_model_info_stacked.html)
are now declared in _WIP_FILES.md as uncontracted and out-of-scope for the
runtime matcher; promote/remove is gated on #42. The 11/11 tracked +
contracted baseline is unchanged. A new pytest enforces tracked families ↔
frame_contracts.yaml set-equality modulo the WIP allowlist parsed from
_WIP_FILES.md, so future drift fails fast in CI before #42 expands to 32
frames.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Append IMP-50 audit completion row referencing INTEGRATION-AUDIT-01-REPORT
(commit 8c7d693) with CONDITIONAL GO for #19 decision.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
P4 had two production issues blocking #50 integration audit deployment:
1. Stage 3 guard had no baseline awareness — flagged ALL forbidden-path
changes including pre-existing dirty WIP. Empirical: 328 such files
already in current working tree (tests/matching/ artifacts etc).
#50 would have hit reject loops immediately without Claude doing
anything wrong.
2. Stage 5 had no commit-scope guard — if Claude ran `git add -A` and
committed user's existing WIP, audit commit would be polluted with
unrelated production changes.
P4a additions:
- _audit_baseline_path / _ensure_audit_baseline / _load_audit_baseline:
snapshot working-tree dirty paths at run_issue entry for audit issues.
Resumed runs preserve existing baseline (no overwrite).
- _check_audit_only_violations(baseline=None): accept baseline set,
subtract from violations — only flags NEW forbidden changes introduced
after audit start.
- _check_audit_commit_scope: verify HEAD commit's file list matches
AUDIT_ALLOWED_COMMIT_GLOBS (INTEGRATION-AUDIT-*.md, BACKLOG.md).
- run_issue: save baseline on audit-mode entry only — no impact on
normal issues.
- Stage 5 (commit-push) YES gate: new guard rejects on out-of-scope
files with remediation prompt (git reset --soft + force-with-lease).
19 new tests:
- baseline subtraction (5): pre-existing removed, None=keep-all,
empty-set=catch-all, full-coverage filter, Windows path normalize.
- baseline persist (5): roundtrip, no-overwrite on resume, missing
fallback, corrupt JSON fallback, non-list fallback.
- commit scope detection (7): report-only allowed, backlog allowed,
src/ rejected, unrelated docs rejected, git error fail-open,
Windows backslash, empty commit pass.
- allowed globs sanity (2): every glob has audit marker, all under
docs/architecture/.
Total: 94/94 pytest pass (75 prior + 19 new).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Issue: #48 (IMP-15 실행-4, axis 4: debug.json + spec doc trace).
Parent: #15. Depends on 실행-1/2/3 (events + classifier outputs).
Surfaces the image/table event streams that 실행-1/2/3 already produced
and consumed, mirroring the existing `zone_geometries_px` top-level
precedent (no new pattern introduced). Adds the matching taxonomy row
to the Phase Z fit-classifier/router spec.
src/phase_z2_pipeline.py (+3):
- write_debug_json now lifts `image_events` and `table_events` to
top-level of `debug.json` via `(visual_runtime_check or {}).get(<k>, [])`,
exactly mirroring the immediately preceding `zone_geometries_px`
surfacing line. Defaults to `[]` when `visual_runtime_check` is None
— additive, no consumer-visible breakage.
docs/architecture/PHASE-Z-FIT-CLASSIFIER-ROUTER-SPEC.md (+1):
- §3.1 taxonomy adds `image_aspect_mismatch` row. Row text explicitly
marks the signal as post-render `fail_reasons` from Step 14
visual_runtime_check (rendered vs declared aspect ratio mismatch),
NOT a router-routed fit_classifier output, and notes the separate
`image_events` stream surface. Prevents future readers from wiring
this taxonomy into §3.2 priority list or §4 router action map.
tests/phase_z2/test_debug_json_event_surfacing.py (new, 2 tests):
- `test_write_debug_json_surfaces_image_and_table_events` invokes
write_debug_json with synthetic visual_runtime_check containing
both event lists; reads back the on-disk debug.json and asserts
both keys are present at top level with the exact payloads.
- `test_write_debug_json_defaults_when_visual_runtime_check_none`
asserts both new keys default to `[]` when visual_runtime_check
is None — guards the defensive `(… or {})` pattern.
tests/phase_z2/test_spec_taxonomy_image_aspect_mismatch.py (new, 2 tests):
- `test_spec_has_image_aspect_mismatch_row` opens the spec file and
asserts exactly one `^\| image_aspect_mismatch \|` row exists
inside the §3.1 table block (no markdown-parser dependency).
- `test_spec_row_marks_post_render_fail_reasons_semantic` asserts the
row text carries both "Post-render" and "fail_reasons" tokens —
enforces the Stage 1 guardrail wording.
