Server: ui-stream.ts is verbatim pi AgentEvent passthrough (SSE data
lines + [DONE]); only app-level events are the loop-level error and
turn-settled terminal; onError carries raw provider text (owner ruling).
Client: useChat + the AI-SDK transport replaced by components/session-core.ts
(SessionModel: 409 identity-retry, optimistic c_ appends, absorbed()
adoption, needsContinuation, shared consume for POST + reattach) and the
session.ts hook; the golden-master grouping canon moved client-side into
core/view.ts (projectView); scanners read the native record (phaseAt3Lh,
danglingQuestionOf, graph toolParts, plan-blocks, runner, extraction).
Coverage: the native message is the unit (fork 3) with the call+result
inseparability bond rule; view-line-less messages (markers, error
results) are invisible to coverage and never checkpointed.
Deletes: chat-transport, projection, ui-message, their tests (ported to
tests/session-core.test.ts), and the ai + @ai-sdk/react dependencies —
the Vercel SDK is fully out of the tree. Wire pins in chat-flow rewritten
onto the AgentEvent dialect (eventsOf parser).
Gates: tsc clean, 994/994 tests, eslint clean, build green; live smoke
on PORT=3004 (verbatim wire → native record).
The onSubmitAnswer path was the one Wave B send site missing its stash:
addToolOutput triggers the SDK's automatic continuation send, the
transport found nothing stashed, and threw — the answer never POSTed, so
the question stayed open. One line (plus the contract comment). Caught
live by the owner; the suite drives answer events directly at the
handler, so this client seam stays untested until Wave D's client work.
The chat record is now pi-native agent history (LhMessage[]), the AI SDK's
UIMessage transcript contract is gone from the server, and every client turn
arrives as ONE explicit input event.
Core (new):
- src/core/native-messages.ts — the LhMessage taxonomy (user/assistant/
toolResult + lhMarker customs: verdict, probe-handoff, subagent-steer,
subagent-abort, plan-block), constructors, and toLlmMessages — the single
model-facing seam (convertToLlm): strips ids, flattens markers to user
text, drops UI-only plan-blocks, synthesizes INTERRUPTED results for
dangling aborted calls, never throws.
- src/core/projection.ts — projectNative: the one-way native → UIMessage
projection (consecutive assistant+toolResult runs group into one
assistant view message; toolResults attach backward by toolCallId;
markers project to data parts; plan-blocks splice after the last
successful plan_init). Every existing consumer reads through it.
- src/core/input-events.ts — the ChatInput union (text/answer/verdict/
continue), parseChatInput validation, and inputAppends (answer targets
the dangling question call or 400s; verdict = optional note user
message + native verdict marker).
Server:
- chat-handler.ts — the turn root takes {chatId, input}; write-ahead
persists input appends + claimed steers before the turn; claimSteers
coalesces the whole queue with the handler's steerDelivered filter as
the dedupe seam; teacher-vs-prober branch on hasProbeDone.
- turn-service.ts — persistTurn stamps ids, drops clean question
placeholder toolResults (the record ends at the unanswered call),
emits plan-block markers idempotently (plan-blocks.ts rewrite), and
persists abort partials.
- ui-stream.ts — PiTurnSpec takes the native base; the wire dialect is
unchanged. jobs.ts steers are native markers (steer_<jobId> dedup ids);
waitForIdle reads danglingQuestionOf. teacher-agent.ts wires
convertToLlm: toLlmMessages. record-adapter.ts DELETED wholesale.
- GET /api/chats/[id] serves messages: projectNative(record.messages).
Client:
- chat-transport.ts — the 409 retry is IDENTITY (re-POST the same event
body after the turn settles — the event was never persisted
pre-acceptance; the merge logic is gone). The transport consumes the
stashed input event via prepareSendMessagesRequest and throws loudly on
a missing stash.
- chat.tsx — every send call site stashes exactly one ChatInput event
(text, answer, verdict, continue) before invoking the SDK send.
Tests: full native migration (chat-flow rewritten to input events and
native record pins; jobs/extraction/demos/graph/usage/plan-blocks/
probe-done/reattach-transport converted; record-adapter.test.ts deleted;
chat-transport tests pin the identity retry). 982/982, typecheck, lint,
build all green.
Reattach to a turn that outlived its view through the SDK's native
resumeStream() so replay burst + live deltas flow through the exact
same processUIMessageStream chunk pipeline as first-subscriber
streaming — no second rendering dialect.
chat-transport.ts: ReattachingChatTransport subclasses
DefaultChatTransport and repoints reconnectToStream at GET
/api/chats/[id]/stream (the server's reattach endpoint, not the SDK
default /api/chat/{id}/stream). It peeks the first SSE event:
- turn-settled terminal: cancel body, fire onSettled (refetch), return
null — the SDK treats null as no active stream, status stays ready.
- start chunk whose messageId equals the seeded last assistant message
id (continuation case): rewritten to a fresh id so the resumed run
mounts beside the seeded message instead of replacing its parts —
settle-time record sync merges to the server's assembled message.
chat.tsx: mount effect resumes when initiallyInFlight (no once-ref:
React StrictMode's dev double-mount aborts the first attempt at
dispose — the retry must run). Mid-session discovery sites (jobs
poll, handoff watch) funnel into attachToRunningTurn().
turnInFlight is demoted to fallback liveness: the 1500ms record poll
self-heals a failed attach and clears a stale error banner on settle.
A status watcher clears the flag once the local stream represents
the turn. Stop/Esc unified in abortInFlightTurn: local stop() when a
stream runs here, always POST /api/activity/abort {kind:'turn'} with
409 not_running treated as a benign raced settle. Composer busy now
covers reattached turns (Stop square shows, send gated).
Live-verified on :3001 with real model turns: navigate away mid-turn
and back (replay + live deltas in-view, fallback poll disengaged),
Esc-abort of a reattached turn (partial persisted server-side,
composer restored), two concurrent sessions (raced-settle handled
via turn-settled + record refetch; live settle adopts the final
record), and no leaked stream readers after unmount.
Owner report (2026-09-25): in the teaching phase a new question arrives
collapsed behind the reveal button, but the transcript scrolled up as if
it had OPENED — the room it made was the expanded form's height, never
shown. Root cause: the expanded state is stale-true from the previous
question (nothing resets it when the dock returns to the composer), so
the new question's FIRST paint rendered open; the dock measured itself
tall and offered that height to the spacer, and the collapse one tick
later could never give it back (fold immunity refuses drops while a
question pends). Collapsed bar, tall reserve — scroll as-if-opened.
Fix (owner-shaped, deliberately minimal): apply the arrival state DURING
RENDER, before the first paint (the render-time state adjustment
pattern) — revealArrival(wasQuestion) decides how a teach question
lands: after the composer → collapsed (the learner reads the teaching
first); after ANOTHER question (a queued parallel check) → OPEN — the
reading already happened, and the height it already has is the honest
height. The paint measures the truth, so the reserve never learns a
height that isn't on screen. Probe arrivals are untouched — instant-open
at the bottom stays. The old [pendingId] effect keeps only its probe
branch.
Live-verified on :3001 with a fabricated teach-phase session driven
through the real settle/adoption path (zero model calls): composer→teach
arrival lands as the 59px bar with the spacer growing 206→211 (+5) and
the transcript scrolling only 77px (the new message itself) — the
previous behavior jumped the reserve +339px and scrolled 660px;
teach-check reveal click still expands (425px) and scrolls to bottom;
a queued parallel question arrives open with the reserve unchanged
(577→577) and +40px scroll.
Pins: tests/reveal-arrival.test.ts — revealArrival(false) = collapsed,
revealArrival(true) = open (the owner ruling). 946/946.
