Auto-update could not work at all: the installer was 189 MB because it carried
the local speech engine and ffmpeg, and the download feed rejects uploads over
about 100 MiB, so update metadata could never be published.
The installer now leaves those components out and the app fetches them the first
time they are needed, verifying every part and the joined archive before
installing. The installer is 90.6 MiB, the update feed is published again, and
updates stay small because the engine is not re-sent on every release.
The fetch is visible and recoverable: the download runs with progress, a failed
install cleans up after itself, and Settings > STT shows the runtime status with
a manual download action for when the automatic one cannot run.
The manual install path still needed 7-Zip, which the target machine does not
have, so "installable without a certificate" was not yet true.
The channel now also publishes the app as byte-split zip parts, and the install
script joins them and extracts with the built-in Windows Expand-Archive after
verifying every part and the joined archive. Version 1.3.1 republishes the
channel from a single build, because a version's artifacts can only match one
build and published volumes are never overwritten.
Local dictation had never produced a transcript on an installed build. The
engine itself was healthy; every connection to it was broken.
Installed builds shipped no speech engine at all: the packaging config had no
entry for the faster-whisper sidecar and no pipeline step built one, so the app
always fell back to a system Python without the runtime. Development was broken
too, because the sidecar and SoX paths were resolved against the Vite output
directory instead of the app root, which also meant recording failed with a SoX
ENOENT. On hosts where localhost resolves only to IPv6, every local request was
refused outright, which silently disabled both local transcription and the local
LLM.
The sidecar is now built and bundled (including the Silero VAD data it needs),
gated by a packaging check that fails when the engine or its data is missing.
Paths are discovered from the app root and fail loudly when the engine is
absent. Local engine URLs are normalized to the IPv4 loopback, decoding is tuned
so repeated hallucinations cannot compound (the same transcript now takes about
a fifth of the time), the engine is warmed up at startup, and holding the hotkey
now shows the text forming live in the recording tip.
Version 1.1.0 was published on 2026-09-15, so its tag is closed. Every commit
since then, the update-feed rework, dictionary import and export, the extra
push transports, and entitlement gating, needs its own immutable version.
Product version, Android version code, iOS build number, and all package and
store surfaces move to 1.2.0 / 1020001, with Korean and English store notes for
the new version code. The download centers read that version and its release
date from one place instead of repeating them inline.
Desktop clients had two competing update sources: the runtime pointed at a
legacy GitLab registry while the Forgejo packages were filled in by
hardcoded, version-pinned scripts. Operators could not tell which feed was
authoritative, and no release could be reproduced from a tag.
Auto-update now reads a single canonical Forgejo registry feed, updated by
a version-agnostic publisher that runs from the tag on Forgejo, GitLab, and
GitHub CI alike. Channel, minimum supported version, forced install,
full-versus-delta thresholds, staged rollout, and a remote kill switch come
from one policy file the client fetches alongside the feed. Tag creation is
gated on a clean tree, matching version surfaces, and a changelog section.
RN 0.85 + React 19 기반 apps/mobile-rn 프로젝트 생성.
6개 화면 이식 (Login, Dash, History, Record, Talk, Settings).
@react-navigation/native + bottom-tabs 네비게이션 구성.
Windows NDK CMake libc++_shared 링크 버그 워크아라운드 포함.
Galaxy Fold (SM_F956N) 무선 디버깅 APK 설치 확인.
DS 컴포넌트 9개 (ScreenPanel, WaveBars, AppStatusBar, Header, FilterChip 신규),
5개 탭 + 로그인 화면 D3RO 인스트루먼트 미학 적용,
i18n 연결 (I18nProvider + 모바일 전용 키 80+개),
모노레포 workspaces에 apps/mobile 추가.
