Owner polish pair before the settings milestone:
- Model menu: arrows walk the filtered results (clamped, no wrap) while
focus stays in the search field; Enter picks the walked row or, with no
walk yet, the first result; typing resets the walk; click and Enter
share one pickModel path; walked row gets a muted ring + scrollIntoView;
cheat-sheet display entries added.
- Graph: window resizes re-anchor the camera so the world point under the
pre-resize viewport center stays under the new center — the graph stays
visually still while scaling (was top-left pinned). viewW/H tracked in
syncSize; re-anchor applied through the zoom behavior (persistence and
card-dismiss unchanged); kFit refresh as before. Pixel-verified: 7/7
center samples identical across a 300x200 shrink.
Owner live-test rounds on waves A/B:
- Alt+arrows no longer ALSO walk the question rows (dock ignores
alt/meta/ctrl combos; session cycling keeps them exclusively)
- question walk clamps at the ends — no wrap (faster to navigate)
- plain Tab toggles focus between the thinking field and the answer
rows (walked row by default); focus follows the walk while in the
rows; Shift+Tab stays the global thinking-level cycle
- Esc in the sessions search blurs back to normal (claims the key so
the window Esc chain does not fire)
- cheat sheet gains the Tab-in-question display entry
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>
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.
11 commits on fbeca36: minimal-color rounds 1-2 (phase color retired,
the braille spinner is the one working indicator; the chip is label-only),
typography rounds 1-5 (three-font rule enforced — CMU titles, Source Sans
session titles, JetBrains mono; snapped type ladder in globals.css with
the xxs rung merged into xs; lh-input caret at the font-metric floor;
session titles at prose size, message counts and active-row weight dropped,
ISO dates), header rounds 6/6b/6c/6d (60px top bar aligned to the sidebar
LH band; the open-sidebar toggle matches the close glyph at size-4 both
ways; SidebarToggle always-mounted with a lockstep animated slot — no more
one-frame title jump; title at text-3xl).
Semantic change folded in: `busy` in the chats list now means exactly "a
teach turn is in flight" — background jobs no longer spin the session row
(the sidebar's Background agents section is their channel).
Merge fallout fixed in the merge: components/graph-coverage.tsx (main's
coverage panel) held the last three text-xxs uses; with the rung deleted
they silently lost their size — mechanically swapped to text-xs, matching
lh-ui's own swap in graph-context-marker.tsx.
No file overlap with main's waves since fbeca36 (extraction coverage,
lh-graph merge, world-scale bake) — clean auto-merge. Gates: typecheck,
eslint, 496/496 tests, build. Live-verified on the dev server: 60px
header with 32px CMU title, label-only chip, snapped ladder active
(--lh-prose-size 16px, no --text-xxs rule anywhere), sidebar rows with
ISO dates, animated sidebar toggle with no header jump.
The world-scale pivot (2026-09-17): node/label/arrowhead sizes were
calibrated as "screen px at fit zoom", a unit whose meaning depended on
the graph's own extent — a 13-node graph at kFit 3.65 rendered 5.3x
smaller nodes than a 338-node graph at kFit 0.69, with no slider values
able to reproduce the dense feel. Sizes are now canonical WORLD units
(graph-canvas.ts drops the ÷kFit division in geometry/regions/hit-test);
kFit survives only as the label-gate 1x reference. The renderer's module
doc and the SIZE SEMANTICS block in render-settings.ts carry the story.
Camera becomes pure persisted view state: {k, x, y} stored to localStorage
(lh.graph.camera.v1, 250ms debounce, flush on unmount), restored on mount
via the new initialCamera constructor arg; the hardcoded fit() runs only
on first-ever open. A pin test locks the d3 zoomIdentity composition.
Region blur becomes world units too (the last span-coupled quantity): the
old bitmap-px reach scaled with graph span, so the 2026-09-14
search-filter collapse fix is now structural (blurW = tuning.blur; the
cellRef anchor is gone). Charge default bumps to -240 (single source:
CHARGE_STRENGTH in layout.ts; CHARGE_RANGE re-anchored so -240 sits
dead-center).
The 2026-09-17 bake — the user's live-tuned defaults (nodeSize 2,
edgeWidth 1, arrowLength 2, labelSize 3, labelZoom 4 / "8x", dilation 20,
blur 20, edgeAlpha 0.2, bitmap 256) — plus the ladder pivot: every
non-physics knob moves on a hand-authored 11-stop ladder (x1.5/x1.33
series, quarter/half steps at the low end), default dead-center at slider
position 50, step 10 = one stop per detent. The sanitizer snaps ANY
incoming value to the nearest stop (ties break low), so stored files can
never hold 2.18137...977-style artifacts; pre-pivot files (sizeUnits !=
"world") reset only the four resemanticized size fields. Physics sliders
keep the fine-grained log mapping.
