Two small Phase 1 follow-ups for the Android target:
1. tauri.conf.json: add `bundle.android.minSdkVersion: 24`. Android 7.0
is the floor — gives us Vulkan availability (for the eventual GPU
feature flag), AAudio for cpal, and is what Pixel-class hardware
tests against. Keeps the global `identifier` on `uk.co.corbel.kon`
for now; the Corbie rebrand sweep will land `corbel.technology.corbie`
as a single coherent commit.
2. src/lib/utils/runtime.ts: add `isAndroid()` and `isMobile()` helpers
alongside the existing `hasTauriRuntime()`. Both use UA sniffing —
sufficient for feature-gating UI, never for security decisions.
These are how the Svelte side will hide:
- hotkey config (no global hotkey API on Android)
- paste-mode picker (auto-paste maps to a copy-only flow)
- meeting auto-capture toggle (process list unavailable)
- multi-window buttons (open-viewer, open-float, etc.)
- system-tray-related affordances
Tauri 2 doesn't expose a synchronous platform-detection helper that
works during initial render, so UA sniffing is the pragmatic choice.
https://claude.ai/code/session_0189xUb6ie6t9qHkzatGZ9Rb
The Phase 9 bulk export utility had a single success toast that was emitted
even when zero of N writes succeeded — "Exported 0 of 5 transcripts to
folder/" reading like the user just deliberately exported nothing.
Branch on the result count:
- 0 of N: error toast pointing at the console for write failures.
- N of N: success toast.
- M of N: warn toast — partial export, with the same console pointer.
Single-file save (`saveTranscriptAsMarkdown`) was already correct:
explicit success on save, error on failure, silent on user-cancelled —
left untouched.
https://claude.ai/code/session_0189xUb6ie6t9qHkzatGZ9Rb
Previously the 15-second meeting-detection setInterval was started in
onMount unconditionally (when Tauri runtime was available). When
`settings.meetingAutoCapture` was disabled the callback still fired every
15 s, just to early-return — burning a wakeup that did no useful work
and confusing "is this firing? did the toggle take effect?" debugging.
Move the timer into a `$effect` whose only tracked dependency is
`settings.meetingAutoCapture`. Toggling off now clears the interval; toggling
on creates a fresh one. Reads of `meetingAutoCaptureApps` and `globalHotkey`
happen inside the interval callback (post-setup) so they don't trigger the
effect to tear down on every keystroke in the apps editor.
The `meetingCapturePoller` variable and its `onDestroy` cleanup are gone —
the effect's own cleanup return takes care of it on unmount.
https://claude.ai/code/session_0189xUb6ie6t9qHkzatGZ9Rb
Error toasts are sticky (duration: 0) so a misbehaving command that fires
errors in a loop — a backend that flaps, a polling effect over a broken
endpoint — accumulates toast items in the in-memory store indefinitely.
The audit found no other unbounded $state arrays in the frontend stores
(history caps at 500, recentNudges prunes by time, tasks/taskLists
replace-on-load), but `items.push(toast)` had no upper bound.
Add MAX_TOASTS = 50 with FIFO eviction. Doesn't change behaviour for
realistic toast volumes; only kicks in if something is genuinely wrong.
https://claude.ai/code/session_0189xUb6ie6t9qHkzatGZ9Rb
HistoryPage previously serialised the full TranscriptDto — text, segments,
manual + LLM tags, audio path — into `localStorage["kon_viewer_item"]` so
the viewer window could pick it up on mount. On a multi-hour transcript
that's MB-scale of user voice content sitting in storage that any
same-origin script in any open Kon window can read.
Hand off only `{ id }` (and a timestamp on re-saves). The viewer fetches
the canonical row from SQLite via the existing `get_transcript` Tauri
command and hydrates via the now-exported `mapTranscriptRow`. Cross-window
sync via the `storage` event still works — the receiving window re-fetches
on event instead of trusting the payload.
- HistoryPage `openViewer` + `openEditor`: write `{ id }` only.
- viewer `onMount` + `handleStorageChange`: route through new
`loadFromHandoff` which calls `invoke("get_transcript", { id })`.
- viewer `saveItemToHistory`: re-stamp localStorage with `{ id, stamp }`
to retrigger the storage event in sibling windows without leaking
content.
- `mapTranscriptRow` exported from page.svelte.ts for the viewer's use.
Backward-compatible at the parse layer: the `{ id }` shape extracts cleanly
from a stale full-DTO payload (TranscriptEntry already carries `id` at top
level), so a session that survives the upgrade picks up the new path on
next handoff without manual cleanup.
https://claude.ai/code/session_0189xUb6ie6t9qHkzatGZ9Rb
The Phase 9 bulk-delete path passed UUID strings to deleteFromHistory(index),
which expected an integer; JS coerced the string to NaN and splice(NaN, 1)
collapsed to splice(0, 1), so bulk-delete silently removed the first N visible
rows instead of the selected ones, then fired delete_transcript against the
wrong IDs.
Clear-all called saveHistory(), which was a no-op stub left over from the
same incomplete-refactor pattern that produced the manualTags persistence bug
fixed in 7eb52d9. The in-memory array was emptied, but SQLite still held
every transcript, so they reappeared on next loadHistory().
- Add deleteFromHistoryById(id) next to the index-keyed deleteFromHistory.
- bulkDelete now calls deleteFromHistoryById.
- clearAll now awaits an explicit per-id delete loop and surfaces a toast on
partial failure.
- Remove the saveHistory() stub and its sole caller's import.
https://claude.ai/code/session_0189xUb6ie6t9qHkzatGZ9Rb
Sparkline: friendlier aria-label ("3 completed today. 14 total over
the last 7 days." rather than a numeric list), per-bar <title>
tooltips with absolute date + count, 30 ms stagger entrance via
scaleY animation. Badge: 180 ms opacity + translateY entrance on
mount; conditional render means each new badge re-fires the
animation. Both animations respect prefers-reduced-motion.