Owner-approved full pass (tooltip wave, 2026-09-25). Mechanism: icon/pill
controls ride the styled Radix tooltip via a new Hint composition
(components/ui/hint.tsx) — the shared primitive adopted as-is, no
restyling; the shortcut appends as an em-dash suffix off the LIVE registry
(a rebind in Settings → Keyboard changes the suffix): "Dictate — ⌘D".
Surface-owned keys (Esc closes, Enter submits) ride a raw binding prop.
Native title= stays only for truncation-reveals (dock spans, the rename
field) and menu-interior abbreviations (MenuShell untouched). One
em-dash pattern; per-state truth on the send square (busy = "Stop the
reply — Esc", idle = "Send — ↵"), Enter on the plan verdict's accept,
"Stop dictation (Esc)" kept verbatim per the owner's flag.
Copy: the dev-speak cull — no pipeline/prober/tier/wave vocabulary
anywhere a user reads. New pins in tests/tooltip-mechanism.test.tsx
(hintText composition, asChild no-wrapper-DOM, the auto-update hint
constant); dock truncation titles pinned native; graph-chrome + settings
pins updated to the new copy.
Live-verified on :3001 — 12.6ms hover appearance (no native 1s lag);
glyphs render on every suffixed surface (⌘D, ⌘K, ⌘⇧F, ⌘,, ⌘B, F2, ↵,
Esc); per-page probes confirm Settings entry, Home logo, session search,
Start session, Close sidebar, model + thinking pills, stats summary,
rename title, plan chip, coverage ring, send square, cost button, node
search (⌘F), graph gear, settings close (Close — Esc), keyboard section
"Fixed — part of this screen", auto-update hint, light-mode "Coming
soon", the three slot rosters, graph-tools hint. Esc still closes the
settings dialog.
Over Platform's 66dbea9 key-presence seam, the five owner-locked tasks:
1. Zero-keys pill — components/setup-cta.tsx: pure hasProviderKey
(classifyKeys reuse — tool keys like Deepgram never count), useZeroKeys
(GET /api/keys, re-read on the new lh:settings-saved 'keys' signal),
ZeroKeysPill (testid setup-cta-keys) taking PRECEDENCE over the models
CTA on chat/home/new-session. SettingsSection gains 'keys'; keys-section
dispatches after save AND delete.
2. Setup stage 1 is self-contained: provider picker (MenuShell +
ProviderMenuList) + key paste + Save via PUT /api/keys/:id — the dialog
STAYS MOUNTED; its own save dispatches the keys signal, the fetch effect
re-runs on it, and the stage flips to the pickers on its own.
onOpenKeys + setup-open-keys die. SetupKeysBody exported for SSR pins.
3. parseProviderKeyRefusal (chat.tsx, envelope-verbatim twin of
parseModelsRefusal) joins the revert-draft-and-open-setup interception;
activity-dock reads 'needs a provider key'; dispatch-card shows
'Add a key' — both deep-link setup, which lands on stage 1 by itself.
4. Dictation trio: the token route's only 503 (missing Deepgram key)
becomes the client-authored DICTATION_KEYLESS short line; the errored
idle click clears the error and deep-links Settings -> Keys ->
DEEPGRAM_API_KEY via the new open(section, row) seam (dialog landOnRow
-> SettingsDialogBody initialRow -> KeysSection openRow follow-effect).
Live-verified on :3001 under LH_IGNORE_ENV_KEYS=1 (the fresh-user switch):
pill on all three surfaces + precedence over the CTA, stage-1 save ->
auto-advance with dialog mounted, real 503 send refusal reverting the
draft and opening setup, dictation 503 -> short line -> Settings lands on
the unfolded Deepgram row with the error cleared, keys-section save and
delete both flipping the pill live, and a fabricated refused research job
rendering the 'Add a key' dispatch card. 926/926, tsc/lint/build green.
The scroll-to-bottom disc (ConversationScrollButton) was the scroller's
last child — DOM order painted it ABOVE the dock overlay, so it floated
at the scroller's bottom edge, partly covering the composer (owner
refinement round 3, 2026-09-24: it must ALWAYS sit above the composer,
never over any element, and stay positioned with respect to the
composer — which grows a lot taller when a question owns the band).
The fix is a DOM move, not a coordinate: the button now renders INSIDE
the DockOverlay, as a sibling of the DOCK container within the dock
block — `absolute bottom-full mb-2`, anchored to the BLOCK's top edge.
Because the anchor is the block itself (not the scroller's bottom):
- it sits above the composer shell AND the stats row, in the
gradient's transparent band;
- it rides every height change LIVE: typing growth, a pending
question expanding the band (verified: the disc moves with the
393px question form, 9px clear), the 180ms fold morph to the slim
bar (8px clear of the folded bar — no overlap mid-morph either,
height animations reflow per frame);
- pointer-events-auto (the overlay is none by default); the
isAtBottom/scrollToBottom context still reaches it — DockOverlay
is a Conversation child.
The use-stick-to-bottom pattern comment updated; the shared
ConversationScrollButton component itself is untouched (positioning
arrives via className, twMerge swaps bottom-4 → bottom-full).
Live-verified on :3001 with a tall transcript, all three states:
chat mode (15px clear of the composer textarea), expanded question
(9px clear of the form top), folded slim bar (8px clear). 902/902.
The owner-locked extraction-UX redesign (dispatch 2026-09-24), UI
chunk over Platform's data layer (eed4b4e — GET /api/chats/[id]
carries `coverage`, the threshold is Graph ▸ extractionChunkTokens):
1. SIDEBAR BLOCK (components/extraction-coverage.tsx +
sidebar's session menu): the coverage info block at the TOP of
the context menu — uncovered tokens, uncovered messages, running
state — then the Extract affordance (optimistic "extracting… /
Running" until the server says running; 409 = benign, the trigger
raced us), fetched ON MENU OPEN (the block mounts with the menu —
on-demand, never eagerly), polled at 2s while a run is in flight.
Visually separated from Rename/Delete by a hairline; those stay
the two session-management actions. Failed read → nothing (never
fake zeros).
2. RING METER (CoverageRing, chat top bar, before the PhaseChip):
a 16px donut filling uncoveredTokens / extractionChunkTokens —
"how close this session is to its next extraction run". Click →
the exact numbers (covered/uncovered tokens, messages, the
threshold, lastFoldedAt via timeAgo). Refresh chains: mount, the
turn-settle bump (chat.tsx coverageTick), the settings signal
(threshold/kill-switch re-read live), a 2s drain-poll while a run
is in flight, and the ACTIVITY FEED — a run for this session
appearing or settling in the dock's feed re-reads coverage (runs
fire from other surfaces — the sidebar menu, the turn pipeline —
and no turn-settle bump announces them; the runs-empty-transition
the graph view rides). Vanishes entirely when the graph is off or
before the first honest read. Running = the whole ring PULSES
(implementer's visual call: motion is the working-state signal in
this app's quiet register).
3. PHASECHIP UN-PILL: the phase label is a flat tag — no border, no
rounding, no padding. The Plan chip is the only button-shaped
thing in the top bar. Drag-surface (pointer-events-none) and aria
semantics unchanged; the transcript PhaseDivider keeps its look.
4. PANEL DEATH: components/graph-coverage.tsx, its graph-view
mount, and tests/render-graph-coverage.test.tsx die whole.
Confirmed: after this, NOTHING client-side consumes GET
/api/graph/coverage — the only remaining client use of the
coverage endpoints is the Extract POST (kept, by design); the
routes stay (Platform's call).
Pure seams exported for tests: ringFill (clamp math),
CoverageMenuBody, CoverageRingGlyph, CoverageRingDetail.
901/901.
Live-verified on :3001: ring fill arcs match the API byte-for-byte
(483/4096 = 4.45/37.70; 1991/4096 = 18.32/37.70); the sidebar block
reads "2k uncovered / 4 messages unextracted"; a real Extract run
went optimistic → dock run → menu settled "fully extracted" AND the
ring drained to the empty arc (the feed chain); graph off → the ring
vanishes; the flat phase label renders chrome-free beside the Plan
pill.