Verification (Stage 4 PASS, Claude + Codex independent):
- pytest -q tests/phase_z2/test_debug_json_event_surfacing.py \
tests/phase_z2/test_spec_taxonomy_image_aspect_mismatch.py
→ 4 passed in 0.07s.
- git diff scope: 4 files, +148 insertions / 0 deletions.
Scope-locked: no edits to classifier (실행-3), event generation
(실행-1/2), Step 21 viewer, §3.2 priority list, §4 router action
mapping, or `table_self_overflow` taxonomy row. Pre-existing
dirty/untracked working-tree files left untouched.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Issue #47 (IMP-15 실행-3 axis 3): extend `classify_visual_runtime_check`
to consume the `image_events[]` and `table_events[]` arrays produced by
`run_overflow_check` (실행-1/2) and widen `visual_check_passed`.
Changes (src/phase_z2_classifier.py):
- Remove `overflow.passed=True` early-return so image/table event scans
always run, even when zone-level overflow was clean.
- Deferred import of `IMAGE_ASPECT_DELTA_TOL` and `TABLE_SCROLL_TOL_PX`
from `phase_z2_pipeline` (circular-safe SSoT; no duplicate literals).
- New `image_events` scan emits `image_aspect_mismatch` when
`delta is not None AND |delta| > IMAGE_ASPECT_DELTA_TOL`
(delta=None ⇒ skip, image not loaded).
- New `table_events` scan emits `tabular_overflow` when
`wrapper_clipped_index is None AND (excess_x or excess_y > TABLE_SCROLL_TOL_PX)`
(wrapper-clipped tables deduped against the existing zone cascade).
- `visual_check_passed = overflow.passed AND not classifications` —
any image/table classification now flips the gate.
Guardrails preserved:
- §3.2 8-rule zone cascade (clipped_inner / zone-self) untouched —
the new emitters are ADDITIONAL.
- `placement_diagnostics`, `categories_seen`, `unclassified_signals`
return-shape preserved.
- No `pipeline.py` production changes; no router action or
`debug.json` passthrough changes.
Tests (tests/phase_z2/test_phase_z2_visual_classifier.py — new):
- `test_image_aspect_mismatch_emits_classification` (|delta|>TOL fires)
- `test_image_aspect_delta_below_tol_no_classification` (≤TOL skipped)
- `test_standalone_table_overflow_emits_classification`
(wrapper_clipped_index=None, excess>TOL fires)
- `test_table_dedup_when_wrapper_clipped`
(wrapper_clipped_index set ⇒ no `tabular_overflow` emit)
All 4 pure-dict (no Selenium / chromedriver / pipeline execution).
Tolerances imported from `phase_z2_pipeline` (SSoT enforced via test
import — no classifier-local literals).
Verification (Stage 4):
- New classifier tests: 4/4 PASS.
- Regression `tests/phase_z2/` excluding new file: 93/93 PASS.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Add table self-overflow detection with element-identity wrapper dedup,
mirroring the image_aspect_mismatch axis pattern (#45).
JS layer: TABLE_SCROLL_TOL_PX=5 module constant; clippedWrapperMap
built as Map<Element,int> keyed by DOM node reference (NOT className)
so two wrappers with identical class strings remain distinguishable;
table_events collected via querySelectorAll('table').forEach with
closest()-ancestor walk resolving wrapper_clipped_index = int|null.
Py layer: aggregate result['table_events'] and append fail_reason
'table_self_overflow' only when (excess_x>TOL OR excess_y>TOL)
AND wrapper_clipped_index is None; wrapper-clipped path continues
to fail via existing clipped_inner reporting.
Tests (Selenium, chromedriver guard mirrored from image_check):
- Fixture D: standalone <table> overflow → table_self_overflow fail
- Fixture E: <table> in clipped wrapper → dedup suppresses table fail
- Fixture F (F1 acceptance): two wrappers with identical className
f13b-cell, W1 clipped by non-table child, W2 hosts self-overflow
<table> with W2 itself NOT clipped → element-identity ensures W2's
table is not suppressed by W1's class; both fails emitted.
Out of scope: image_events behavior (intact from #45), classifier
pass/fail consumer (→실행-3), debug.json surfacing (→실행-4).
Refs: #46
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
scripts/generate_frame_previews.py iterates figma_to_html_agent/blocks/{frame_id}/index.html,
renders preview.png via Selenium headless (capture_slide_screenshot pattern reuse), and writes
_preview_manifest.json (schema v1) with idempotent stale-detect (mtime+sha256). Build-time only
— no runtime pipeline integration, no AI calls, no MDX/Jinja regen. Stage 2 baseline (commit
56619a0): total=33, renderable=20, missing_index_html=13, orphan=1 (1171281192).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Stage 3 lock implementation: extend build_layout_css dispatch beyond
the horizontal-2 / vertical-2 1-D dynamic paths. T / inverted-T /
side-T-left / side-T-right / 2x2 now flow through a 2-D track solver
instead of the fr_default sink, with length-locked heights_px (R) +
widths_px (C) on every return path (default and override).