The earlier draft tabindex=0 on the SVG was correctly flagged by
svelte-check as noninteractive_tabindex; SVG role="img" + aria-label
is sufficient for SR navigation without putting it in the keyboard
tab order.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Phase 8 carryover backlog: the showMomentumSparkline toggle was
sitting under Rituals and visually claimed by the Launch-at-login
border-t subgroup. New top-level Tasks section hosts it, ready to
absorb future task-page settings (energy default, WIP limit, etc.).
The deeper Phase 9 SettingsPage restructure (search box + 7-group
progressive disclosure via the new SettingsGroup component) is
deferred to a follow-up polish session: the existing 2309-line file
uses a hand-rolled accordion that needs careful unwinding, and is
not in this session's scope. SettingsGroup component remains
available for that future pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Reusable progressive-disclosure wrapper around the native <details>
element. Animated chevron, hover + focus-visible affordances, and
prefers-reduced-motion guard. Designed to host the six collapsed
groups in the Phase 9 SettingsPage restructure.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Per-row Tag button calls extract_content_tags_cmd, persists via
saveTranscriptMeta. Dashed-italic chips render the AI tags distinct
from manual; clicking a chip promotes it into manualTags (the
LLM tag disappears, the manual one stays). Top toolbar gains "Tag all
untagged" for batch tagging across the corpus, with progress text.
Existing addManualTag / removeManualTag handlers swap their no-op
saveHistory() calls for saveTranscriptMeta — picks up the latent
manualTags persistence bug as a side effect.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
TranscriptEntry gains llmTags: string[]; TranscriptRow gains the
storage-shape llmTags: string. ContentTags type added. mapTranscriptRow
hydrates llmTags from the comma-joined column. saveTranscriptMeta
forwards llmTags through to update_transcript_meta_cmd, mirroring the
existing manualTags handling.
buildFrontmatter unions auto + manual + llm tags so exported markdown
surfaces every source in one tags: list.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Slim leading checkbox on every row, tinted-row state when selected.
Bulk-action toolbar appears only when selection is non-empty: select
all (visible), clear, export selected (via exportTranscriptsToDir),
delete selected (single confirm). Esc clears selection. Cmd/Ctrl+A
selects all visible when focus is inside the list and not in a text
input. Stop-propagation on the checkbox keeps the click off the
row-expand toggle.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Replaces the clipboard-only path with saveTranscriptAsMarkdown. User
picks the location via the OS save dialog; cancellation leaves no side
effect, no toast, no fallback. Toast on success names the file.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Centralises the save-dialog plus write-file plumbing. suggestedFilename
slugs the title into "<slug>-<YYYY-MM-DD>.md". saveTranscriptAsMarkdown
opens the system save dialog and writes via write_text_file_cmd; on
cancel returns null with no toast or fallback. exportTranscriptsToDir
writes one .md per item to a chosen folder, in-batch collision suffix
" (2)" etc. Documents the deliberate non-check of pre-existing files
in the chosen directory.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Default on. Controls only the sparkline; the "N today" badge is
unconditional. Copy kept in the zero-loss register: "Never counts
against you." Co-located with the Rituals section for Phase 8.
Can move to a dedicated section in Phase 9 polish.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Phase 8 completionStats store listens on these events to refresh the
daily count. Keeps the badge + sparkline accurate after un-tick / delete
paths, which don't currently fire kon:task-completed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Badge renders when today's count > 0. Sparkline renders when the
setting is enabled and any of the last 7 days has a completion.
Wrapped in a narrow aria-live region so increments announce without
re-reading the rest of the header.
Fix: converted todayCount from $derived module export to a getter
function (Svelte 5 derived_invalid_export constraint).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Tiny inline SVG. Seven bars, zero-days render as 1 px baseline stubs.
fill=currentColor so the parent's text colour (tertiary ink on the
Tasks header) drives it. Self-hides if all 7 days are zero.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Owns the last-7-days completion series. Refreshes on task-completed /
step-completed / task-uncompleted / task-deleted window events and on
window focus (for day rollover). Derives todayCount from the newest
entry.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds the Phase 8 frontend types. New setting defaults to true (sparkline
on for everyone on upgrade) and persists via the existing settings
envelope; no migration needed because missing keys spread over the
defaults object on load.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Small if-then automation layer. Rules persist in SQLite; the runner
lives on the frontend and binds to the Phase 6 event bus so the rule
pipeline reuses the same delivery primitives (timer events, TTS,
Tasks navigation).
Storage:
- Migration v12 adds implementation_rules (id, enabled, trigger_kind,
trigger_value, actions_json, last_fired_key, created_at,
updated_at) with enabled+trigger_kind index for the runner's hot
path.
- CRUD helpers: insert / list / get / set-enabled / mark-fired /
delete, plus a round-trip test.
Commands (all main-window-guarded via ensure_main_window):
- list_implementation_rules
- create_implementation_rule — validates HH:MM, checks the target
task exists at save time for surface-task actions, caps
speak-line at 240 chars, pins v1 timers to 5 minutes.
- set_implementation_rule_enabled
- mark_implementation_rule_fired — main-thread idempotency shim so
the runner can atomically claim a fire.
- delete_implementation_rule
Runner (implementationIntentions.svelte.ts):
- Subscribes to kon:task-completed and kon:morning-triage-finished
(MorningTriageModal now emits on all three exit paths — empty,
skipped, picked — so skip counts as finishing).
- 30 s poll for time-of-day rules, plus an immediate check on startup
so a rule whose time has already passed today catches up once.
