Marks F2 in dogfood-feedback as 🚀 promoted and adds the standalone
spec at docs/superpowers/specs/2026-04-26-f2-tag-click.md capturing
the mini-brainstorm decisions, scope, and follow-ups (multi-tag
filter, rename/merge, source preservation on undo).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Splits the tag chip into two actions per F2 dogfood feedback:
- short click on chip text → applies the tag to the inbox filter
(Inbox header shows "🔎 필터: #tag (n개)" banner with ✕ 해제 button)
- × button on chip → immediately removes the tag and surfaces a
module-level pub/sub undo toast for 5 seconds; clicking the toast
restores the tag
`TagUndoToast` is a tiny self-contained component: `pushTagUndo()` from
NoteCard publishes an entry, the mounted `<TagUndoToast />` near the
end of `<App>` subscribes and renders it. Auto-dismiss after 5 s,
click-to-undo cancels the timer and runs the restore callback.
AI vs user tags share the same behaviour — only the chip background
colour distinguishes them, matching the F2 decision table.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds `tagFilter: string | null` and `setTagFilter` to the inbox store, plus
an extracted pure `selectFilteredNotes` selector so unit tests can import it
under vitest's `node` environment without dragging `api.ts` (which touches
`window.inkling` at module load).
Tests cover four cases: null filter passes through, single-tag match,
no-match empty result, and any-tag-matches semantics.
F2 dogfood feedback step 1/3.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Extracted to own spec with mini-brainstorm decisions captured.
F1 in dogfood-feedback.md marked 🚀 promoted with link.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Inline 📅 YYYY-MM-DD badge appears in NoteCard between summary and
tags. Click to edit (HTML date input). Past dates: gray + line-through.
AI label shown when not user-edited (mirrors title/summary AI badge
policy). Empty state shows '📅 마감일 추가' link in gray.
New IPC inbox:setDueDate routes to NoteRepository.setDueDate which
sets due_date_edited_by_user=1 (per slice §1.1 invariant 2 — user
edit blocks future AI overwrite). Preload bridge + InboxApi type
extended.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
GenerateInput gains todayKst field. AiWorker computes KST-aligned
date once per job, runs parseDueDate on rawText, calls provider.generate
with todayKst, then merges: rule.iso wins if matched (deterministic),
else AI's due_date, else null. Logs dueDateSource (rule|ai|none) for
debugging. now() injection for testability.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
AiResponse extends with dueDate: string|null. zod regex
^\d{4}-\d{2}-\d{2}$, follow-up roundtrip check coerces invalid
dates (2026-13-99 etc.) to null. PROMPT_VERSION → 2: prompt now
takes todayKst arg, asks model to extract due_date as ISO or null.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Regex + KST math, returns ISO YYYY-MM-DD or null. 14 high-confidence
rules (literal date, N월 N일, MM/DD, N일/주/개월 뒤, 모레/내일/글피/오늘,
다음/이번 주 X요일, 다음 달). Ambiguous tokens (월말, 주말, 퇴근 전,
시각) return iso=null + confidence='medium' so caller (AiWorker)
can defer to AI. 26+ golden fixtures.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three-state outcome per note: inserted (new id), skipped (id+rawText
match), forked (id match but rawText differs → new uuidv7 to preserve
raw_text invariant from slice §1.1). Media files copied into
MediaStore convention <profileDir>/media/{noteId}/{n}.{ext} with
new media DB rows.
NoteRepository.importNote handles full provenance: ai_status='done',
ai_provider, ai_generated_at, edited flags, intent fields, tags
with source preserved.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Extracted to own spec with mini-brainstorm decisions captured.
F5 in dogfood-feedback.md marked 🚀 promoted with link.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Tray now has 4th callback that opens directory chooser, exports all
notes via ExportService with includeMedia=true default. Dialog
message warns about raw_text plain-text + recommends private location.
Native toast on success/failure with note + media counts.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
ExportService composes pure exportFormat layer + reads from
NoteRepository.listAll (new, asc-ordered) + MediaStore.absolutePath
(new helper). Writes notes/{date-id8-slug.md}, media/{id8__n.ext},
index.jsonl, manifest.json, README.md to user-picked dir.