PR 2 scope (u1~u5):
- u1: _aggregate_zone_signals_per_track — per-row + per-col virtual
zones via max(weight) + max(min_height_px) of single-span zones,
falling back to all-span when a track has none.
- u2: _build_grid_dynamic_2d default builder — feeds virtual zones
into compute_zone_layout + compute_zone_layout_cols; emits
computation="2d_dynamic_aggregated", dynamic_rows=True,
dynamic_cols=True.
- u3: _override_to_grid_tracks override builder — single-span
aggregation (max h per row, max w per col), normalize, multiply
by avail_h/avail_w, last-element diff absorb; emits
computation="user_override_geometry"; falls back to u2 when
total_h or total_w == 0.
- u4: build_layout_css dispatcher wiring — topology in
{T, inverted-T, side-T-left, side-T-right, 2x2} routes to
_build_grid_dynamic_2d (default) or _override_to_grid_tracks
(override); legacy [override-warning] stderr removed for the
5 presets; step08 trace gains a 2-D-aware print line that fires
before the dynamic_rows / dynamic_cols branches.
- u5: PR 1 lock test test_top_1_bottom_2_fr_default_populates_geometry
renamed to test_top_1_bottom_2_dynamic_2d_populates_geometry and
flipped to PR 2 reality (computation="2d_dynamic_aggregated",
dynamic_rows=True, dynamic_cols=True).
Fixtures: 10 build_layout_css (5 presets × {default, override}) +
5 retry_gate *_dynamic_2d.yaml locking the retry gate skip reason
"dynamic_cols (2-D topology) ... IMP-09 lock" for the 5 presets.
Tests: python -m pytest -q tests = 104 passed (Stage 2 baseline
10 RED → GREEN, 0 regressions). Kei archive
(build_containers_type_b / page_structure) untouched —
rg "build_containers_type_b|page_structure" src/phase_z2_pipeline.py
returns 0 hits.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Codex #1 (Stage 5) reproduced a smoke regression on the actual checkout :
when V4 carries the parent exact key (e.g., `04-2`) AND the drag/drop
override targets a sub-id (`primary=04-2-sub-1`), the aligner kept the
parent at parent granularity and emit `['04-1', '04-2']`, so the override
flag failed with `unknown section_id(s) ['04-2-sub-1']`.
Fix : `align_sections_to_v4_granularity` gains an optional
`override_target_section_ids` keyword. From each canonical
`${parent}-sub-N` target it derives the parent id and adds it to a
`force_drill_parents` set. Sections in that set are drilled into
sub-sections regardless of whether V4 carries the parent exact key.
Top-level override targets (no derived parent) do not trigger
force-drill, so backward-compat is preserved for parent-granularity
overrides.
The call site in `run_phase_z2_mvp1` collects sub-ids from
`override_section_assignments` and forwards them to the aligner.
Generalization (RULE 0) :
- Trigger is the override schema (`X-sub-N`), not a specific MDX / section /
frame id. Applies to all 32-frame MDX uniformly.
- Decision is deterministic on the override target shape, independent of
V4 yaml content.
- Default (no override) path is unchanged byte-for-byte.
Side fixes (forward-only RULE 1 cleanup, no history rewrite) :
- `align_sections_to_v4_granularity` docstring rewritten in English
(overwrites the Korean docstring committed in 5191aca).
- Step 9 diagnostic comment quoted-string rewritten in English
(overwrites `"V4 entry 없음"` committed in a422d72).
Tests : 3 new cases in `test_phase_z2_subsection_schema.py` —
`test_align_parent_v4_exact_keeps_section_when_no_override_targets_sub`
(backward-compat axis), `test_align_force_drills_when_override_targets_sub_id_with_parent_in_v4`
(blocker regression), `test_align_top_level_override_target_does_not_force_drill_other_sections`
(force-drill scope guard). Pytest scope-qualified result :
`test_phase_z2_subsection_schema.py` + `_section_assignment_override.py` +
`_v4_fallback.py` = 40 / 40 PASS.
Smoke (axis = sub-id override -> aligner -> assignment plan, both V4 yaml
shapes) :
- HEAD V4 yaml (`04-1`, `04-2.1`, `04-2.2` only) :
`--override-section-assignment primary=04-2-sub-1` ->
`aligned_section_ids=['04-1', '04-2-sub-1', '04-2-sub-2']`,
`plan[0].assignment_source='cli_override'`,
`plan[0].source_section_ids=['04-2-sub-1']`.
- V4 yaml with `04-2` exact key (Codex's stress case) : identical
aligned output and identical assignment plan.
Downstream `composition_planner` abort
(`phase_z_status_not_allowed:extract_matched_zone`) is IMP-05 territory,
unchanged in both shapes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
