028042aaa9
feat( #68 ): IMP-39 u1~u8 ranking_sort_policy single-source + backend↔frontend label-priority mirror
...
Multi-MDX Regression (IMP-91) / multi-mdx-regression (push) Failing after 23s
u1: templates/phase_z2/catalog/ranking_sort_policy.yaml — single-source policy
(label_priority asc {use_as_is:0, light_edit:1, restructure:2, reject:3}
+ confidence desc + v4_rank asc tie-break).
u2: src/phase_z2_pipeline.py — apply_ranking_sort helper + lookup_v4_match_with_fallback
applies policy AFTER IMP-38 raw-window selection (raw default_window + usable_count
preserved on RAW all_judgments).
u3: src/phase_z2_pipeline.py — _build_application_plan_unit forwards ranking_sort_policy
+ sorted_candidate_evidence into Step 9 payload.
u4: Front/client/src/services/designAgentApi.ts — frame_candidates builder reads
unit.sorted_candidate_evidence + unit.ranking_sort_policy first; local LABEL_PRIORITY
retained only on warn-fallback path.
u5: tests/test_ranking_sort_policy.py — pure permutation coverage (sample-agnostic).
u6: tests/phase_z2/test_label_priority_synthetic.py + fixtures/ranking_sort_policy/
synthetic_divergence.yaml — low-conf use_as_is behind high-conf restructure.
u7: tests/phase_z2/test_imp39_mdx04_env_toggle_e2e.py — samples/mdx_batch/04.mdx with
AI_FALLBACK_ENABLED=off; backend selected_v4_rank == frontend frame_candidates[0].
u8: tests/phase_z2/test_imp39_corpus_audit.py — real corpus sweep over
tests/matching/v4_full32_result.yaml (10 MDX sections); section IDs loaded
dynamically (RULE 0 / RULE 7 sample-agnostic).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com >
2026-05-24 17:12:07 +09:00
896f273ffa
feat( #92 ): IMP-92 u1~u5 AI fallback config validation (model ping + operational error classification)
...
Replaces #84 UI-noise removal plan with positive operational-alert contract.
Five-axis stack lands together: (1) default model literal moved to current
Opus-family ID, (2) Anthropic SDK error classifier mapping exceptions to
quota/billing/auth/other, (3) api_error_kind plumbed through ai_repair_status
summary + per-record retention, (4) Step 0 preflight ping gated under
ai_fallback_enabled (default OFF preserved) with fail-fast on invalid
model/key, (5) frontend formatter rewritten to surface only operational
quota/billing/auth toasts (non-operational paths return null per
feedback_auto_pipeline_first silent-pipeline policy).
u1 - default model literal claude-opus-4-6-20250415 -> claude-opus-4-7
(src/config.py + tests/test_phase_z2_ai_fallback_config.py lock mirror)
u2 - classify_operational_error type+status_code dispatch + Step 12
api_error_kind stamp on except path (src/phase_z2_ai_fallback/client.py
+ src/phase_z2_ai_fallback/step12.py + tests/phase_z2_ai_fallback/test_step12.py)
u3 - _summarize_ai_repair_status aggregates api_error_kinds {quota,billing,
auth,other}; error_records[i].api_error_kind retained per-record
(src/phase_z2_pipeline.py + tests/test_imp47b_failure_surface.py)
u4 - _run_step0_ai_preflight + Step0PreflightError; preflight only fires
when ai_fallback_enabled=true; one-token ping; invalid key/model =>
setup failure before Step 1 (src/phase_z2_pipeline.py +
tests/phase_z2/test_pipeline_step0_preflight.py NEW)
u5 - AiRepairStatus.api_error_kinds? interface + formatAiRepairHumanReview
Message rewritten: operational quota/billing/auth -> Korean copy
verbatim from issue body (tie-break quota -> billing -> auth);
validation/coverage_violated/unsupported_kind/generic-other/legacy
payload -> null (Front/client/src/services/designAgentApi.ts +
Front/client/tests/imp47b_human_review_toast.test.tsx)
Guardrails respected:
- feedback_demo_env_toggle_policy: default OFF preserved; preflight skipped
when ai_fallback_enabled=false (test_preflight_skipped_when_disabled
asserts anthropic.Anthropic() not called).
- feedback_auto_pipeline_first: non-operational AI failures stay silent;
only quota/billing/auth reach user toast.
- feedback_ai_isolation_contract: AI remains fallback-only; no normal-path
migration; MDX preserved.
- project_imp46_carveout_caveat: cache_key/fingerprints fields untouched on
every record; no overlap with #62 cache region.
- feedback_no_hardcoding: zero MDX-sample-specific literals; classifier
dispatch by SDK type, not by string parsing.
- feedback_artifact_status_naming: operational toast scoped to alert axis,
not overall PASS signal.
Tests:
- Targeted u1+u2+u3+u4: 63 passed
- u5 vitest (Front/): 10/10 passed
- tests/phase_z2_ai_fallback dir regression: 240 passed
- tests/phase_z2 dir regression: 323 passed
- IMP-92-adjacent (-k "imp47b or ai_fallback or preflight or step12 or step0"): 299 passed (808 deselected)
- u1 baseline lock (test_client_mock.py): 8 passed
Zero failures, zero regressions outside scope.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com >
2026-05-23 22:07:25 +09:00
2ef02f5f18
feat( #76 ): IMP-47B u11 frontend human_review surfacing (hunk-split from IMP-41)
...