6 unit tests against tmp dirs + :memory: DB.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
final reviewer 가 식별한 forward-looking polish 4건을 후속 리스트에
명시. 4번째는 F1 (Due Date 마이그레이션 v2) PR 시 즉시 반영 권장:
openDb() 가 마이그레이션을 BackupService 인스턴스화 전 호출하므로
마이그레이션 결함 시 첫 실행 직전 상태 회수 불가.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Extracted to its own spec, dogfood-feedback.md F6 header reflects
L1 promoted status while L2/L3 remain raw.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Instantiate BackupService at app.whenReady, run daily snapshot then
again before quit (synchronous-blocking via preventDefault). Tray menu
gets '지금 백업' entry that triggers manual runDaily with native
toast feedback.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Skips when marker matches today's KST date. Marker written after
successful snapshot, before rotation. lastSnapshotAt() exposed for UI.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Code review I1: wrap snapshot's backup+rename in try/catch, unlink
orphan tmp file on failure so the next run does not encounter a
confusing 'existing file is not a database' error from sqlite.
Code review I2: JSDoc note that snapshot() is not safe for concurrent
calls — callers should serialize via runDaily()'s marker.
New unit test injects a fake db whose backup() rejects after a partial
write, asserts no .tmp / .sqlite remains.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Code reviewer minor nitpicks:
- Add test for inkling-2026-02-30.sqlite (locks roundtrip-validation contract)
- Add test for weekly window inclusive at oldest boundary
- Precompute today=startOfDayUtc(now) once outside the loop, pass to helpers
No behavior change; tests added cover existing semantics.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Spec reviewer flagged the weekly window keeps 5 Mondays (anchor + 4
prior), not 4 as the plan prose said. Code preserved (tests pass);
constant renamed and comment made honest about the actual semantic.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 1: 순수 GFS retention 함수 + 7 단위 테스트
Task 2: BackupService.snapshot() — KST 날짜·tmp+rename 원자성 + 6 단위 테스트
Task 3: runDaily() — .last-snapshot 마커 + lastSnapshotAt + 7 단위 테스트
Task 4: main/index.ts wiring (whenReady + before-quit) + tray '지금 백업'
Task 5: F6-L1 promotion (별 spec 분기 + dogfood-feedback.md 상태 갱신)
backup 위치: <profileDir>/backups/ (mini-brainstorm 결과 A 채택).
스키마 변경 0, 외부 dep 0. better-sqlite3.backup() API 가정.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
8개 항목 순차 작업 (F6-L1 → F5 → F6-L3 → F1 → F2 → F3+F4-E →
F6-L2 → F4-C·F) + 데이터 안전 우선 + 머지+테스트 게이트 + 단일
v0.2.1 cut 후 dogfood 재설치 + 1주 soak. F4-A·D 는 측정 후
별 brainstorm 으로 deferred. 항목별 mini-brainstorm 에서
decision-pending 답변하는 라이프사이클.
본 spec 은 순서·범위·게이트만 정의, 항목 내부 설계는 per-item.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
slice + strategy.md §3 가 회의·퇴근 같은 강한 contextual cue 만 다루고
샤워·산책·자기 직전 같은 ambient 떠오름은 사각지대. 6개 심리 메커니즘
(habit stacking, ambient if-then, 환경 앵커, variable interval prompt,
Zeigarnik, 정체성 고리) 후보 + 슬라이스 내/후속 분류 + H1~H5 가설.
권장 순서: E (카피 priming) → 데이터 → C·F → A·D.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
dogfood 첫 알림 노출에서 '구출' 표현이 일상 한국어로 부자연스럽다는
피드백. 5개 UI 표면 + e2e 단언 + strategy.md §1·§3·§7 의 어휘 결정에
함께 묶임. drafting 시 결정 대기 #1(strategy 재검토 동반 여부)이
promoted 경로 좌우.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
v4/v7 imports stay API-compatible — typecheck and unit suite (52/52)
both pass against the new resolution. The caret range deviates from
the slice's strict-pin invariant (spec §7.1); revisit if reproducible
builds become an issue.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Reframe the file as a living register for in-flight dogfood feedback
rather than a single-feature stub. Each item carries a status label
(raw / drafting / ready-for-spec / promoted / rejected) and a fixed
six-slot template (관찰 / 제안 방향 / 결정 대기 / 가설·측정 / 범위 /
영향). Items graduate to their own spec file once mature; the entry
here then collapses to a one-line link.