Part 1 — the surface:
- Settings ▸ Agent ▸ Tools gains the prober GRAPH-TOOLS row as its FIRST
row (owner ruling: in the Tools section, not the tool-toggles list —
its subject is the map, its mechanism is agent capability). Writes
agent.graphTools {enabled} (absent = on, the toggle discipline) via
the section PUT; composeGraphTools pure seam. Its own key in the agent
contract — never in the tools map.
- The row carries the search-row hint discipline: "Connects prior
sessions' knowledge into new ones — the prober reads the knowledge
map" (owner wording). The honest interplay composes both: when
graph.enabled is off the row dims + disables with a "Graph is
disabled" hint (the effective gate itself is server-side, per-turn
fresh). AgentSection reads the graph section (record.graph, passed by
the dialog) — read-only.
- Cluster labels wrap: wave 7 made labels session names and FILE PATHS,
which the old single-line fillText could not bound. wrapRegionLabel
(pure, exported): word-wrap first, char-split long paths (the
node-label discipline), capped at 3 lines, overflow folded into an
ellipsized last line (owner: wrap, cap 2–3, ellipsize). Pre-wrapped
once per geometry build at the SCREEN-fixed size (viewer-relative
labels — the wrap never goes stale mid-camera-move), drawn centered
line-by-line at the centroid. Deleted-session clusters keep their
hull-only honest silence (label null → no lines).
- The orphan sweep: StatsBar's leftSlot prop is dead (its one consumer,
the knowledge-context slider, died in c690181) — removed, and pre-chat
views now render NOTHING on the left (owner ruling this round: a
"— · —" summary over an empty popover is pure noise) — the dictation
middle just widens. Stale knowledge-context comments swept in
chat.tsx (adoption-window + activity-nudge docs), activity-dock.tsx
(useActivity ground rules), dictation.tsx (×2 adaptive-meter docs),
home.tsx (pre-chat dock comment).
Part 2 — Teacher's c690181 review: clean. Imports, KIND_ICON/KIND_LABEL
entries, render branches and event seams all excised whole; the probe-
handoff branch re-indented intact and merge-adopted; tests updated in
lockstep (merge-adopted, record-adapter, render-new-session retired).
Platform's 4b60e0f roster-string fixes (labeler dropped from both
SLOT_META blocks) reviewed — consistent across agent-section +
setup-dialog.
Gates: tsc clean, 854/854 (8 new: wrapRegionLabel cases, graphTools
SSR rows ×3; stats-bar + ladder fixtures updated to the new rulings),
lint 0, build ok. Live on :3001: graphTools row first-in-group with
hint, off→{enabled:false}→on round-trip, graph-off dimming + hint +
disabled toggle, pre-chat rows summary-free, chat view keeps its
summary, long-title cluster label wrapped without console errors
(visual read unavailable this session — pure wrap is unit-pinned).
The owner's zero-configuration ruling (2026-09-23): nothing configures
the graph beyond chat models. The prior-knowledge injection — its budget
dial, its runner, its marker part, its dock/dispatch entries — is
removed across all bands; the prober now reads the learner's persistent
knowledge map on demand through two gated tools instead:
- list_sessions: every birth session, names + kinds + counts + dates
(graphSessionIndex, Platform's core seam)
- list_propositions(session): what was born there + the session's own
latest understanding level + that evaluation's date
(graphSessionPropositions) — names/levels/dating ONLY, no statement
text (owner ruling: the map's organization is the prober's business,
its prose is not)
Registered via createProberTools when chat-handler passes graphStore
(the graphToolsEnabled gate decided server-side, per-turn fresh); the
prober doctrine gains the map-review nudge at session start.
Also: runner's JobKind special-casing collapses (KIND_ROLE/KIND_RUNNERS
are plain records again), semantic-clusters.ts dies with its test, the
embedder dep's doc tells the wave-7 stub story, and ports.ts's
readClusterLabels/readNamingRuns surface stays untouched for Platform's
ledger-port interlude.
848/848 tests, typecheck/lint/build green.
- Right-click any session row: MenuShell context menu, exactly two actions
(owner ruling): Rename and Delete… — MenuShell discipline (placement,
focus, arrow cycling, lh:menu-open announcements), agent-section's
outside-click/Esc contract with [data-lh-menu] carve-outs.
- Rename is INLINE in the row (owner ruling): select-all entry, Enter/blur
commits through the existing plumbing — PUT /api/chats/:id then
lh:chat-renamed. The open chat header now LISTENS for the event too, so
renaming the active session from the sidebar updates the header in place
(it only dispatched before).
- Delete opens the two-option dialog (locked design, DELETE_OPTIONS pure
seam for render pins): "Delete the session only" (preserve-graph — the
graph keeps every contribution) vs "Also remove its graph traces"
(delete-nodes — shared topics survive, only-this-session topics die).
409 turn_in_flight/jobs_running and 503 extraction_winding_down render
the server's line verbatim with a same-mode Try-again.
- Success: the sidebar patches its snapshot instantly, the active session
navigates home (chat unmounts — drafts die with it), and lh:graph-revision
announces the returned revision — the graph view refetches the snapshot
AND costs (deletion is the one writer the status poll can't see).
- ChatDeleteMode type re-exported through the core barrel (the shell's one
needed seam from core/graph-deletion.ts).
Gates: tsc clean, 909/909, lint 0, build ok. Live-verified on :3001:
context menu open/Esc, inline rename (row + API + active header), both
delete modes on scratch sessions, simulated 409 + retry-then-succeed,
active-session navigation home, graph+costs refetch deltas on home.
- lh:settings-saved CustomEvent (the lh:chat-renamed pattern, owner
ruling): every writer (composer save, SetupDialog save, the settings
write seam — one dispatch point carrying the section) notifies after a
successful PUT; every reader re-reads its own seam: ModelControl's
record read, the Agent section's slots/catalog, and chat.tsx's
configured state (clearing slots mid-session now surfaces the setup
CTA without a reload). Self-notifications re-read idempotently.
- MenuShell now announces lh:menu-open/lh:menu-close (still placement-
only); the settings dialog's content column locks (overflow-hidden,
count-based) while any menu is open — wheel-scrolling with a dropdown
down is dead; the re-place-on-scroll fallback stays for un-lockable
surfaces.
- The settings thinking menus' 'Default' noneRow is retired (owner:
the unset state can't re-enter once chunk 2's auto-populate lands; no
unpin is needed).
- Live-caught regression from the MenuShell migration fixed: the
composer's outside-click handler never learned the portaled menu body
isn't a container descendant — menu-row pointerdowns counted as
outside and ATE THE PICK (composer picks have been dead for pointer
users since 221c99d); rows now carve out via [data-lh-menu], the
settings sections' discipline.
- Item 4 (resolved-level display for legacy records) deferred to chunk 2
per owner ruling — the payload omits the level for legacy no-level
records and DEFAULT_THINKING_ORDER was server-only (now in the core
barrel via bd68486, exactly the seam chunk 2 needs).