- AiRepairStatus interface mirrors backend step20 u8 schema
- formatAiRepairHumanReviewMessage(): pure helper for the three failure axes
(error / coverage_violated / unsupported_kind) — null on success/no-AI
- Home.tsx: toast.error(aiReviewMsg) after run completion
- FramePanel.tsx: reject-click window.confirm guard ("frame 유지 + AI 재구성")
- imp47b_human_review_toast.test.tsx: 6 vitest cases (null/false/3 axes/other)
Verification (frontend node_modules junction from main worktree):
- vitest imp47b_human_review_toast.test.tsx: 6/6 passed
- vitest full suite: 19/19 passed (imp41_application_mode 13 + u11 6, zero regression)
Hunk-split rationale:
- stash@{0} (imp47b-frontend-u11-pre-rebase, captured before IMP-41 merged)
contained inline IMP-41 helpers alongside u11 changes
- HEAD already has IMP-41 helper-based implementation (buildBadgeTitle /
mergeApplicationCandidates from services/applicationMode.ts, f358604noreply@anthropic.com >
2026-05-22 00:34:32 +09:00
f358604fb3
feat( #70 ): IMP-41 application_mode forwarding to FramePanel V4 badge tooltip (u1~u5)
...
Forward backend Step 9 `unit.application_candidates[]` (application_mode /
auto_applicable / delegated_to) onto FrameCandidate and surface the
application_mode as a Korean consequence phrase in the FramePanel V4-label
inline badge tooltip. Deterministic frontend-only refactor; no LLM call,
no V4-label color change, no outer composedTitle change.
u1: types/designAgent.ts — add optional applicationMode / autoApplicable /
delegatedTo on FrameCandidate (legacy fixtures keep undefined).
u2: services/applicationMode.ts (new) — pure helper exporting
ApplicationMode union, APPLICATION_MODE_TOOLTIP_KR (keyed by backend
mode VALUE, NOT V4 label), buildBadgeTitle, mergeApplicationCandidates.
u3: tests/imp41_application_mode.test.ts (new) — 13 Vitest cases pinning
composite output per mode, undefined/unknown→legacy fallback, merge by
template_id, skip missing/empty/non-string keys, first-wins on dupes,
empty/null/non-array input.
u4: services/designAgentApi.ts — bridge consumes mergeApplicationCandidates
and forwards three fields onto FrameCandidate while preserving
LABEL_PRIORITY sort and TOP_N_FRAMES slicing.
u5: components/FramePanel.tsx — V4-label badge `title` now calls
`buildBadgeTitle(candidate.label, candidate.applicationMode)`;
badge color className map preserved verbatim; outer composedTitle
untouched.
Scope-qualified verification (5 files, IMP-41 axis only):
- Vitest: client/tests/imp41_application_mode.test.ts — 13/13 PASS.
- Diff↔Plan parity: 5 files match Stage 2 plan, no scope creep.
- AI-isolation contract honored: tooltip values originate from backend
enum; no frontend re-derivation from V4 label.
- No spacing/font shrink; clipping resolution stays at layout/zone/frame
layer (feedback_phase_z_spacing_direction).
Pre-existing unrelated diagnostics (BottomActions.tsx,
imp47b_human_review_toast.test.tsx) remain open on their own axes and are
not gated by this commit.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com >
2026-05-22 00:17:32 +09:00
15ef7c65e9
fix( #75 ): IMP-47A mdx03 frontend execution stabilization (u1~u4)
...
u1: SlideCanvas iframe sandbox += allow-scripts (allow-same-origin preserved)
→ embedded-mode script in slide_base.html now applies html.embedded
→ standalone CSS reset deactivates inside iframe; no clipping
u2: designAgentApi.loadRun merges candidate_evidence + v4_all_judgments
+ v4_candidates via Map<template_id|id|frame_id> dedup,
LABEL_PRIORITY (use_as_is<light_edit<restructure<reject) then
confidence desc, capped TOP_N_FRAMES=6
u3: Home.handleGenerate useCallback deps = [uploadedFile, slidePlan,
userSelection, pendingZones, pendingLayout] (5-tuple, stale-closure fix)
u4: tests/manual/imp47a_e2e.md — mdx03 manual e2e spec (5 axes)
Frontend-only. Backend src/ untouched. No template/catalog edits.
Determinism preserved (no LLM in frontend merge logic).
Baseline: pytest -q tests → 623 passed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com >
2026-05-21 14:56:56 +09:00
b4872ba6ce
feat( #38 ): IMP-29 frontend zone-level evidence bridge (candidate_evidence reader + types + UI)
2026-05-20 21:53:47 +09:00
a79bd8bc43
feat(IMP-11): D-2 — frame min_height_px hint (backend → UI)
...