- V3 모바일 마스터 플랜 작성 (docs/v3/00-mobile-master-plan.md)
11개 Phase, 5주 로드맵, 전략 결정사항 확정
- 7개 디자인 HTML 레퍼런스 저장 (docs/v3/designs/)
- app.json → app.config.ts 전환 + 환경변수 체계
- 5탭 네비게이션 (DASH/HIST/REC FAB/TALK/SET)
- metro.config.js monorepo 격리 (root RN 0.84 충돌 해결)
- ui-native 테마 디자인 목업 색상 동기화
- SafeArea 적용 (useSafeAreaInsets)
- DEV 로그인 우회 (__DEV__ 전용)
- Supabase .env 연동
- iOS 시뮬레이터 검증 완료
[A] Dark/Light/6종 + Auto 테마 토글
- theme-mode-context.tsx: localStorage 영속 + prefers-color-scheme
- theme-provider.tsx: 컨텍스트 기반 동적 getTheme
- sidebar.tsx: MUI Select 토글 (PaletteIcon)
- 12개 locale theme.* 키 추가 (label/dark/light/auto/nord/solarized/catppuccin/dracula)
[B] DocumentEditor Mermaid + Markdown 렌더
- markdown-preview.tsx: react-markdown + remark-gfm + lazy mermaid
- mermaid 코드 블록 자동 SVG 렌더
- dark theme + d3ro 폰트
- document-editor.tsx: Edit/Preview Tabs
[C] 회의 오디오 재생
- meeting-audio-player.tsx: Supabase Storage signed URL (1h) → <audio controls>
- meetings/[id] AUDIO 카드 추가
[D] Dashboard 14일 추이 차트
- meetings-trend-chart.tsx: recharts LineChart (회의/문서)
- dashboard/page.tsx: 14일 데이터 집계 + 이번 주 stat 자동 계산
[W] fix: Emotion key 'd3ro-mui' → 'mui'
- 'd3ro-mui'에 숫자 3이 포함돼 Emotion 키 검증 실패
- dev SSR /login 500 에러 차단했음
deps:
- react-markdown@10, remark-gfm@4, mermaid@11, recharts@3
묶음 A — UI 정리 + i18n 완성:
[A1] desktop SettingsModal에 CloudSyncSection 통합 (탭 6번째)
- CloudIcon import, settings.tabs.cloud 키 추가
- 기존 about 탭은 index 5 -> 6
[A2] apps/web Sidebar 공유 layout 리팩터링
- app/(app)/layout.tsx 신규 (route group)
- dashboard/meetings/record/teams/billing을 (app)/ 아래로 git mv
- app/(app)/layout.tsx에 auth 가드 + Sidebar 통합
- 기존 개별 page.tsx에서 Sidebar/auth 중복 제거
- app/(app)/dashboard/layout.tsx 제거 (루트 layout이 처리)
[A3] 11개 locale에 V2 새 키 추가 (en/ja/zh/zh-TW/es/fr/de/pt/ru/vi/th)
- nav.meetings/record/teams/billing/logout
- login.subtitle/google/github/terms
- settings.tabs.cloud
묶음 B — V2-4b pull 동기화:
- CloudSyncService.pullAll() 신규
- history/dictionary 테이블 원격에서 fetch
- last_sync_at 이후 updated_at만 필터
- Last-Write-Wins 충돌 해결 (remote.updated_at > local.updated_at)
- 로컬에 없는 행은 INSERT, 있는 행은 UPDATE (구체적 컬럼 지정)
- IPC CLOUD_SYNC.PULL_ALL 채널 + handler + preload api
- CloudSyncSection에 Pull 버튼 추가 (Push 옆에 위치)
묶음 C — V2-8b Stripe webhook 서명 검증:
- stripe-webhook Edge Function에 Web Crypto API 기반 HMAC-SHA256 검증
- Stripe-Signature 헤더 파싱 (t=, v1= 엔트리)
- Replay 방지 (timestamp tolerance 300초)
- constantTimeEqual로 타이밍 공격 방지
- crypto.subtle.importKey/sign으로 HMAC 계산
- stripe-portal Edge Function 신규 (Customer Portal)
- JWT 인증 -> 기존 customer_id 조회 -> billing_portal/sessions 생성
- return_url 지원
- apps/web/components/billing/portal-button.tsx (구독 관리 버튼)
- billing 페이지에 Free 외 tier 사용자에게 PortalButton 표시
- config.toml에 stripe-portal 함수 등록 (verify_jwt=true)
묶음 D — V2-6b packages/ui-native:
- @d3ro/ui-native 신규 패키지 (React Native 전용 DS)
- theme.ts: d3roNativePalette/Typo/Radius (MUI 없는 정적 값)
- components/MetalCard.tsx: View + 섀시 섀도우
- components/PhosphorText.tsx: Text + 앰버 glow (textShadow)
- components/Led.tsx: View 원 + glow
- components/PhysicalButton.tsx: Pressable + 누름 느낌
- React/React-Native는 peerDependencies
- apps/mobile/package.json에 @d3ro/ui-native를 file: 의존성으로 추가
(mobile은 npm workspace 제외이므로 file path 필요)
검증:
- desktop typecheck OK
- web typecheck OK
- web next build OK (11 라우트, (app) 그룹 반영)
- 회귀 없음
이전 커밋(45a5808)에서 다음 파일들이 누락되어 추가:
- package.json: 루트가 monorepo workspace로 재구성된 내용
- tsconfig.json: apps/desktop references 구조
- package-lock.json: workspace install 결과
파일시스템에는 반영되어 있었으나 스테이징 누락으로 git 히스토리에서
빠져 있던 것을 보충. 체크아웃 시 빌드 깨지는 문제 방지.