The label cross-fade becomes PROPORTIONAL to the gate (span = gate x
LABELS_FADE_RATIO): the old absolute 0.5-ratio span read as a 100%
relative change at the old 0.5 gate but a 12% sliver at the tuned 8x —
"lower zoom levels naturally produce a longer transitionary period".
At the 8x default the node→cluster handoff now fades across one full
ladder stop (8x -> 12x). labelZoomDisplay is absolute (ratio / 0.5) so
the tuned gate keeps reading "8x" now that 4 IS the default.
Gates: 495/495 tests, tsc/eslint/build clean.
- Draw order: the cluster-label block moves from layer 5 to THE LAST
layer of draw() — labels can never be occluded by node labels or
hover names (user 2026-09-17; chose "topmost" after reassessing that
discs were never actually on top — the white-out over discs is
source-over compositing of the 85%-alpha text over near-white node
fill, an optical contrast flip, not a draw-order bug).
- clusterLabelSize: new persisted setting (default 14 = the old
REGION_LABEL_SIZE constant, so rendering is pixel-identical). Screen
px at ANY zoom (drawn at size/k — camera-compensated, UI-like, same
law as before). Log-slider range [7, 28] puts the default dead-center
per the house slider law. Sanitized: old persisted files (no field)
fall back to 14.
- Panel: the Cluster labels checkbox moves from the Labels section
into Regions (user: cluster settings live with regions), joined by
the new "Cluster label size" slider. Labels stay decoupled from the
Background colors compute switch, as designed on 2026-09-16.
The region bitmap rebuilds only every REGION_EVERY-th frame while the
live sim runs, so during drags the tints stepped at 1/3 the frame
rate. Between rebuilds draw() now linearly blends the previous final
bitmap into the current one ('lighter' + globalAlpha weights = a true
lerp of the premultiplied pixels — no source-over alpha stacking).
Because box blur and the end-alpha scale are linear operators, lerping
the finished bitmaps is equivalent to fading coverage before the blur,
the mathematically exact form of the user's preferred pipeline. Two
tiny GPU blits per frame; splat/blur/readback still only on rebuilds.
New RegionTuning.enabled is a compute switch (not a paint toggle):
off skips the entire build (splat, blur, readback, blit) — the settings
panel exposes it as "Background colors". Cluster labels stay decoupled
and follow clusterLabels. Old persisted files predate the field and
sanitize to enabled: true.
THE MESSAGE IS THE UNIT OF COVERAGE (vision-extraction-coverage.md): a
per-chat append-only ledger (coverage.jsonl) records which messages the
pipeline has extracted, by content fingerprint over exactly what the
extraction view sees (one shared line builder — core/coverage.ts); the
uncovered remainder is DERIVED at read time, never dual-tracked. A quiz
answer merging into the asking message re-opens only that message;
plan-block data-part attach never re-opens anything.
- The trigger (turn-service) replaces every-3-quiz-answers with
uncovered-tokens >= EXTRACTION.chunkTokens (4096) — quiz-less sessions
extract too, content size is the honest signal.
- runExtraction is now chunked: token-budgeted batches of the oldest
uncovered consecutive messages (a tool call + result never split), each
merged batch checkpointing its messages' coverage atomically — an abort
or error loses at most the in-flight batch; merged batches stay covered.
- Observation refs are batch-tagged (obs:<batchTag>:<sourceId>): re-runs
of the same batch dedupe; a re-opened message's fresh understanding
lands as a new event. Fixes the old obs:p1 position-collision.
- New API: GET /api/graph/coverage (exact per-session derived coverage +
running flags, most-uncovered first) and POST /api/graph/coverage/extract
(manual run for one session, 409 while in flight).
- Graph view chrome: the coverage button opens a panel with every
session's covered/uncovered tokens and a per-session Extract button —
the manual answer to aborted runs, errors, and dormant tails. Renders
even when the graph is empty: an empty graph with uncovered sessions is
exactly when it matters. Polls only while open.
- Notes imports keep the whole-file checkpoint path (coverage undefined).
- Tests: pure-core coverage suite (11), chunked-pipeline suite (3:
multi-batch + no-op, abort-keeps-earlier-batches + remainder re-run,
re-open alone + data-part invisibility), panel render suite (5);
pendingQuizAnswers and its tests deleted as dead code. 486/486.