- existing sessions: a send that lands the 503 models_not_configured envelope
renders as the SETUP FLOW, never the error banner — the record refetch drops
the refused orphan (the gate refuses before the write-ahead persist), the
draft goes home to the composer, and Save auto-replays it (owner ruling);
parseModelsRefusal reads the envelope from the transport's verbatim error
text (exported pure, tested)
- persistent 'Set up your models' CTA above the chat composer while
slotsConfigured is false (server-read seed via app/chat/[id]/page.tsx) —
legacy per-chat overrides may still send with the pill present; the server
gate stays the truth
- SetupDialog prefill + 'needed' marks (owner ruling): pinned tiers open with
their pick (matched in the federated catalog — lapsed keys don't prefill),
unpinned tiers carry the mark; Save still requires all three;
prefillFromSlots exported pure, tested
- refused jobs read as INFO: dispatch card shows a quiet 'needs model setup'
line + 'Set up models' deep-link (ChatActionsContext.openSetup); dock blobs'
activity reads 'needs model setup' instead of 'failed'; real failures keep
the red tone; isModelsRefusal added to the envelope's home module
- agent-section ModelsBody.slots type honest to the nullable contract
Item 1 — no Default in the slot selectors (owner, fix wave): a
"default" model is not a thing in this app. Picking Default used to
delete the slot and re-arm the first-run gate — incoherent. Now the
model menus carry NO clear row at all (a slot holds a pick or reads
"—"), the thinking menus carry a "—" row (menu-none) that clears only
the thinking pin, and the hint line is the roster only — the resolved
factory pin is never named. Master's amendment confirmed: the
factory-pin hint is SUPERSEDED entirely (Platform removes the pins).
Item 2 — the composer control IS the strong slot (owner: "one and the
same"): ModelControl (model-picker.tsx) reads the raw models section
(GET /api/settings) and the federated catalog, and writes
PUT /api/settings/models via composeSlotModel — the same seam the
settings Agent section and SetupDialog use. Props retired: chatId,
settings, onSaved, fallbackModelId, noDefaultModel — the control is
self-sufficient and identical on every surface (composer, home dock,
/chat/new). The per-chat model PUT is gone from the UI; legacy
per-chat overrides keep resolving server-side. StatsBar/DockOverlay
plumbing slimmed accordingly; chat.tsx drops its per-chat settings
state and initialSettings prop.
Item 3 — the composer catalog cache no longer freezes per page load:
fetchCatalog(force) invalidates, and every model-menu open revalidates
in the background (stale-while-revalidate — cached list shows
instantly, the fresh federation swaps in). Owner's repro verified live:
a key saved mid-session widens the menu without a reload.
Item 4 — stale doc comments fixed: model-catalog.ts's "OpenRouter
catalog"/"multi-provider future" (Platform's file — mechanical,
flagged for review) and model-picker's retired "config-pinned"/
per-chat-PUT narrative.
Live-verified on 3001: unpicked slots read "—" everywhere; model
menus have no clear row; the "—" thinking row clears the pin but
keeps the model; composer pick writes strong and preserves cheap/
vision siblings; thinking pill writes strong.thinkingLevel; groq key
added mid-session appears in the open menu without reload. Caught in
live verification: the record fetch was mapping the WHOLE settings
record into the models section (picks never wrote) — fixed to read
body.settings.models.
Gates: typecheck, lint, 807/807, build.
Root cause of the live-test bugs (2026-09-19): pi passes every run's
AbortSignal to tool execute, but toPiTool dropped it — agent.abort()
only landed at the next step boundary, so a long local grep held its
turn hostage until the dev server's memory restart killed the run
mid-tool, freezing the record at its write-ahead state (the 'went
back in time' rollback).
- lh-tool: execute ctx gains signal (parent + timeout combined); every
call races a 30s system-level timeout with a 250ms cooperative grace —
abort kills a stuck tool but lets a poll-the-signal tool deliver its
partial results. Pre-aborted signals never fire listeners: checked
upfront. toolTimeoutMs is the test seam (pdfModelOverride idiom).
- fs-shared: walkFiles returns {timedOut, aborted} instead of throwing —
abort yields partial results.
- grep: the scan phase was unbounded (the walk's 10s deadline only
bounded collection) — same budget now checked per file; find/grep
check the signal and return partials with a notice.
- pending/chat: streamingToolActivity names the ACTUAL running tool —
a part sits input-streaming through its whole execution, so the old
shimmer called minutes-long greps 'Writing a question…'.
Tests: toPiTool guard (forward, abort, timeout, grace, question
bypass, ToolInputError passthrough), walk mid-abort, find/grep
pre-abort + mid-run abort, activity labels. Gates: 617/617.
The browser e2e check on the live dispatch session found the tool-pdf
part rendered NO card: the artifact renderer registry had no pdf entry
(the doc comment literally says "add your renderer here"), and the typed
TeacherUITools union + the chat.tsx job-part enumerations (poll cadence,
dispatch nudge, settle-sync) all missed the new tool kind.
- PdfJobCard in job-card.tsx: the research JobCard idiom, kind pdf —
registered in artifacts/index.ts.
- ui-message.ts: pdf entry in TeacherUITools (fire-and-forget ack output).
- chat.tsx: two mechanical "tool-pdf" additions to the research/demo/visual
enumerations — UI-stream file, flagged for review at merge (same
treatment as the activity-dock type-forced entries).
- demos.test.tsx: live/settled dispatch-card renders + registry pin.
The shortcut architecture (owner-approved plan, 2026-09-17):
- src/keyboard.ts: pure action registry (id, label, scope, default
binding) + lh:keymap localStorage overrides + event normalization.
The settings milestone renders this same registry as its editor.
- components/keyboard-shortcuts.tsx: the single document-level keydown
listener (window level, bubble) owning the pieces no surface owns:
- Tab is DEAD (owner decision): browser tab-focus navigation is
abandoned; Tab is suppressed everywhere, Shift+Tab cycles the
thinking level.
- Esc is a layered dismiss: topmost popover first (popovers
stopPropagation their own Esc — usePopoverDismiss + ModelControl),
then dictation stop (lh:kbd-stop-dictation), then abort of an
in-flight turn (lh:kbd-abort-turn; server-side wake turns are
honestly non-stoppable — the seam only stops local streams).
- Global actions dispatch lh:kbd-<action> window events (the
lh:chat-renamed pattern); surfaces that aren't mounted simply
don't listen.
- The input guard: inside a typing surface only mod/alt combos (and
Tab) fire; bare keys pass through to the field.
- Cheat sheet overlay (?) renders from the registry, grouped by scope.
Bindings wired this wave: cmd+d dictation toggle, cmd+k model picker,
shift+tab thinking cycle, cmd+b sidebar, cmd+shift+f sessions search,
alt+up/down + alt+enter session cycling, cmd+g graph view, cmd+n new
session, f2 rename (chat), ? cheat sheet. Wave B adds the question
keyboard machine; wave C the per-surface bindings (find, import).
Verified live in the dev shell (CDP): all of the above via
Input.dispatchKeyEvent.
Co-Authored-By: Claude Fable 4.5 plus-plus <noreply@anthropic.com>
Owner's first live mic test (2026-09-17) drove four changes:
- ONE primary transcription input at all times: while a quiz/ask
question waits, finals land in its field (text mode → 'Your answer',
select modes → 'Your thinking') via a module-level claim registry
(useDictationTarget in dictation.tsx); the composer is the fallback.
Esc in a question field also stops dictation (threaded through
DockOverlay → QuestionDock).
- No interim results: live words were distracting and shifted the meter.
Params now match the pi dictate extension exactly (no interim_results,
endpointing=300, no utterance_end_ms); the meter's width stays fixed.
- Optimistic meter: the click enters the meter view immediately (Stop
pill + meter within one frame, verified at 150ms in the dev shell);
token mint and mic acquisition run concurrently behind it.
- Wide idle button: the stats-bar middle is one full-width affordance —
a quiet bordered 'Dictate' button when idle (failure text replaces the
label), Stop pill + fixed meter while live.
Live-verified in the Electron dev shell: wide button renders on home
(167px) and chat (358px), meter + Stop appear at 150ms after click,
flat 2px bars in silence, clean return to 'Dictate' on stop.
Two-band architecture from docs/research/stt.md: renderer captures mic
audio in an AudioWorklet (linear16 16kHz mono, ~128ms frames, per-quantum
RMS for the meter) and streams it directly to Deepgram over WebSocket —
audio never loops through our server. The real key stays server-side;
/api/stt/token mints a short-lived grant token (TTL 300s) per run.