- F1: Due-date 추출 (drafting) — content from the previous stub
normalized to the new template.
- F2: 태그 클릭 = 즉시 삭제 + undo 부재 (raw) — NoteCard.tsx:110
binds chip onClick to removeTag, no confirm or undo, and there is
no "filter by tag" affordance to match the user's mental model.
Direction: split click=filter, ✕-icon=remove with 5s undo toast.
README docs map updated to point at the new path.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The due-date extraction work originated as feedback during dogfood, but
the dogfood strategy doc itself should stay feature-agnostic — it's the
generic operating manual for the 2-week dogfood, not a feedback log.
- Remove the "시간 표현 포함 노트 수" row from dogfood-strategy §3.1.
- Rephrase the due-date spec stub so H1 / §7 / §9 reference the spec's
own sample-collection plan instead of relying on the dogfood retro.
- Spec is now framed as "independent of slice exit"; entry timing is a
separate decision once an accumulated sample meets H1.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Captures the design skeleton, hypotheses, and open questions for the
hybrid (rule-based first → AI fallback) Korean date-phrase extractor.
Schema impact, prompt change, UX copy, and library tradeoffs are
sketched but deferred — formalization waits for slice v0.4 dogfood
Pass and the H1 (time-phrase frequency ≥ 30%) check in week-1 retro.
Cross-link added to dogfood-strategy daily-metric table so the H1
data is collected during dogfood, and to README docs map.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When the OllamaBanner appears, it was generic enough that the user
couldn't tell whether the env var was missing, the LAN host was
unreachable, the model was uninstalled, or DNS was failing. Log the
resolved endpoint at startup with a `fromEnv` flag so we can confirm
INKLING_OLLAMA_ENDPOINT was actually read, and render the underlying
health-check reason as a small subtitle under the banner copy.
The user-facing primary message still avoids the forbidden tone words
("실패"/"끊김"/"연속 실패"); the diagnostic line is technical and only
appears when status.reason is set.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Vitest needs the node-ABI prebuilt sqlite binary; Electron 41 needs ABI 145.
Add `rebuild:node` / `rebuild:electron` scripts and wire them as
pre-hooks for `test`, `test:e2e`, `start`, `dev`, `test:integration`.
- Smoke test was racing on `firstWindow()`, which non-deterministically
returned the quickcapture window. Strip ELECTRON_RUN_AS_NODE from the
launch env, wait for both windows to register, then pick the inbox by
title and assert the heading via `getByRole`.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Follow-up to the CJS refactor: destructuring electron.BrowserWindow /
electron.Tray pulled the values into scope but lost the type
namespace, breaking sites that used the bare class name as a type
(let win: BrowserWindow | null, return-type annotations,
implicitly-any close handlers). Added 'import type { ... as ...Type }'
for each affected file and replaced the type-position references
with the aliased names. Runtime semantics unchanged; typecheck
exits 0 again.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 32 of the slice plan. playwright.config.ts (testDir
tests/e2e, single worker, 60s default timeout, list reporter)
and a smoke spec that launches the built main bundle, waits
for the inbox renderer, and asserts the v0.2 empty-state copy
'첫 기억을 구출해보세요.' is rendered.
Build verified end-to-end (`npm run build` exits 0, produces
out/main/index.js + out/preload/index.mjs + the inbox /
quickcapture renderer chunks). Playwright's chromium bundle
installed.
Known issue (deferred to a follow-up debug session): on this
specific Windows machine, electron@41.3.0's main-process
module hook is not injecting the real `electron` API into the
launched bundle. `require('electron')` from the entry script
returns the on-disk path string (the package's index.js
default), and `process.electronBinding` / `process.type` are
undefined inside the spawned process. This reproduces with a
minimal package.json + main.js outside this repo — i.e. it's
not a slice bug. The smoke test currently fails with
'Process failed to launch!' at the
`app.whenReady().then(...)` line because `app` is undefined.