Live-smoked on real GLM: manual extract of "teach me modal realism" ran
one batch, wrote 4 coverage entries, panel flipped to "1.3k covered ·
complete" with the button disabled, graph refetched (nodes 11), usage
persisted with real stats.
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.
The continuous 0-4096 dial was precision theater (a 128-token budget is
~2 propositions; nobody wants that granularity). Decided stops (owner,
2026-09-15): 0 / 64 / 256 / 1024 / 4096 / 16384 — the native range walks
positions 0-5 and maps through GRAPH_CONTEXT_TOKENS_STOPS in contracts.
Max rises to 16384 (route range follows); the server still accepts any
integer in range so a stale client cannot brick settings; non-stop
legacy values snap to the nearest stop for the thumb, readout stays
honest.
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.
Per-stroke alpha baked into every stroke/disc accumulated through
source-over compositing: one shape 0.35, an orphan disc over anything
1-0.65^2 ~ 0.58, hubs toward 1.0 — a lone node's glow read as a
darker double-counted spot on top of other regions. The pipeline is
now dilate at full opacity -> blur -> scale the whole field by
edgeAlpha once at the end. Painted at 1, the alpha channel is a pure
coverage indicator (overlaps of any count contribute exactly 1.0),
the blur smooths it into a coverage falloff, and the final scaling
gives a constant interior alpha everywhere covered, varying only
across the blur edges. The blur-off path also runs the readback now
(radius 0 = copy) so the end-alpha applies uniformly without blur.
Interaction change (user-flagged, accepted for now): different
clusters' overlapping halos no longer blend hues proportionally —
the later-painted cluster covers the earlier, ownership boundaries
softened only by the blur reach. Verified: old build spread 36k of
55k tinted pixels above alpha 100 (stacking); new build concentrates
at 70-90 (0.35*255=89 interior with falloff), and an orphan disc now
reads no brighter than the skeleton it sits on.
Non-integer blur radii broke the rendering (the log slider mapped to
fractional values), so the control is now integer-only by design:
REGION_BLUR_OPTIONS with snap-to-nearest sanitization for any input,
same law as the bitmap resolution. The old off zone (blur 0) is gone;
stored 0s snap back into the set.
All graph panel controls are now user-facing and server-persisted:
the old session-only dev number inputs are gone. GraphRenderSettings
gains nested regions/physics sections (backward compatible: old flat
files load with defaults), and RegionTuning/PhysicsTuning moved from
the renderer into core/render-settings.ts — one shape, one sanitizer,
one source for the store, the panel, and the renderer.
Physics/region clamps are sweep-validated safe ranges (each field at
both ends + all four corners settles cleanly on the real graph: no
NaN, no explosion — linkStrength caps at 4 since 8 explodes and 5
wobbles; friction floors at 0.1 since 0 never damps). Panel: log
sliders (default = geometric mean = position 50, step 5), off-zone
sliders for blur/cluster pull (position 0 = exactly 0), a fixed-set
dropdown for bitmap resolution, depth-anchor modes promoted to the
select.
Two follow-ups on the semantic-cluster wave, both user-reported:
1. Region coloring: edge-only dilation (the anti-double-counting rule
from the Louvain era) left 114 of 338 cluster members completely
uncolored — semantic clusters contain members that share no
intra-cluster edge with anyone (90 edge-orphans + isolated
micro-topics), and 20 of 45 clusters painted nothing at all. Fix:
a disc (radius = dilation, same as a round cap) for exactly the
members no intra-cluster edge stroke reached. Every member is
painted by its edges' caps OR one disc, never both — the
no-double-counting rule survives. Verified live: all 74
orphans-in-clusters sit on tinted region pixels.
2. Depth modes (ORPHAN RULE): a node with no edges has no dependency
structure, so depth-0 stamping put 90 unconnected propositions on
the vertical baseline / radial base ring — meaning they don't have.
Orphans are now exempt from the depth force and get the regular
view's unprojected forces instead (full center gravity + full
cluster pull). Structured-node projections unchanged (x-only
gravity in vertical, none in radial; tangential-only cluster pull
in radial). Gravity in depth modes is a custom force with EXACTLY
d3's forceX/forceY math (strength × alpha — the first draft
forgot the alpha and shifted the settle). Verified live: radial
orphans follow their topics instead of the base ring; vertical
orphans spread 1006 world units (6/90 near the baseline band vs
90/90 before).