Patterns ported from the owner's pi dictate extension (tuned in real
use): generation counter guarding every async handler, the
latest-RMS-never-resets meter (-50..-10dB window, 60ms tick), mic-first
stop + CloseStream + 3s safety net, endpointing=300 with the
utterance_end_ms=1000 backstop.
UI (owner placement): mic button in the stats bar's middle slot on all
three surfaces (home graph-view, chat, draft); while recording the meter
unfurls to fill the row's free width with the interim ghost at its
flank. Esc ends dictation from the composer. Token mint comes before
getUserMedia so a keyless app fails fast without a TCC prompt; one
automatic reconnect on abnormal socket close, errors surface inline.
Live-smoked the full error path in the dev shell (bar renders on all
three pages, 403 grant failure shows inline, mic untouched). The full
audio path needs a Deepgram key with Member permission to mint tokens.
Traffic lights (owner report: 'way too low, not aligned with the phase pill'):
measured in the running app — circles centered at 30.75pt vs the pill box's
29.25pt (band 0-59.5, pill box 19-39.5, pill ink 28.75; uppercase glyphs sit
high). trafficLightPosition y 24 -> 22.5 lands the circle centers at 29.25,
exactly on the pill. Owner confirmed aligned after this change.
Rename escape hitbox (owner report): entering rename was already width-of-
title only, but the editing input spanned flex-1 so clicking the rest of the
bar stayed inside the input. The input is now shrink-to-fit
([field-sizing:content], min 8ch / max 70%) and any mousedown outside the
input blurs it, running the existing onBlur commit path; the header drops
lh-drag while renaming so the drag region cannot eat the escape click.
Verified: click-away commits (exactly one PUT), Escape discards, Enter
commits.
Phase pill is pure decoration (owner report: 'should not be interactable
in any way'): no tooltip, pointer-events-none, select-none — part of the
drag surface instead of a carved-out target.
Graph chrome was unclickable in Electron (owner report): home.tsx's
floating drag strip underlays the z-10 controls row, and Electron's
app-region mask ignores z-index, so the region ate their clicks (plain
browser unaffected — DOM hit-testing is fine). lh-no-drag carve-outs on
the chrome row and the node detail card (it is placed imperatively
anywhere, including the top strip).
LhMark home link now mirrors the ghost toggle: same size-9 chip geometry
and hover:bg-accent treatment.
The system title bar disappears (titleBarStyle hiddenInset +
trafficLightPosition {16,24}): the app's own 60px header bands become the
window's drag surface, and the traffic lights float over the sidebar's
mark band, vertically centered on the same line as the chat header's
phase pill (pixel-verified: circle centers y 24–36.5pt, pill box center
30pt).
- electron/main.cjs: hiddenInset chrome; dev shell (LH_DEV_URL) gets its
own userData (-dev suffix) so it never fights the packaged app's
single-instance lock — verified: the lock silently killed the dev
shell while the packaged app was open (that also proved the lock works
across separately-launched builds).
- app/globals.css: .lh-electron-gated utilities — drag/no-drag bands,
92px traffic-light gutter (.lh-lights-gutter / .lh-closed-toggle:
lights end ~75pt + the lights' own 16px edge inset as the gap),
home's floating strip becomes a drag surface only in the shell.
Everything is inert in the plain browser.
- sidebar.tsx: SidebarProvider sets <html class=lh-electron> post-mount
(UA detection; no hydration mismatch, absent in browser); LH wordmark
text replaced by the owner's mark (components/lh-mark.tsx, inlined
icon-no-padding.svg, 24px, home link kept); SidebarToggle gains the
92px margin when closed so it never slides under the static lights.
- chat.tsx: the rename hit-box shrinks to the title text itself —
flex-1 spanned the whole bar, so every click started a rename and the
window lost its grip. Shrink-to-fit + a flex-1 drag spacer (chips stay
right) + hover:bg-accent affordance (clear when hover renames vs
drags; -mx/px cancel for zero shift). The renaming input keeps flex-1
for edit room.
- home.tsx / new-session.tsx / phase-divider.tsx: drag bands + carve-outs.
Live-verified in the dev shell (pixel analysis) and headless browser:
lights/pill/mark centered, 17pt lights→mark gap, rename click→select-all
→Escape roundtrip, browser layout untouched.
Gates: typecheck, eslint, 500/500 tests, build.
Rename (owner ask): clicking the header title opens edit mode with the
whole title selected — typing replaces, arrows move inside. Enter or a
click away commits (PUT /api/chats/:id); Escape and an empty commit
discard. renameChat joins the ChatStore port; like saveSettings it is
NOT conversation activity — updatedAt stays put so sidebar order never
churns on a rename. The sidebar patch is instant: the header dispatches
lh:chat-renamed and SessionsSidebar folds it into its snapshot (the 2.5s
poll is only a backstop). Sidebar right-click rename is DELIBERATELY
deferred: it needs a real context-menu system, which gets its own pass
later (owner decision, 2026-09-17).
Graph chrome:
- The extraction-coverage button moves between search and import; the
gear keeps the far-right corner. Coverage still renders on an EMPTY
graph (that is exactly when it matters).
- The no-match search note no longer shifts its siblings: the chrome
row is items-start (every button is 44px tall, so top alignment pins
the row while the search container grows downward).
- Popovers dismiss properly: pointerdown anywhere outside the control's
root (button + panel) closes it, and so does Escape. Two panels can
never be open at once — the sibling chrome buttons are outside the open
root, so pointerdown closes the first before the click opens the
second. New shared hook components/popover-dismiss.ts, wired into
both the settings gear and the coverage control.
Gates: typecheck, eslint, 498/498 (+2 rename store tests), build.
Live-verified: select-all/focus/Enter/Escape/empty/blur round-trip,
updatedAt untouched, sidebar instant patch; chrome order + frozen
positions under a no-match query; popover close via canvas, search
input, Esc, mutual exclusion, toggle intact.
h-10 → h-15 (60px) in both the chat header and the /chat/new draft
header. 60px is an alignment, not just a size: it equals the sidebar's
LH wordmark band, so the header's border-b continues that band's bottom
edge as one horizontal line across the app — the sidebar's 'New session'
row begins exactly where the chat column's chrome ends.
Live: header border-b at y=60, wordmark band bottom at y=60. Gates:
typecheck, lint, 469/469.
The round-1 experiment put the braille spinner in the chip's old dot
slot; rejected live: the pill's curve and the square braille cell fought
for space (visually squeezed), and the chat surface already shows
in-flight work (thinking shimmer, the dock) — the chip stating the phase
is its whole job. PhaseChip back to a label-only pill (busy prop
removed, single-child gap dropped); work-state lives in exactly one
place: the sidebar row's spinner. Doc note + test pin the decision so
it doesn't get re-tried blind. Gates: typecheck, lint, 469/469, build.
Nothing exists purely for decoration: phase color-coding is retired
app-wide (phases only advance — every session ends Teaching — so
phase-as-signal was decoration with a one-time function), and what
state genuinely deserved — 'a turn is in flight' — is now carried by
MOTION alone, in the terminal's own glyph language.
- BrailleSpinner (components/braille-spinner.tsx): the canonical
cli-spinner sequence (⠋⠙⠹⠸⠼⠴⠦⠧⠇⠏) as exact dot bitmaps in a 2×4
braille cell, 80ms/frame (800ms cycle). Pure CSS: one shared keyframes,
frames staggered via animation-delay; no JS timer, no re-render per
frame — cheaper than the old breathing dot. Static contexts (SSR,
prefers-reduced-motion) show frame 0 — a legible glyph, not a blank.