- Idempotency via last_fired_key composed as YYYY-MM-DD@HH:MM for
time rules; new time rules whose HH:MM has already passed today
are pre-seeded so they don't fire retroactively on save.
- Rules are paused when Nudges "Mute for now" is on — a hard mute
stops all rule delivery in addition to OS notifications.
- Stale-task safety: if a surface-task action's target has been
deleted, the runner opens Tasks and warns clearly rather than
pretending to surface something that's gone.
Editor (ImplementationRulesEditor.svelte):
- Lives in Settings under a new "If-then rules" accordion section.
- `If` picker: time of day (with time input), a task completes,
morning triage finishes.
- `Then` composer: optional surface (inbox / today / all tasks /
specific task), optional 5-min timer, optional speak-aloud line.
- Saved rules list with enable toggle + delete.
Rules table integration for Phase 10b rename sweep: add
implementation_rules to the kon.db → corbie.db migration shim when
that phase lands.
Gates: fmt, clippy -D warnings, cargo test 265/0, svelte-check
0/0, npm build green. Pre-existing Vite chunk warning on sounds.ts
is unrelated to Phase 7.
Frontend-owned nudge bus that consumes in-app signals Corbie already
produces, applies suppression, and fans out to OS notification + an
optional TTS read-aloud. OS-wide keyboard/window activity detection
stays deferred per the revised roadmap — the plan before rewrite
would have been brittle on Wayland, permission-heavy on macOS, and
low-quality everywhere.
Triggers (v1, all in-app signals):
- inactivity_with_active_timer — timer running, window blurred ≥ 90 s,
at least 60 s into the timer.
- pending_morning_triage — ritual enabled, past 10:00 local, last
shown ≠ today. Polls every 5 min while focused.
- micro_step_idle — micro-step decomposition created, no child step
or parent task completed within 15 min.
Suppression:
- Respects nudgesEnabled + nudgesMuted.
- No nudge while the app has focus (document.hasFocus).
- Hard cap 3 per rolling hour.
- Permission requested via @tauri-apps/plugin-notification on first
delivery; denial is silently respected.
Rust side:
- tauri-plugin-notification registered + ACL entries on the main-
window capability only (secondary windows can't fire nudges).
- commands::nudges::deliver_nudge — thin wrapper, security-guarded
via ensure_main_window, delegates to the plugin. No DB writes —
the roadmap's nudges-audit table is deferred until a concrete need
emerges.
Frontend glue:
- nudgeBus.svelte.ts — subscribes to window events, applies
suppression, calls deliver_nudge (+ tts_speak when speakAloud is
on).
- kon:task-completed now dispatched on complete_task_cmd success.
- kon:microstep-generated + kon:step-completed dispatched from
MicroSteps so the idle trigger can clear itself on any engagement.
- kon:focus-timer-cancelled added to focusTimer so the bus can reset
its inactivity state on cancel, not only on natural completion.
- nudgeBus started from +layout.svelte onMount, stopped on destroy.
Settings:
- New "Nudges" section with three toggles: Enable nudges, Mute for
now (separate so a hard mute doesn't lose preferences),
Speak nudges aloud (reuses Phase 4 TTS).
- All default OFF. No first-run prompt — nudges are a Settings-found
feature rather than a walkthrough step.
Out of scope per the revised Phase 6 spec: OS-wide keyboard/window
hooks, biometric signals, custom trigger editor (Phase 7), notification
sound (platform variance too high for Layer-1 — revisit in Phase 9
polish with a bundled .wav).
Gates: fmt clean, clippy -D warnings clean, cargo test 262/0 (4
existing + 262 current), npm run check 0/0, npm run build green.
Three opt-in rituals, all default OFF. Research-anchored (Barkley's
point-of-performance, Sweller cognitive-load theory, Newport shutdown
ritual, Gollwitzer implementation intentions, Thaler/Sunstein nudge
with informed consent for the ADHD audience).
Morning triage: modal gated on ritualsMorning toggle, configurable
trigger time (default 08:00 to respect ADHD sleep inertia rather than
the spec's 06:00), "pick up to three for today" with a gentle swap
message on the fourth attempt. Skip sets last-shown-today so it never
re-prompts the same calendar day. last-shown persists via kon_storage.
Evening wind-down: dedicated page, user-triggered only (tray menu +
Settings button). Mechanical closure + physical reset + intentional
cue — the whole Newport template. Open loops are read-only reflection;
Tasks page owns transactions. Copy is additive throughout: "you
finished X today", never "you didn't finish Y".
Autostart: tauri-plugin-autostart registered (LaunchAgent on macOS,
.desktop on Linux, registry Run on Windows). No bespoke Rust commands
— frontend calls the plugin's invoke-handlers directly. Toggle in
Settings is one-way (click → OS call → state update) to avoid the UI
lying during the round-trip. First-run presents a forced-choice prompt
for all three options, with "skip all" escape hatches per step.
Copy audit against RSD literature: no "overdue", "failed", or
day-to-day comparison framing anywhere in ritual surfaces.
Post-v0.1 ideas captured in the roadmap: calendar integration
(read-only ICS as interim, cloud sync parked) and right-click-to-task
(in-app simple, system-wide a separate phase).
Platform-dispatched TTS (spd-say + espeak-ng fallback on Linux, say on
macOS, PowerShell System.Speech on Windows) with a shared SpeakerButton
component. Tap to speak, tap again to stop; only one button speaks at
a time so two surfaces don't talk over each other. Text always travels
via argv (or a PowerShell here-string delivered through -EncodedCommand
on Windows) so user content never enters a shell string.
Mount points: DictationPage transcript footer, transcript viewer header,
per-step in MicroSteps. Settings gains a "Read aloud" accordion with
voice picker (lazy-loaded from the OS synth), rate slider 0.5-2.0x,
and a British-English test utterance.