Louvain only ever found the reasoning chains (the graph is disconnected
by nature — every connected component became its own cluster) and
edge-orphans could never join anything. Clusters are now RELATED TOPICS:
propositions grouped by meaning.
- Embedder (src/server/embeddings.ts): in-process transformers.js v4
running EmbeddingGemma-300M q8 (MTEB 68.4, ~300MB one-time download),
MRL-truncated to 512d, "title: none | text:" prefix on 100% of inputs,
L2-normalized so cosine = dot. Model/dim mismatch voids the ledger
(full re-embed). Injected at the composition root — the agent layer
never imports the server layer (architecture law).
- Ledger (data/graph/embeddings.json): model + dim + vectors + the
persisted cluster partition. Reclustering runs at the END of every
embedding batch (post-merge hook, both front doors) — never per read
(the agglomerative pass is ~1.5s at 338 nodes).
- Clustering (src/core/semantic-clusters.ts): hand-rolled average-linkage
agglomerative over cosine (npm HDBSCAN is an abandoned alpha). Stragglers
stay singletons by default; attachStragglers is the all-in toggle.
Threshold 0.60 swept on the real graph: 26 topics, 74/90 edge-orphans
land in real topics, 10 stragglers.
- Consumers (snapshot, settle, namer, replay) read the persisted
partition — one cluster concept everywhere; core-survival label
matching unchanged (20/26 topics kept labels).
- Louvain retired: detectCommunities + graphology* deps deleted
(graphology-louvain had leaked into the client bundle since the
clusters wave); scripts/bench-graph.ts (dead FA2 bench) removed.
- scripts/embed-graph.ts: backfill/cold-start (338 nodes in 150s
including the one-time model download).
The graph's follows/supports edges encode derivation (premise ->
conclusion); the layout can now express that rank spatially, as a
session-only playground toggle (nothing promoted yet):
- inferenceDepths: longest chain from a base node, CYCLE-SAFE by
SCC condensation (Tarjan, iterative) — the DAG property is typical,
not guaranteed; cycle members share their component's depth (mutual
support = equal footing). O(V+E), pure, deterministic.
- DepthAnchor force: per-node target scalar — vertical mode rows
(baseline at the anchor centroid, arrows up), radial mode rings
(foundations center, outer boundary = edge of knowledge). The rank
is computed, not simulated: pairwise up-forces can't give baselines.
- Mode projections: gravity x-only in vertical / dropped in radial;
cluster pull x-only in vertical / tangential-only in radial; the
anchor re-blend preserves user arrangement in the FREE dimension
only (x / angle) — meaning owns the constrained axis, memory owns
the free one.
- setPhysicsTuning reheats (alpha 1): toggling a mode re-forms the
graph live and commits the new rest state; drag release is springy
along the constrained axis (node returns toward its level/ring).
- Sweep on the real graph: strength 0.8 = 215/216 arrows correct with
organic sag; 1.5 = crisp rows. 174 base nodes, max chain depth 7,
zero cycles in current data. replay-layout --depth=<mode> bakes a
mode into the ledger for rest-state experiments.
The 'virtual node' realized as a centroid force: two allocation-free
O(n) passes per tick (membership resolved once at sim creation;
Float64Array sums). Measured indistinguishable from force-off at 338
nodes (271ms vs 272ms per 400-tick settle).
- createGraphSimulation: SimOptions.clusterPull + linkStrength +
velocityDecay (physics controls plumbing); degree-scaled link
stiffness so linkStrength=1 is exactly d3's default
- settleGraphLayout split into layout-settle.ts (it computes the
Louvain partition for the pull); layout.ts stays browser-barrel-pure
- barrel no longer re-exports clusters.ts: graphology + louvain are
fully out of the client bundle (verified against built chunks)
- replay passes the per-batch partition; CLUSTER_PULL_STRENGTH=0.15
tuned by sweep: cluster spread 63->36 world units, centroid
separation and total span unchanged (no blob merge, no collapse)
One engine owns the equilibrium: the merge settle and (next commit)
the live drag both run createGraphSimulation — d3-force link springs,
1/d charge, weak centroid gravity — so a drag's rest state IS the
server's rest state. The FA2-era anchor mechanism survives as a
post-tick POSITION-SPACE re-blend (retention 0.85): d3's forceLink is
a Hooke spring (strength*(d-L)*alpha), which overpowers any force-space
anchor at long range — the re-blend is alpha-independent and always
wins as the sim anneals (the 675-unit drift bug). Gravity retuned for
d3's 1/d many-body falloff (bootstrap 0.2 / incremental 0.12: cluster
spacing lands ~400 world units, the FA2-era density). Rim orphans seed
(and anchor) at 1.7x the CONNECTED core's mean radius — a max-span-based
rim fed back on itself and ran the layout away to 6000 units. Replay:
338 positions in 1.6s, old-node shift within a batch 4.8 (FA2: 23).