- Sidebar rows: spinner ONLY while a teach turn is in flight; idle and
question-pending render NOTHING (a resting state has no scan value —
the meta line's 'Waiting for your answer' text carries it). The
three-state phase dot (pulse/solid/hollow) and PHASE_FILL/RING maps
are gone.
- Top bar PhaseChip: label-only pill; the old dot slot becomes the
spinner's home while a turn is in flight (busy prop wired from
ChatView's busy = localStream || turnInFlight). Transcript
PhaseDividers drop phase colors for monochrome rules + labels.
- API narrowing (app/api/chats/route.ts + ChatListItem contract):
busy = isTurnRunning ONLY — background jobs no longer count (the
sidebar's Background agents section is their channel; a per-row
spinner would double-signal them).
- globals.css: --phase-* tokens + lh-dot-breathe removed (dead after
the sweep — no remaining users).
Live-verified: idle/pending rows render no indicator, /chat/new chip is
label-only, 10 CSSAnimations at 800ms with correct 80ms stagger, static
frame 0 under reduced motion. Gates: typecheck, lint, 470/470, build.
Lane crossings to flag at merge: app/api/chats/route.ts + the
ChatListItem contract comment (src/core) — the busy-semantics narrowing;
components/chat.tsx one-line busy prop.
The owner's three-part test of the context-injection flow surfaced
timing/visibility bugs, fixed as standing ground rules for ALL background
processes (owner rules, 2026-09-15):
1. THE JOB STARTS AT SEND. triggerGraphContext moves from onFinish to
right after the write-ahead persist in handleChatRequest: its only
input is the first typed message (the topic), just made durable — so
embedding + cosine scan run CONCURRENT with the prober's turn and
the dock blob appears the moment the session starts, not after turn
one finishes. Extraction stays at onFinish (it needs answers
accumulated DURING the turn).
2. DELIVERED CONTENT SHOWS MID-TURN. Steers inserted server-side
(claimed at turn start, persisted write-ahead) used to wait for the
next turn's completion to appear — the full record sync mid-stream
is forbidden (it breaks the SDK's in-place continuation merge), so
adoption was deferred to idle. New append-only seam: core
mergeAdopted() inserts the record's new message ids in canonical
order (in-flight streaming tail kept last) and never rewrites a
local message; any other shape bails to the full-sync-at-idle net.
Wired at three points, no new standing polls:
- turn-start window: three bounded record fetches (400ms/1s/2s)
after a send catch the claimed steer while the turn streams;
self-terminating, survives the submitted->streaming flip.
- jobs-poll settle: mid-turn settles adopt append-only NOW, the
owed full sync still runs at idle.
- wake-adoption poll: each tick adopts persisted wake content, so
steers show at the wake turn's START, not its finish.
The dock stays honest with zero gap: jobs hold "running" while
queued behind a pending question ("delivering context"), then flip
done at claim/delivery — visible from start to delivery, and the
turn-start nudge (useActivity().refresh) clears it within ~150ms.
3. THE MARKER IS CLICKABLE (observability). The knowledge-context
SystemLine becomes GraphContextMarker: quiet line, popover with the
delivered propositions IN DELIVERY ORDER (most similar first — the
order the budget filled), each with its folded understanding level;
Esc / outside pointerdown close. The part data was already the audit
trail — now it's readable at the seam where it landed.
Tests: mergeAdopted unit pins (6), marker render pins (3) — 468/468,
tsc, eslint, build all green; live-smoked end-to-end: blob at send,
marker at t+500ms mid-turn, popover order + levels, instant dock clear.
One shot per chat, fired at the first turn's onFinish: the session's
topic (its only typed message) is embedded as an EmbeddingGemma retrieval
query; the semantically closest graph propositions are selected under the
session's token budget (pure budget, no similarity threshold — the
decided doctrine) and delivered as a marker-only steer through the wake
pipeline, queuing behind pending questions like every sub-agent result.
- contracts: ChatSettings.graphContextTokens (default 1024, max 4096,
step 128, 0 = off); GraphContextData + graphContextText; the
graph-context JobKind + buildGraphContextSteer (id steer_ctx_<chatId>)
- core/graph-context.ts: brute-force cosine top-K, greedy chars/4 fill
under budget, most-recent UnderstandingLevel fold, count cap 64,
parse-don't-cast at the marker seam
- server: embedQuery (role-asymmetric query format, same warm singleton
— vector-space parity by construction); the runner (startGraphContext-
Injection one-shot gates; runGraphContextJob validates ledger parity);
wake branches (chatIdOf/jobIdOf, finalizeJob done, recoverStaleJobs
re-deliver or contextRespawn); abortJob kills context jobs SILENTLY
(nothing was promised in the lesson); turn-service trigger seam
- adapter: data-graph-context flattens to the canonical block (malformed
parts have no model face)
- UI: the budget slider (home dock + /chat/new draft, the session-data
strip); the quiet SystemLine "~537 · 10 propositions"; dock kind
"Knowledge context"; model-picker PUT now spreads settings (a model
pick can no longer clear the sibling override)
Live-smoked end to end on the real embedder + real GLM: seeded ledger
(10 propositions), session on "modal logic: possible worlds and the
necessity box" — job in the dock, queued behind the prober's question,
delivered on the answer, SystemLine in the transcript, wake turn reads
the block and continues the probe. Gates: tsc, lint, 459/459, build.
Three fixes from the owner's resize report:
- Continuous shrink: the marginalia now mounts whenever a question owns
the band and its box width is pure CSS, so it shrinks continuously as
the display narrows. Below MARGIN_MIN_W (220, the owner's floor) it
FADES over the normal 300ms transition instead of hard-unmounting;
widening past the floor fades it back in. No mid-question unmount, no
pop, no invisible limbo.
- Live flank tracking: a ResizeObserver on the dock block replaces the
render-time measurement — the fade floor reacts instantly at any
width (window resize, sidebar toggle, scrollbar changes). The old
measurement went stale during drag-resizes (nothing re-rendered on
macOS until the question rewrapped), so the mount decision popped
late and left a sub-floor sliver visible.
- Phantom-fold mute: resizing rewraps the question text, the browser
re-clamps scrollTop and emits the clamp as a scroll event — the fold
machine read the transient position and flickered the box to the bar
and back. Scroll events within 400ms of a WIDTH change are now
ignored (width-only: stamping the dock's own height morphs
self-muted the machine and swallowed the at-bottom auto re-expand,
leaving the box stuck collapsed — caught in live verification).
Scroll-up/scroll-down fold behavior itself unchanged (verified live:
mute-window scroll ignored, post-mute scroll folds, bottom re-expand
works). Gates: typecheck, lint, 423/423 tests.
Retires the fixed bottom-left plan window (which sat ON the sidebar and
bled into the session). Two placements replace it, per the owner's
lock-in:
- The plan chip in the top bar beside the phase chip — session-level
structure beside session state; opens the full-DAG popup; present at
every width; the access point.
- The plan marginalia: the BARE diagram (no card, no border, no shadow)
in the dock band's right flank — marginalia on the lesson's page, not a
floating widget. The band's reserved height (tall while a quiz is
open) is its canvas; meet-fitted (PlanDiagram's new fill mode) so it
fits both the flank's width and the band's height; live statuses +
the yellow current-node glow; click opens the popup. Mounted only
while a question owns the band — composer mode keeps the reading
view pristine. Folding the dock fades + blurs it out of existence
(300ms), returning to the bottom fades it back in. Rendered only when
the flank holds a real diagram (>= 240px); narrower windows keep the
chip alone.
Files: chat.tsx (chip, DockOverlay marginalia wrapper + flank measure,
PlanWindow removed), plan-window.tsx (PlanWindow → PlanMarginalia),
plan-block.tsx (PlanDiagram fill prop: h-full w-full meet-fit),
tests/plan-window.test.tsx (rewritten for the marginalia).