Rust tests cover rate mapping, NaN handling, and Windows here-string
terminator safety. No pause/resume, no SSML, no cloud voices — that
stays out of scope per the Layer-1 roadmap.
Codex post-implementation review flagged one MAJOR: the energy segmented
control declared `role="radiogroup"` / `role="radio"` but only wired
`onclick`. No arrow-key navigation, no Home/End, no roving tabindex.
Keyboard users got four independent tab stops while assistive tech was
told it was a single radio group — a broken ARIA contract.
Fix (W3C APG Radio Group pattern):
- Extract the options list as `energyOptions` so the render loop and
the keyboard handler share one source of truth.
- `energyRadioKeydown` handles ArrowLeft/Right/Up/Down (cycle wraps),
Home (first), End (last).
- Roving tabindex: the currently-checked button gets `tabindex=0`,
the rest get `tabindex=-1`, matching the APG recipe. Focus moves
with selection.
- The radiogroup container gets `tabindex="-1"` to satisfy the
svelte-check a11y rule without creating its own tab stop.
All green: 251 tests, clippy -D warnings, fmt, svelte-check 0/0, build.
Closes Phase 3 of the 2026-04-23 feature-complete roadmap. Incorporates
the Codex plan-review fixes from this session: profile-free index, tri-
state update command, and de-prioritise-not-hide semantics.
Storage (kon-storage):
- Migration v11 adds `energy TEXT` to `tasks` with a CHECK constraint on
`high | medium | brain_dead | NULL`. Index `(energy, created_at DESC)`
— deliberately not per-profile because the tasks table carries no
profile_id column yet (tracked as a separate gap in HANDOVER).
- `TaskRow.energy: Option<String>` plus `task_row_from` read.
- `insert_task` signature grows by one optional arg (`energy`). Allowed
`too_many_arguments` with a rationale comment — the positional shape
matches the column order and flipping to a params struct would have
rippled through every caller for cosmetic benefit only.
- New `set_task_energy(pool, id, Option<&str>) -> TaskRow`. Lives as its
own function because `update_task` uses COALESCE to let `None` mean
"preserve" — which would make clearing the tag impossible.
- Two new tests: round-trip including explicit NULL clear, and CHECK
constraint rejection of unknown values.
- Tests updated for the v10 → v11 version bump.
Tauri (src-tauri):
- `TaskDto.energy`. `CreateTaskRequest.energy` (optional). Inline
validation against the allowed set before hitting the DB, so frontend
bugs surface as friendly errors instead of CHECK-constraint failures.
- New `set_task_energy_cmd` command mirroring the storage tri-state API.
Frontend (svelte):
- `EnergyLevel` type added to `types/app.ts`. `TaskDto`, `TaskEntry`, and
`TaskDraft` grow an `energy` field.
- `SettingsState.currentEnergy` (persisted) + `matchMyEnergy` (persisted
toggle). Defaults: null + false — no surface change until user opts in.
- `setTaskEnergy(id, EnergyLevel | null)` action on the task store.
Calls the dedicated Tauri command, updates local state, broadcasts to
sibling windows.
- `EnergyChip.svelte` — new component. Cycles unset → High → Medium →
Brain-Dead → unset on click. Colour tokens: accent / warning /
text-tertiary (deliberately not danger-red for Brain-Dead — the brief
is explicit that this state must not feel pathologised).
- Chip rendered on every task row in TasksPage and every row in
WipTaskList. Hidden-until-hover when energy is unset so untagged rows
stay calm; always visible once tagged because the colour is the signal.
- Tasks page header gains a "I feel" segmented control and a
"Match my energy" toggle. When both are active, matching tasks sort
to the top — unset tasks are treated as Medium-equivalent. Nothing is
ever hidden; this is a de-prioritisation, not a filter.
Deferred / out of scope:
- LLM-driven surfacing (brief says "The AI surfaces...") — deterministic
client-side sort is v1; LLM layer is a later phase.
- tasks.profile_id + per-profile energy sort — separate migration.
All green: cargo build + 251 tests + clippy -D warnings (0 warnings)
+ fmt + svelte-check (0/0) + npm run build.
Independent review surfaced three majors and one nit. All actioned.
MAJOR 1 — profile scoping:
`decompose_and_store` and `extract_tasks_from_transcript_cmd` now
accept an optional `profile_id` (wired from `profilesStore.activeProfileId`
in MicroSteps.svelte and DictationPage.svelte), and thread it into the
feedback-retrieval query so per-profile decomposition styles do not
leak into each other. `record_feedback` gets the same treatment.
MAJOR 2 — prompt-budget regression on long inputs:
New `trim_to_budget` helper + `FEW_SHOT_CHAR_BUDGET = 2000` char cap
in `src-tauri/src/commands/tasks.rs`. Retrieval still pulls up to 5
rows but they are char-counted and truncated against the budget
before being sent to the LLM. Char cost matches the `Input: ...\n
Good output: ...` render path so the budget maps cleanly to ~570
Qwen3 tokens, well inside the 8192-context reserve after the 512-
or 768-token response allocation. Oldest-first drop order (iteration
stops at cost exceeded) preserves the most recent correction which
is the one carrying the user's live preference.
MAJOR 3 — inline edit stale-rollback race:
`saveEdit` in MicroSteps.svelte now stamps a monotonic per-step
`saveToken`. Each edit bumps the token; on failure the rollback
only fires if `saveToken[step.id] === myToken`, so a slow-failing
first save can no longer overwrite a faster successful second save.
NIT — retrieval ordering stability:
`list_feedback_examples` ORDER BY now `created_at DESC, id DESC`.
SQLite timestamp precision is one second; without the secondary
key, bursty feedback within the same second would select
non-deterministically.
Also: malformed `context_json` now warns via eprintln! rather than
disappearing silently — Codex minor.