PLAUD 수준의 기능 확장:
- 전사 편집: 인라인 편집 + 원본 보존 + 수정됨 표시(점선 밑줄)
- 다중 문서 생성: 회의록/보고서/아이디어노트/커스텀 템플릿
- MD 에디터: 렌더링/편집 토글 (react-markdown + remark-gfm)
- 상세 페이지 리디자인: 풀스크린 탭 전환 구조
- 내보내기: MD, PDF, TXT, DOCX 4종 (docx 패키지)
- 커스텀 프롬프트 템플릿 저장/재사용 (electron-store)
- SSOT: 마크다운 파싱 유틸 추출, 후처리에서 자동 문서 생성 제거
DB: meeting_documents 테이블 + edited_transcript 컬럼
IPC: 8+4 채널, 서비스 2개(MeetingModeService 확장 + MeetingDocTemplateService 신규)
UI: 8개 신규 컴포넌트 (meeting/ 디렉토리), 12개 locale i18n
Phase 12:
- FileTranscriptionService: ffmpeg PCM 변환 + 30초 청크 순차 STT
- MeetingSummaryService: 자막 세션 → LLM 자동 요약 + DB summaryText
- DictationTemplateService: 필드별 음성 입력 상태 머신 + 프리셋 3개
Phase 13.1:
- VoiceConversationService: STT→Ollama /api/chat→TTS 대화 루프 (10턴)
- TTSPlaybackService: Windows SAPI 문장 단위 큐 재생
- LocalLLMService.chatStream: Ollama /api/chat 스트리밍
Phase 13.2:
- RAGService: Ollama 임베딩 + SQLite 벡터 + 코사인 유사도 검색
- KnowledgeBasePage: 문서 관리 + 질문/답변 UI
- PDF 파서: zlib FlateDecode 해제 + BT/ET 텍스트 추출
Phase 13.3:
- VoiceActionService: LLM JSON 액션 플랜 생성 + 실행
- 프리셋 6개 (크롬/메모장/탐색기/볼륨), 위험 명령 차단
공통: IPC ~70채널, 에러코드 780-878, i18n 100+키
버그픽스: 라이선스 로컬 키 우선, i18n featureLabel, DOM 중첩
- DB: better-sqlite3 + drizzle-orm (history/dictionary/stats, WAL 모드)
- HistoryService: CRUD + 검색 + 통계 + 30일 보존 정책, 세션 완료 시 자동 저장
- DictionaryService: CRUD + 검색 + 사용 횟수 추적 + STT 프롬프트 힌트
- HistoryPage: 목록 + 검색 + 삭제 + 복사 + 페이지네이션
- DictionaryPage: 목록 + 검색 + 추가 다이얼로그 + 삭제
- DashboardPage: 실데이터 통계 (총/오늘 세션, 시간, 단어수, 연속일수)
- IPC: history/dictionary/stats 핸들러 + preload API
- Bootstrap: DB 초기화 (critical) + 이력 자동 저장 연동