When the host issue is sorted, no test changes should be
required: build, config, and spec are correct. Likely fixes:
clean-reinstall electron + node-modules, try electron@latest,
or invoke through `npx electron` rather than the binary
directly. Captured in project_inkling_status memory under
'open issues'.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two related runtime defects surfaced when first attempting
`npm run build` + Electron launch:
1. Renderer html files referenced `/src/.../main.tsx`, which
vite dev resolves but vite production rollup cannot. Changed
both inbox and quickcapture to `./main.tsx` (sibling-relative).
2. Electron 41's main-process module hook only injects the
real `electron` API into `require('electron')` calls inside
CommonJS modules. With package.json `"type": "module"` set,
electron-vite emitted ESM bundles that did
`import { app } from "electron"`; Node's ESM CJS-interop
then loaded the on-disk index.js (which exports the binary
path string) instead, leaving `app` undefined at startup.
Fix: drop `"type": "module"` so electron-vite emits CJS for
main + preload (renderer is unaffected — vite handles its own
ESM transform regardless). Source files keep modern import
syntax via the default-import + destructure pattern
(`import electron from 'electron'; const { app } = electron;`)
which transpiles correctly to `const { app } = require('electron')`
in the CJS output.
Also updated `electron-log/main` -> `electron-log/main.js`
because Node ESM strict resolution requires the `.js` extension
even though the package's CJS entry serves both.
Verification: `npm run build` produces 47-module renderer +
1005KB main bundle without errors. `npm run typecheck` clean.
Unit suite (52 tests) unaffected.
Plan deviation log: Task 4 (logger), Task 30 (main wiring), and
Tasks 17/18/22/30 referencing electron imports landed with named
imports per plan; this commit refactors them to default+destructure
without changing semantics.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 31 of the slice plan. Diff list of media/ subdirectories
against current note ids (SELECT id FROM notes); deletes
directories whose noteId is no longer referenced. Runs once on
startup from main; returns { removed } so logger captures the
count. Covers the 'media delete failure leaves orphan' branch
of spec §6.1 row 8.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 30 of the slice plan. Replaces the Task 2 placeholder main
entry with the final whenReady wiring:
- profile path resolution + better-sqlite3 open + migrations.
- repo / store / continuity / intent / health constructed
against the open db.
- LocalOllamaProvider reads INKLING_OLLAMA_ENDPOINT for LAN
dogfood, falls back to localhost:11434 otherwise.
- AiWorker registers an onUpdate that fans note:updated through
pushNoteUpdated(getInboxWindow, note).
- NotificationService is plumbed to electron.Notification with
isSupported gating; CaptureService gets the worker.enqueue +
notify.celebrate hooks.
- IPC bindings for both capture and inbox surfaces.
- HotkeyService registers Ctrl/Cmd+Shift+J -> showQuickCapture;
failure logged but not fatal.
- Tray menu '구출한 메모 보기' / '기억 구출하기' / '종료'
with click-to-show-inbox.
- worker.loadFromDb() resumes pending jobs at startup.
- MediaGc runs once and logs the count of removed orphan dirs.
- Two logger.info(..., {...obj} as Record<string, unknown>)
casts at the health and gc result sites to satisfy the
index-signature requirement on logger meta — the result
objects (HealthResult, {removed: number}) are assignment-
compatible with Record<string, unknown> at runtime but TS
refuses without the spread + cast.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 29 of the slice plan. Subscribes to ollamaStatus from the
inbox store and renders one of two messages: 'ollama pull
gemma4:e4b 실행 후 앱을 재시작해주세요.' when the model isn't
installed (status.reason includes 'not installed'), or
'Inkling 정리가 잠시 멈췄습니다. Ollama를 실행해주세요.' for
any other unhealthy state. Capture continues regardless.
HealthChecker.ts itself was pulled forward in Task 21 so its
import would resolve from inboxApi.ts at compile time.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