All green: cargo build + 249 tests + clippy -D warnings + fmt
+ svelte-check (0/0) + npm run build.
Closes the human-in-the-loop gap from docs/brief/feature-set.md and
Phase 2 of the 2026-04-23 feature-complete roadmap.
Storage (kon-storage):
- Migration v10 adds the `feedback` table: (target_type, target_id,
rating, original_text, corrected_text, context_json, profile_id,
created_at) with CHECK constraints on target_type and rating, plus
indexes on (target_type, rating, created_at DESC) for prompt-time
retrieval and (profile_id, target_type, created_at DESC) for
per-profile scoping.
- New public API: `FeedbackTargetType`, `RecordFeedbackParams`,
`FeedbackRow`, `record_feedback`, `list_feedback_examples`.
- Tests updated — the RB-02 rollback regression now discovers the
real max version at runtime instead of hard-coding v10 for its
poison migration.
LLM (kon-llm):
- `prompts::FeedbackExample` — local shape for few-shot exemplars so
kon-llm stays independent of kon-storage.
- `prompts::build_conditioned_system_prompt` — appends a "here is
the style this user prefers" block to the base system prompt
when examples are available; returns the base prompt unchanged
when empty, so new users and early sessions see generic output.
- `LlmEngine::decompose_task_with_feedback` and
`LlmEngine::extract_tasks_with_feedback` thread examples through
to the builder. The old one-arg variants are preserved and now
call through with an empty slice.
- 4 unit tests covering empty, empty-input-skip, correction-wins,
and thumbs-up-only fallback.
Tauri (src-tauri):
- New commands::feedback module: `record_feedback`,
`list_feedback_examples_cmd`.
- `decompose_and_store` and `extract_tasks_from_transcript_cmd`
now fetch the last 5 positive/neutral feedback rows for their
target type and pass them through to the LLM, wiring the
learning loop end-to-end.
- Shared `to_llm_examples` helper parses the `context_json.input`
field (where the recorder stashes the parent task text / transcript
chunk) back into the exemplar shape.
Frontend (MicroSteps.svelte):
- Thumbs-up and thumbs-down buttons on every micro-step row.
Hover-revealed; the vote recolours the icon; clicking again
clears the local highlight (the row itself stays in the audit
trail).
- Pencil icon + double-click to edit step text. Save flows through
update_task_cmd for persistence and records a correction feedback
row with (original_text, corrected_text) — the highest-value
training signal.
- Parent task text is captured in context_json.input at record time
so the prompt builder can reconstruct the (input, preferred-output)
pair on subsequent decompositions.
- Feedback capture is best-effort — a record_feedback failure never
interrupts the primary action.
What's deferred to a later phase:
- Thumbs + corrections on extracted tasks (same pipeline, different
surface — probably TasksPage after the AI-extraction path)
- Thumbs on transcript cleanup output
- Semantic retrieval over the feedback corpus (once there is enough
data to justify embedding infrastructure; the storage shape is
already ready for it)
Jake's feedback on Phase 1: make the timer pinnable / always-on-top,
combined with the existing Now-list pop-out. Two changes:
1. Mount <FocusTimer /> in src/routes/float/+layout@.svelte so the
running countdown stays visible in the always-on-top float window
alongside the WIP task list. No content change to the float page
itself — the timer is a global overlay.
2. Add a pop-out icon to the main-window focus timer that opens the
existing /float route via window.open. One click → timer + Now
list pinned on top without touching main window focus. Hidden
inside the float window itself (detected via URL) so you cannot
recursively pop out.
Result matches the Todo float-out UX the user already knows:
click ExternalLink, you get a small always-on-top window with
tasks + a live countdown ring in the top-right.
Closes the Core MVP gap in docs/brief/feature-set.md ("visual time
representation") and wires the dangling kon:start-timer emit that
MicroSteps.svelte has been firing into the void since the stub was
written. Implements phase 1 of the 2026-04-23 feature-complete
roadmap.
New:
- src/lib/stores/focusTimer.svelte.ts — singleton timer store with
localStorage persistence so a timer started in Dictation survives
page nav, window close, and reopen. Gentle WebAudio chime at
completion (no bundled asset). 250 ms tick. Completion flash for
3 s before auto-clear.
- src/lib/components/FocusTimer.svelte — floating top-right overlay
with SVG progress ring (shrinking colour: accent -> warning in
the final 15% -> success on completion). Cancel + "+1 min" on
hover. Renders nothing when idle.
Wired in +layout.svelte next to ToastViewport.
Two triggers now in the app:
- MicroSteps row 2-min button (pre-existing emit, previously no
listener)
- WipTaskList row 5-min button (new; the brief's "just-start"
from the Now column)
Respects prefers-reduced-motion via the existing
[data-reduce-motion] attribute.
Out of scope, carried to later phases: rhythmic voice anchoring
(Phase 6), custom-duration picker, multi-timer UI, native OS
notification (deferred to Phase 6 with the full nudge pipeline).
resolveVolume previously clamped any value >=1 to full blast. If a
future settings change ever leaks a 0–100 scale through (instead of
0–1), the user gets jump-scared at max volume. Treat any v>1 as a
scale-drift bug and play at DEFAULT_VOLUME instead.
Also reject NaN / Infinity explicitly rather than relying on the
<=0 branch catching them.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two new Settings → AI knobs that compose cleanly with what already
shipped (aiTier, LLM model, translator prompt framing).
**B.1 #15 — Named cleanup presets.** LlmPromptPreset enum
(Default / Email / Notes / Code) appends a short context hint onto
the CLEANUP_PROMPT just before generation. Presets shape tone and
structure ("email paragraph", "bulleted meeting notes", "preserve
technical terms") without licensing the content-editing the
translator-not-editor framing forbids. cleanup_transcript_text_cmd
now takes `preset: Option<String>` which runs through the new
LlmPromptPreset::parse (normalises aliases like "meeting-notes",
collapses unknown values to Default).