The rhythm agreed with the owner: the stats/controls row hugs the
composer (8px — reads as the card's metadata, one object), and the
whole cluster floats with deliberate air above the screen bottom
(20px). The 8px straight-edge gap optically matches the 6px pill-to-
pill gap: two rounded pills' corridor reads wider than geometric, so
rounded-rounded gaps run ~2px tighter than straight-edge gaps.
- stats-bar.tsx: pt-4 -> pt-2 (8px under the composer) — shared by
chat and home automatically
- chat.tsx dock block + home.tsx composer wrapper: pb-4 -> pb-5
The dock's reserve auto-measures the block, so the spacer/fold
machinery adapts on its own.
border-b border-border — the same 1px border-border line that
separates the left sidebar (border-r on its container), so the top
bar now reads as its own block with the app's standard separator.
Rounds 1–3 chased a soft fade under the title; the owner called the
final overlay look wrong too. Back to the simple picture: the top bar
is its own block — in flow above the scroller, plain page background,
no gradient, no border, no tone change. It reads as a block because
the transcript starts below it. (Also reverts round 3's overlay
header + pt-14 entirely.)
Root cause of the 'solid bar' look: the scroller began BELOW the
header, so the fade zone sat over plain page background — any opacity
there read as a solid bar, no matter the stops. The owner's picture:
content goes under the title bar, easing beneath it.
- The header is now an overlay (absolute top, z-20) over the
scroller, which fills the column to the very top — same floating-
chrome pattern as the home graph view.
- The wash lives INSIDE the header, behind the title content:
60% background at the very top, ramping down to transparent over
48px — never solid, supports the title, content melts beneath it.
- The transcript gets pt-14 (header 40px + its original 16px), so
scrolled-to-top content clears the floating title.
- Header is pointer-events-none except the sidebar toggle, so the
canvas area under the wash stays draggable/selectable.
Verified live: scroller top = 0 under the 40px overlay header; wash
48px ramping from 60%; first message at exactly 56px on scroll-to-top;
the old below-header edge fade is gone.
Round 1 (h-24, 65% hold via 25% mid) read as a faint tail plus an
abruptly solid band and stretched far into the chat. What the owner
wanted: peak ~60% right under the title, ramping down to transparent
a little below — no full opacity anywhere, short reach. Now h-10
from-background/60 to-transparent: a single gentle ramp.
The header fade was h-5 (20px) from full background opacity — it
read as a solid bar with a quick blur on the edge. Now a 96px wash:
65% under the title, holding softly through ~25% at the midpoint,
melting to transparent — supports the title, never reads as a bar.
Two round-1 follow-ups (owner):
- The overlay painted over the scroller's scrollbar track. The scroller
reserves symmetric gutters (scrollbar-gutter: stable both-edges), so the
overlay's right edge is now inset by the measured gutter — half of
offsetWidth − clientWidth; zero on macOS overlay scrollbars, the real
width on classic platforms. Re-measured on window resize. The headless
shell renders classic scrollbars (11px), so the path is live-verified.
- Wheel over the dock did nothing: the overlay is a SIBLING of the
scroller, so native scroll chaining stops at the overlay (the old
in-flow dock chained for free — a real regression, not a preference).
A non-passive wheel listener on the dock block forwards the delta to
the transcript ONLY when nothing under the cursor can take it: the
question area, textarea overflow, and the model menus keep their
native scrolling, including their own edge chaining; an unhandled
wheel (over a fully-visible box, or at an internal scroller's edge)
scrolls the transcript — so wheel-up over the quiz itself now folds
the box under the cursor.
Verified live: wheel-up over the question box scrolled the transcript
(9460→9100) and folded to the bar; wheel-down handled cleanly; the
overlay inset matches the measured gutter exactly.
The question box folded far too late because the old dock was IN FLOW
inside the scroller: folding it shrank the scroll content, so the fold
had to wait until the learner was a full dock-height past the bottom —
any closer and the browser's scrollTop clamp pinned them back to the
bottom, re-expanding the box (the oscillation the old hysteresis
comments worked around).
The dock now overlays the column bottom (absolute — a sibling of the
scroller, the use-stick-to-bottom pattern, like the scroll button), and
a spacer at the transcript's end reserves its height (DockOverlay
measures its settled block and reports upward; ChatView's dockReserve
state feeds the spacer). At-rest geometry is identical: the spacer
reserves exactly the overlay footprint.
The fold is now VISUAL ONLY: COLLAPSE_DIST 64px (was dock-height+64),
EXPAND_DIST 16px, position-based, no velocity. The reserved tail GROWS
immediately (typing, new variants) but never drops while a question is
open — a shrinking tail would reintroduce the clamp; in chat mode drops
apply (the composer typing-shrink is the old harmless behavior).
expandIntent (2s bar-click suppression) and the teach reveal gate are
unchanged.
Verified live (headless chromium, dev 3001, pending-quiz chat): teach
invitation mounts collapsed; click → expand + bottom; scroll up 200px →
folds to the bar at d=200 with the spacer held (no geometry change, no
yank); back to bottom → re-expands; spacer ≡ overlay footprint.
Note: the overlay spans the column's full width, so on platforms with
classic (non-overlay) scrollbars its gradient/background strip covers
the bottom of the scrollbar track — invisible on macOS overlay
scrollbars; flagged for master.
- New session: literal '+' glyph removed — one lucide PlusIcon (the
duplicated glyph was PlusIcon + text '+').
- Top bar: a small pointer-events-none gradient under the header eases
the scroller into the bar exactly like the dock's bottom fade.
- Option subtext (opt.description): .lh-prose — Source Sans 3; 'I don't
know' copy stays JetBrains Mono.
- 'I don't know': now layout-matched with a radio circle; dashed border
preserved as the deliberate non-answer.
- Separator spacing: options group pb-1 → pb-2 so the gap before the
divider equals the 8px inter-option gap.
The plain user-text div renders Source Sans 3 too (.lh-prose) — the
teaching surface reads in one family, chrome labels stay JetBrains Mono.
Build green; gates clean. The reskin's final live-browser pass runs on
this tree (dev on 3001) — no screenshot tooling in this session, SSR
checked.
COLUMN now spans the session area with a thin margin (px-6, no max
width); DOCK (the composer/question shell) is a deliberate narrower
46rem centered column — same elements, same places, width polish only.
- Session title: font-title (CMU Serif bold) at text-2xl, per the mock.
- MessageResponse's markdown container now carries .lh-prose (Source
Sans 3 + emphasis tokens) on every markdown surface; labels stay
JetBrains Mono. Unlayered plain CSS beats the text-sm utility.
- PhaseChip + PhaseDivider sizes now pull from the declared text ladder
(text-xs) instead of pixel literals.
The model-control wave (user-decided design, 2026-09-12): ONE controller
for the conversation agents (prober/teacher are functionally one agent);
background sub-agents stay config-pinned. pi-ai owns the registry — the
shipped 366-model OpenRouter catalog (cost, context, modality,
thinkingLevelMap), getSupportedThinkingLevels/clampThinkingLevel for the
per-model thinking truth, and streamSimple's automatic clamp. No second
registry to drift out of sync.
Seams:
- config.ts: ReasoningEffort retired for pi's ThinkingLevel
(off|minimal|low|medium|high|xhigh|max). GLM roles go honest at "low" —
pi's catalog says GLM-5.3-flash accepts exactly low|high|max, and the old
"minimal" was silently clamped to low by streamSimple since the swap;
nothing changes behaviorally. visualMaker → qwen/qwen3.8-flash (fixes
the dead qwen/qwen3.8-max slug — the catalog only ships -0902; also the
user's cost pick, 0.15/0.47 per-1M).
- src/agent/models.ts: piEngineFor(role, settings?) — provider+id resolved
from the catalog, thinking level from settings ?? config, CLAMPED to the
model's supported list (a settings change between models can never send
an invalid level). PiEngine gains `level` (observability/tests).