Step 9 v4_all_judgments[] now exposes per-candidate min_height_px from
catalog frame_contracts.visual_hints.min_height_px (None when contract
unregistered). SlideCanvas pendingLayout zones render a red ring + 'min H
Npx' badge when zone height falls below the active frame's threshold.
Visual hint only; resize clamp (minSize=0.05) unchanged.
5 axes (single commit per Stage 5 plan):
- u1 backend: src/phase_z2_pipeline.py — Step 9 builder adds min_height_px
via single get_contract(c.template_id) lookup; reuses _contract for
catalog_registered (no double-lookup).
- u2 type: Front/client/src/types/designAgent.ts — FrameCandidate gains
optional minHeightPx?: number.
- u3 mapper: Front/client/src/services/designAgentApi.ts — maps snake-case
min_height_px → camelCase minHeightPx on v4_all_judgments path;
v4_candidates fallback remains undefined (graceful).
- u4 active-frame lookup: Front/client/src/components/SlideCanvas.tsx —
activeFrameId = overrideFrameId ?? defaultFrameId; activeCandidate via
region.frame_candidates.find.
- u5 hint render: Front/client/src/components/SlideCanvas.tsx —
zoneHeightPx = height * SLIDE_H (logical px, no double-apply); compare
against activeCandidate.minHeightPx in pendingLayout mode only; red
border + badge when below.
Tests: 5/5 pass in tests/test_phase_z2_step9_v4_all_judgments_min_height.py
(source-string + catalog-shape guards + None propagation, registered and
unregistered template_ids).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com >
2026-05-17 22:29:17 +09:00
0fb168befc
feat(IMP-10): D-1 — filtered_section_reasons UI (read-only)
...
Surface step20_slide_status.json.data.filtered_section_reasons in the
frontend Home header. Verbatim mirror of backend payload — no enum
redefinition, no translation, no auto-classification.
Units:
- u1: FilteredSectionReason interface mirroring src/phase_z2_pipeline.py
:2217-2278 (10 fields incl. override-uncovered source/position variant).
- u2: RunMeta extension + loadRun() mapping with ?? [] back-compat defaults.
- u3: Header badge + <details> disclosure adjacent to existing status
badge; hidden when filtered_section_ids.length === 0; renders all 10
schema fields + filter_reasons[] verbatim.
Scope:
- Frontend-only, read-only. No backend / sync script / Kei·AI panel
changes. Files: Front/client/src/services/designAgentApi.ts (+20),
Front/client/src/pages/Home.tsx (+25).
Refs: gitea issue #10 (IMP-10 D-1 filtered_section_reasons UI)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com >
2026-05-17 19:43:13 +09:00
ab2764c8d0
feat(IMP-08): U3 — frontend wire (zoneSections override)
...
Wires the frontend drag/drop zone assignment through to the backend
--override-section-assignment CLI flag.
PipelineOverrides gains an optional zoneSections field
(Record<string, string[]>) carrying canonical ordinal section ids
(e.g., "top": ["04-2-sub-1"]).
Vite middleware /api/run accepts overrides.zoneSections and forwards
each non-empty zone as `--override-section-assignment ZONE=sid[,sid]`.
Empty arrays and non-string sids are filtered to avoid bogus
assignments from a partially-built UI state.
Home.tsx builds the override with a diff-vs-default guard per Codex
Stage 3 R3 B3 fix : createInitialUserSelection seeds zone_sections with
the auto plan, so a literal copy would pollute backend assignment-source
provenance even on a fresh re-render. The diff compares each zone's
section list against sourcePlan.zones[].section_ids and only emits zones
that differ. Toast summary now reports zoneSections=N when forwarded.
Smoke verification : python -m src.phase_z2_pipeline samples/mdx_batch/04.mdx
test_imp08_smoke --override-section-assignment primary=04-2-sub-1 produces
section_assignment_plan with assignment_source=cli_override and
v4_selector_trace.candidates populated via the U1 alias resolver
(04-2-sub-1 -> 04-2.1 V4 entry).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com >
2026-05-15 22:36:16 +09:00
0f0d3fa91f
feat(frontend): add Front/ — Vite/React frontend with backend pipeline integration
...
Mirror of design_agent_front/design-agent/ for shipping alongside backend.
Vite plugin (vitePluginPhaseZApi) endpoints :
- POST /api/run — spawn `python -m src.phase_z2_pipeline` with overrides
- GET /api/sample-mdx?mdx=03/04/05 — fixed sample MDX
- GET /frame-preview/{n} — figma preview thumbnails
- GET /data/runs/{run_id}/{path} — pipeline artifacts (final.html, step*.json, ...)
Env toggle forward (보고용) :
PHASE_Z_ALLOW_RESTRUCTURE / PHASE_Z_ALLOW_REJECT / PHASE_Z_MAX_RANK=32
Components :
- LeftMdxPanel (03/04/05 fix list + section tree)
- SlideCanvas (iframe + slideOverrideCss prop for inline CSS inject)
- FramePanel (label priority + confidence sort)
- LayoutPanel
README with mermaid diagrams covering the 5-step demo flow.
node_modules / dist / .manus-logs / .env excluded via .gitignore.
2026-05-14 14:48:42 +09:00