**A.1 #28 — Sequential-GPU guard.** New LocalEngine::unload drops
the backend + model_id so a subsequent load actually reclaims VRAM.
load_llm_model, load_model, and load_parakeet_model Tauri commands
grow an optional `concurrent: bool` argument. When concurrent is
Some(false), loading LLM first unloads whisper+parakeet, and vice
versa — prevents VRAM OOM on tight-VRAM setups. Default is the
previous parallel behaviour so nothing changes for multi-GB cards.
Transcribe-in-progress paths (transcribe_pcm, transcribe_file, live)
pass None, so mid-dictation model loads don't accidentally tear
down the LLM.
Settings UI (AI section):
- Cleanup preset segmented button + descriptive copy for each option.
- GPU concurrency segmented button with explicit trade-off text
("faster transitions vs fits in tight VRAM").
Frontend wiring:
- settings.llmPromptPreset flows from DictationPage's
cleanupTranscriptIfEnabled into the Tauri command.
- settings.aiGpuConcurrency flows from both DictationPage (auto-load
on record) and SettingsPage (manual load/unload buttons) as
`concurrent: "parallel" === true` to the load commands.
Tests: three new preset cases in crates/ai-formatting/src/llm_client.rs
(parse aliases, suffix non-empty for non-default, default suffix
empty). All 139 existing lib tests still pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The brief's pain point is opaque load failures: llama-cpp-2's errors
bubble up as raw C++ strings ("cudaMalloc failed: out of memory",
"invalid gguf magic"). A user seeing that has no path to recovery.
New backend command test_llm_model runs a staged diagnostic:
1. Model not downloaded → `not-downloaded` + download hint.
2. File size ≤90% of expected → `incomplete` (stalled download)
+ re-download hint. Matters because llama-cpp-2 can segfault
on truncated GGUF rather than returning cleanly.
3. Requested model already loaded → `ready`, no side effects.
4. Otherwise attempt a real load. On failure, classify_llm_load_error
maps the raw string to one of:
- load-failed-vram (OOM / cudaMalloc / allocation)
- load-failed-corrupt (GGUF magic / unsupported format)
- load-failed-permission (permission denied / access denied)
- load-failed-other (catch-all)
Each category has a prewritten actionable hint pointing at the
specific Settings surface (tier picker, re-download, file perms).
classify_llm_load_error is pure-string and unit-tested — 8 cases
covering the main categories plus edge cases (OOM alias, Windows
"Access is denied", unknown errors). Ordered narrow-to-broad so
overlap doesn't misclassify.
Settings UI gets a "Test" button in the AI section's action row,
visible whenever the model is downloaded (both downloaded-idle and
loaded states). Shows inline hint below the status line when the
test surfaces one. Refreshes both local and global LLM status after
the test since a successful test implicitly loads the model.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Hands-off feedback for recording lifecycle: a short C5 pitch at
record start, a falling G4 at record stop, and a staggered C5–E5–G5
major third when the finalise flow completes. Synthesised at runtime
via the Web Audio API (OscillatorNode + linear-ramped GainNode
envelope) rather than shipping WAV assets — keeps the binary size
flat and lets us tweak timbre without touching bundled files.
Off by default. Settings → Output exposes the toggle with a volume
slider (0–100%, default 15%) and a "Test" button that plays the
completion cue so the user can confirm loudness without recording.
Hooked at three call sites in DictationPage:
- playStartCue after page.recording = true in startRecording
- playStopCue at the top of stopRecording
- playCompleteCue just before `saved = true` at the end of
finaliseTranscription's transcript-present branch
All three no-op when settings.soundCues is false. The Web Audio
context is lazily constructed on first cue (most browsers suspend
it until a user gesture — Tauri's webview inherits that). If the
AudioContext can't be built we silently drop the cue rather than
throwing.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The LLM runtime has been quiet since it shipped in Phase 3 — users
had no surface-level signal that cleanup was loaded, warming, or
actively generating. Settings has verbose status text internally,
but a dictation-flow user never opens Settings during a run.
New: a shared $state store drives a small chip in the sidebar that
reflects the true LLM state in ≤500 ms of any transition (brief
item #31 acceptance). Five states:
off → hidden (user has aiTier === "off")
warming → model download or first load in flight; amber pulse
ready → loaded + idle; green dot
generating → cleanup_transcript_text_cmd or
extract_tasks_from_transcript_cmd in flight; accent
pulse with Sparkles icon
error → last operation failed; red dot with AlertTriangle
The store exposes three calls: refreshLlmStatus(aiTier) (polls the
backend), markGenerating(detail) / markGenerationDone(success).
DictationPage wraps its cleanup + extract calls in mark-generating
pairs. SettingsPage's LLM load / unload / delete / download paths
also refresh the global store so Settings-initiated transitions
surface in the sidebar immediately. The chip collapses to a
dot-only compact form when the sidebar is collapsed.
No backend changes — everything wires onto the existing
`get_llm_status` Tauri command.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Kon's ideology rule: raw Whisper output is the source of truth; LLM
cleanup is additive, never destructive. The preview overlay already
tracks both rawText and finalText across the listening → live →
cleanup → final phases — but until now the user had no one-click path
from final to raw if cleanup changed their meaning.
Frontend: a context-aware "Use raw" / "Copy raw" button appears in
the preview overlay's final phase, only when rawText and finalText
actually differ (Raw format mode or LLM-off leaves the button hidden).
Two behaviours depending on how the transcript reached the target:
- settings.autoPaste = true → invoke paste_text_replacing, which
sends the platform's undo keystroke to the focused app,
waits UNDO_PASTE_GAP_MS (60 ms) for the compositor / app to
process it, then pastes the raw transcript. The preview hides
itself beforehand so the keystroke doesn't race focus
(existing Wayland-hardening path).