- ChatRecord.settings {model?: {provider, id}, thinkingLevel?} — written
by PUT /api/chats/[id]/settings (replace-whole, no updatedAt churn),
read by chat-handler at turn composition: agents are stateless per turn,
so the change is next-turn-effective by construction. The turns log
still records the ACTUAL billed model (the honest witness).
- GET /api/models: the configured-provider catalog as client-safe DTOs
(core/contracts/models.ts) + config defaults; a second provider joins
the picker by registering it in piModels() — nothing else changes.
- PUT settings validation gate (src/server/model-catalog.ts): unknown
provider/model or unsupported level → 400 carrying the supported list,
never a silent clamp.
- scripts/sync-pricing.ts: regenerated PRICES/CONTEXT_LIMITS from pi-ai's
shipped openrouter.json (offline; same source as the engine + picker).
All 366 models priced; glm's stale 0.15/0.5 → pi's 0.075/0.25, limit
1310720 → 1048576 — cost is inferred at read time, so history re-prices.
- UI (components/model-picker.tsx): stats-bar right side becomes model +
thinking buttons; menus open ABOVE; fuzzy search over "provider id
name" (core/fuzzy.ts — pi's /model pattern); model rows show name/id/
price/context; the thinking menu lists exactly the chosen model's
supported levels (GLM: low|high|max — no dead options, no "off").
Model switch drops an incompatible pinned level (falls back to config,
which clamps honestly) rather than carrying a 400.
Tests (350/350, +15): catalog projection + defaults, the settings gate,
pi thinking truth pinned against upgrades (glm low|high|max; minimal→low;
off→low — GLM cannot disable thinking), the engine factory's
override/fallback/clamp/throw matrix, fuzzy ranking, store round-trip,
and the re-priced folds (usage, stats bar, activity dock).
Live smoke (disposable driver, deleted): catalog route 366 + defaults;
glm+minimal → 400 + supported; turn 1 billed qwen/qwen3.8-flash under
the override; mid-session switch flipped turn 2 to glm; clear-overrides
restored the config pin. ALL CHECKS PASS on real OpenRouter.
The custom-message scheme (user-decided 2026-09-12): system-authored
user-role messages are NOT human messages — they carry marker data parts
only, the adapter owns their model-facing wording, and the UI renders
them under control. Nothing fake is ever persisted as human speech.
The inventory (all migrated):
- probe→teacher handoff wake: was a text-bearing user message rendering
as a fake user bubble; now data-probe-handoff marker-only — the
adapter flattens it to HANDOFF_TEXT (core/contracts/chat.ts), the UI
renders a quiet 'Preparing your plan…' system line.
- plan accept/decline: was boilerplate text ('The plan is accepted…',
'I'd like the plan revised…') masquerading as learner speech — the
decline wasn't even marked; now data-verdict {accepted, at} on its own
message, adapter flattens to verdictText(), UI renders the quiet
verdict marker ('✓ Plan accepted' / '↻ Plan revision requested').
sendVerdict sends TWO REAL MESSAGES when the learner typed a note
(the note as a genuine user message, then the verdict) in ONE turn
via setMessages + sendMessage() — no 409 dance, no double teacher turn.
- sub-agent steers: the abort steer gains its data-subagent-aborted
marker part; the UI keys steer rendering off PARTS (artifact/aborted),
not metadata — metadata stays as pump mechanics only.
- record-adapter: systemTextOf() flattens marker-only messages; steers
keep their text (payload — briefs are content, not boilerplate).
phases.ts: plan→teach advances on data-verdict{accepted:true} only;
learner text alone can never sanction teaching (pinned). data-plan-
accepted retired, no back-compat (user-decided: wiped-data doctrine).
Pinned: record-adapter.test.ts (verdict/handoff flattening, note+verdict
order, steer payload kept, UI-only parts skipped), phases.test.ts
(verdict marker semantics, out-of-order, text-never-fakes), jobs.test.ts
(abort marker part), demos.test.tsx (badge renders off the part, not
metadata).
Live GLM smoke (disposable driver, deleted): probe→conclude→wake→plan→
decline-with-note (revised, still plan)→accept-with-note (teach) — ALL
CHECKS PASS, including marker-only handoff/verdict on the wire and zero
fake human speech in the transcript.
Gates: tsc, lint, build, 335/335.
Two user-reported problems from the muscle-protein-synthesis session,
both diagnosed from the live data (chat_mtxeml9e_6118e59f).
The freeze: after probe_done, the open session never showed the
teacher's wake turn — only navigating away and back loaded it. Not an
agent-switch problem: the server-side handoff worked perfectly
(write-ahead persist, liveness registry, the sidebar's busy dot saw the
wake the whole time). The gap was client attach: the only mid-session
wake detector the open page has is the jobs poll, and probe-phase
sessions have zero job traffic, so that poll is OFF; the post-turn
adoption poll read turnInFlight=false at ~0ms (the wake fires at
~0.75-1.2s, after the pump's grace), adopted, and stopped. Fix: the
post-turn adoption effect now keeps a short WATCH WINDOW after adopting
(~4s, 8x500ms) — if turnInFlight flips true, it hands off to the
existing wake-adoption poll (busy shimmer, adopt at finish).
The mermaid failures: four failed plan_init calls, all caused by ONE
hidden rule — underscores are forbidden in node ids (mps_translation,
net_balance, per_meal). The id regex rejected '_' BEFORE the slug check,
so the specific, actionable message never fired; the generic 'can't read
the edge — expected a[title] --> b[title]' lied (the syntax was exactly
that), and the teacher spent three attempts hunting punctuation it
suspected (colons, commas, multi-word labels) — visible in its own
reasoning trail. Per the user's principle — enforce only the slugs,
free everything else: underscores are first-class slug chars; the id
CANDIDATE class is broad so every bad id reaches the specific slug
message; chained edges (a --> b --> c) expand to pairs; ==>, -.-> are
edge synonyms; any direction header accepted (we layout ourselves);
id["title"] quote pairs are stripped. The tool description, zod
describe, and the teacher's planning prompt now state the one strict
rule instead of the old subset liturgy.
Verified: tsc, 305/305 (the exact muscle-protein mermaid is now a
passing test), lint, build; real-browser puppeteer smoke on live GLM —
the teacher's wake turn appeared in the open page ~50s after the probe
concluded with ZERO navigation, verdict gate + first teach check +
plan window all live in-page (the single console 404 is the app's
missing favicon, environmental noise).
The accept-plan UI lands (docs/probe-teach-split.md §4, formerly the
interim rule): presenting the plan no longer moves the phase — the
learner's ACCEPT does. While the presented plan awaits its verdict, the
composer's send square becomes two stacked buttons (green Accept plan /
red Decline); typed text rides the verdict as his note, Enter keeps
sending plain side questions (phase unmoved), decline asks for a
revision and the buttons stay.
The seam: accept sends a user message carrying a data-plan-accepted
part (PlanAcceptedData in contracts/chat.ts, TeacherUIDataParts
'plan-accepted'); phaseAt3 reads it as the plan→teach event (plan_init
demoted to presentation-only); convertToModelMessages drops data parts,
so the teacher sees only the text. Mid-lesson rewrites need no fresh
verdict — acceptance happens once, at the first presentation.
The teacher's planning prompt now ends the turn at presentation and
waits for the verdict (never teaches into an unaccepted plan).
Gates: tsc clean, 304/304 (phases tests rewritten to the acceptance
seam incl. decline/side-question/failed-init cases; composer verdict
render tests), lint, build. Live GLM smoke ALL PASS: phase pinned at
'plan' after plan_init, presentation turn carried NO check, accept
click flipped the summary to 'teach', then tagged checks moved the map
(current=rates, visited=composite); zero console errors.
amosblomqvist