- settings.autoPaste = false → nothing was pasted in the first
place, so just overwrite the clipboard with raw. User's own
paste yields raw.
Backend: new paste_text_replacing Tauri command plus a mirror of the
paste-backend matrix for undo (wtype -M ctrl z / xdotool key ctrl+z /
ydotool keycodes 29:1 44:1 44:0 29:0 / osascript cmd+z / SendKeys '^z').
Reuses the pick_linux_backend_order Wayland-vs-X11 preference.
Registered in the Tauri command handler.
Acceptance per the brief: "after paste, Ctrl+Z within 5 s replaces
LLM output with raw transcript" — satisfied via the 4 s auto-hide
window on the preview's final phase. The click extends auto-hide so
the user actually sees the confirmation.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds colors_and_type.css token system, fonts (Lexend, Instrument Serif Italic,
JetBrains Mono, Atkinson Hyperlegible Next, OpenDyslexic), SVG assets (wordmark,
waveform mark, grain), HTML preview spec cards, UI kit, and SKILL.md reference
under src/design-system/. Foundation for applying the new Kon visual identity.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Previously the bulk import ran new Set(...) on raw trimmed strings
before lowercasing, so 'ACME' and 'acme' both survived the first
dedupe pass. Neither existed in the store, so both got added —
defeating the commit message's claim that pasting the same block
twice with different casing is a no-op.
Collapse case variants at the initial dedupe step using a lowercase
seen-set, keeping the first occurrence's casing as written.
Co-authored-by: jars <jakejars@users.noreply.github.com>
Adds src/lib/utils/hotkeyValidity.ts with validateHotkey(combo, os)
and wires it into HotkeyRecorder.svelte. Rules:
- X11/Wayland/Linux: reject single-key combos unless the trigger
is F13..F24 (the conventional global-shortcut escape hatch).
- Windows: reject combos whose only modifier is a right-hand
variant (RCtrl/RAlt) — matches Handy #966's failure mode where
RegisterHotKey silently ignores them.
- macOS: reject Fn-only combos and bare-key combos (non-function
keys).
- unknown OS: pass through — better to ship a flawed combo than
reject one we can't validate.
When validation fails, the recorder leaves the previous known-good
hotkey intact and surfaces an inline warning-coloured sentence
explaining why, with actionable copy ("add a modifier", "use the
left-hand equivalent", etc.). The save button disappears because
settings.globalHotkey is never written to — no state drift.
Matches Handy #917 / #1019 / #966 / #956, brief item #5.
Co-authored-by: jars <jakejars@users.noreply.github.com>
A rapid double-tap of the global hotkey, evdev autorepeat, or a
sticky-key compositor quirk (KDE's 'slow keys') can all deliver
the same press twice within ~100ms. Without a guard, the recording
toggles into and out of the same frame and the capture is lost.
Gates the evdev 'kon:hotkey-pressed' forwarder in +layout.svelte
behind a 120ms debounce (Date.now()-based; no timers, so no tail
latency for a legitimate single press). The debounce is intentionally
shorter than a deliberate double-press cadence but longer than any
autorepeat interval we've seen in the wild.
The audio-stream-warming half of brief item #4 (Handy #1143) lives
in Workstream A's Phase A.3 warm-up WAV; this covers the UX side.
Co-authored-by: jars <jakejars@users.noreply.github.com>
Adds src/lib/utils/settingsMigrations.ts exposing
loadSettingsWithMigration() / saveSettingsWithVersion() around a
{version, data} envelope in localStorage["kon_settings"]. The
migration chain is indexed by destination version so adding a v2
is one MIGRATIONS[2] = (prev) => next entry away from working.
Legacy bare-object settings blobs are treated as v0 and folded into
v1 identically to before — no user-facing reset — but an unreadable
blob now surfaces a single Settings reset toast instead of silently
dropping data.
Covers Handy #602 ('settings reset on update') and unblocks the
other B.2/B.3 SettingsState additions listed in
docs/whisper-ecosystem/workstream-B.md: every subsequent field
lands behind a MIGRATIONS step, so older Kon builds stay readable.
Co-authored-by: jars <jakejars@users.noreply.github.com>
Settings → Vocabulary gets a "Bulk add from a list…" disclosure under the
single-term row. Expanding reveals a textarea; paste newline- or
comma-separated terms, hit Import, and the page loops addTerm for each
entry the active profile doesn't already have.
Dedupes case-insensitively against the existing term list so pasting the
same block twice is a no-op. Skipped + failed counts surface via toast;
persistent errors (any failing term) also land in vocabularyError so the
inline panel explains what went wrong.
Covers OpenWhispr issue #460 — one-at-a-time entry becomes friction past
roughly ten terms. No backend changes; addTerm is already in profilesStore.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Ported the best bits of OpenWhispr's TranscriptionPreviewOverlay into Kon's
window conventions. Off by default — toggle in Settings → Output → "Floating
preview when Kon is unfocused". Opens only when the main window isn't
focused at the start of a recording, so it never adds noise when the user
can already see the transcript in the main surface.
Phase state machine (src/routes/preview/+page.svelte):
- listening — pulsing dot, no text yet
- live — animated bars + streaming raw Whisper output
- cleanup — accent bars while the LLM cleanup pass runs
- final — checkmark + formatted text + 4s auto-hide
Data plumbing: raw segment text is captured before post_process_segments in
live.rs (new raw_text field on LiveResultMessage) and in transcription.rs
(new raw_text in the transcription-result payload). DictationPage forwards
raw_text to the overlay via Tauri global events — preview-listening on
start, preview-append per chunk, preview-cleanup before the LLM pass,
preview-final with the formatted text, preview-hide when a run produced no
transcript (empty recording / cancel).
Window is always_on_top, skip_taskbar, focused=false so it never steals
focus from whatever the user is dictating into. open_preview_window shows
an existing hidden preview or builds it fresh; close_preview_window hides
without destroying so the next open is instant.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Default off. When on, the layout polls detect_meeting_processes every 15s
with the user's app-name patterns. On a fresh match (edge-triggered — no
re-toast until the app goes away and comes back) we fire a reminder toast
that tells the user which meeting app appeared and their global hotkey. We
never start recording on this signal; the ideology rule says the user
decides. The signal is a single channel: process list match only — no mic
activity heuristic, no calendar.
Backend adds kon_core::process_watch::{list_running_process_names,
match_meeting_patterns} over sysinfo, exposed to the frontend as the
detect_meeting_processes Tauri command.
Settings ships two new fields — meetingAutoCapture (bool) and
meetingAutoCaptureApps (string[]) — with a comma-separated input in the
Output section. Default app list is ["zoom", "teams"], user-editable.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
initI18n (src/lib/i18n/index.ts) registers three locales and picks the
initial one in order: kon_locale in localStorage > navigator.language short
code > en. +layout.svelte calls it once at app boot; guarded so per-window
re-init is a no-op.
Locale files are deliberately sparse — this is a scaffolding pass so strings
can be migrated incrementally. The Settings → Appearance → Language picker
plus its own description is the first real consumer; everything else
continues to render as hardcoded text until extracted.
Also: split the @chenglou/pretext ambient shim into src/lib/shims.d.ts. The
declaration previously lived in app.d.ts alongside a top-level `export {}`,
which made app.d.ts a module — scoping `declare module` to its own imports
and breaking resolution from src/lib/utils/textMeasure.ts. The fresh
.svelte-kit sync triggered by installing svelte-i18n surfaced it. Ambient
shim files must stay script-scoped (no top-level imports/exports).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds an opt-in "auto-paste into focused window" toggle. When enabled, the
dictation pipeline sets the clipboard and then sends a Ctrl+V / Cmd+V
keystroke to whatever window currently has focus — the common case after a
global-hotkey dictation, since Kon's own window never stole focus.
Backend (src-tauri/src/commands/paste.rs) probes for a platform paste tool
and falls back cleanly:
- Linux Wayland: wtype > ydotool > xdotool
- Linux X11: xdotool > ydotool > wtype
- macOS: osascript System Events keystroke
- Windows: PowerShell WScript.Shell SendKeys
detect_paste_backends is a pure probe used by Settings to describe the
available backend next to the toggle (or nudge the user to install one).
paste_text always copies first, so auto-paste failure degrades to the
existing clipboard-only behaviour and surfaces a warn toast.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Parakeet-TDT scores 85 on any GPU-equipped English-capable system (Instant
speed + Great accuracy + GPU boost + headroom) vs ~75 for the best distilled
Whisper. A new test in recommendation.rs locks this in so future scoring
tweaks don't silently regress it.
FirstRunPage previously stored settings.modelSize by title-casing a lowercased
alias — which worked for Tiny/Base/Small/Medium but produced
"Whisper-distil-small-en" for the new distil ids. Swap to an id→label map
and pass the raw model id through to download_model/load_model; the backend
already accepts full ids via the whisper_model_id fallback.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two new registry entries (crates/core/src/model_registry.rs):
- whisper-distil-small-en — 336 MB, ~6× faster than whisper-small-en
- whisper-distil-large-v3 — 1.55 GB, near large-v3 accuracy at medium size
Both are whisper.cpp-compatible GGML binaries hosted on HF by the
distil-whisper org; no runtime change, just wider model choice. English-only
by design (matches upstream Distil-Whisper).
The Settings model picker widens to six options — Tiny, Base, Small,
Distil-S, Medium, Distil-L — ordered roughly by accuracy. Download/load
commands now take the resolved model id (whisper-distil-*) instead of the
lowercased label, so the frontend owns the label↔id mapping.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
kon-llm now owns a real LlamaBackend + LlamaModel, with three Qwen3 tiers
(1.7B Q4, 4B-Instruct-2507 Q4, 14B Q5) selectable per hardware. Downloads
are resumable with SHA-256 verification and stored under ~/.kon/models/llm.
Engine exposes three high-level surfaces — all greedy/temp-0, GBNF-constrained
where output shape matters:
- cleanup_text (prompt-injection-hardened system prompt; profile terms
appended as "preserve these spellings" suffix)
- decompose_task (3–7 micro-steps, constrained JSON array)
- extract_tasks (optional-array; empty when no explicit commitments)
post_process_segments now takes an Option<&LlmEngine> and, when loaded and
format_mode != Raw, joins segments → cleanup → replaces segments with the
cleaned text (first segment span). Rule-based path still runs first; LLM
errors log and keep rule-based output.
Tauri commands: recommend_llm_tier, check_llm_model, download_llm_model,
load_llm_model, unload_llm_model, delete_llm_model, get_llm_status,
cleanup_transcript_text_cmd, extract_tasks_from_transcript_cmd,
decompose_and_store (LLM-backed subtasks).
Settings: AI tier toggle (off / cleanup / tasks), model picker with
downloaded/loaded status, download progress events via
kon:llm-download-progress.
Dictation: ensureLlmModelLoaded on mount, cleanupTranscriptIfEnabled after
stop when tier != off and format_mode != Raw, LLM task extraction when
tier=tasks (regex fallback on failure).
Interim: both llama-cpp-sys-2 and whisper-rs-sys statically link their own
ggml, so src-tauri/build.rs emits -Wl,--allow-multiple-definition on Linux.
Replace with a system-ggml shared-lib setup as a follow-up.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