The layout now EVOLVES instead of reshuffling on every extraction. A pure
core module (src/core/layout.ts) warm-starts forceAtlas2 from the positions
ledger (data/graph/positions.json, read-modify-write like sources.json,
updatedAt stamp for future live-physics conflict handling), re-blends every
previously-positioned node toward its anchor each iteration (retention 0.85
- old placements stay recognizable, strong forces still win), seeds new
nodes near their positioned neighbors, and seeds orphans on the rim outside
the core so they read as unconnected knowledge.
- Port/store: GraphStore.readPositions/writePositions; snapshot carries
positions when the ledger exists (optional - fresh graphs unchanged).
- Extraction pipeline settles synchronously after every merge (chat and
notes paths), before usage persist - the next /api/graph fetch already
reflects new nodes.
- View: the ledger IS the layout; the local one-shot FA2 pass survives
only as a fallback when no ledger exists.
- scripts/replay-layout.ts: simulates organic growth by replaying the
current 338-node graph one imported file at a time (real import order,
edges activate when both endpoints exist). Real run: 338 positions in
0.3s, 90-degree-0 orphans collecting at mean r=338 vs connected r=194.
- Also restores the cluster region fills (hull background colors) that
6692b19 had removed on a misreading of "no glow".
Live-verified: fit zoom derives from the ledger bbox, pickAt aligns with
ledger positions at every probe, hover/click/detail-card work, 408/408
tests, typecheck/lint/build green.
Future live physics stays cheap by design: the same pure function is the
merge-time settler and the future client simulator; a pinned node is just
an anchor with full retention.
- LABEL_GAP 6 -> 2 (user: labels sit way too far below their nodes).
- Hover glow is now PROPORTIONAL to the disc — outer radius = node radius
x 1.7 (GLOW_SCALE), matching the blur-a-copy-of-the-disc mental model,
instead of the old constant +6px pad that read huge on small nodes and
thin on big ones. GLOW_ALPHA 0.45 -> 0.3.
- Hover-OUT fade was broken by construction: draw() gated the glow on
hoverId, which nulls instantly on leave, so the 180ms fade-out ran
invisibly and the glow popped off in one frame. New hoverNode field
outlives hoverId through the outgoing fade (cleared when the fade
completes); verified live: 8 distinct frames over ~200ms then stable.
- Sliders rescaled per spec: ALL five are now log-scaled 0-100 with the
default dead-center at 50 (ranges chosen as default/4..default*4:
node/arrow/label size 0.5-8, edge width 0.25-4, labelZoom 0.0625-4),
coarse step 5 (21 stops; labels-until-zoom was "way too granular" at
201). Generic toSliderPosition/fromSliderPosition replace the
labelZoom-only pair; sanitize clamps follow the same range constants;
size sliders display compact values ("2", "1.6", "0.75").
EOF
- Defaults per user: nodeSize 2, edgeWidth 1, arrowLength 2, labelSize 2
(labelZoom 0.5 stays — reads as the "1x" they confirmed good).
- Label size slider floor 3 -> 1 so the default 2 is not the smallest
value; sanitize clamp widened to match.
- Cluster hull glow (layered low-alpha fills) removed for now: it read as
jagged up close (straight-line hull approximation at high zoom) and
does not belong until the layout settles. The solid viewer-relative
border and the inverse-gated region labels stay. HULL_FILL_LAYERS
constant deleted; hullShapes cache carries only the stroke path now.
- Saved render-settings.json reset so the new defaults apply on next load.
EOF
The defaults were calibrated in the screen-fixed era; world-scaling renders
sizes at setting x k/kFit, so at the default label gate (2x fit) everything
read ~2x too big ("labels way too large even at the minimum"). Fit-zoom
normalizes layout scale away, so the fix belongs in sizes, not layout:
spreading nodes further apart in world units changes nothing at fit.
- Defaults: nodeSize 6 -> 4, arrowLength 5 -> 4, labelSize 12 -> 7.
- Clamps/slider ranges widened at the low end: nodeSize 1-14 (step 0.5),
edgeWidth 0.25-4 (step 0.25), arrowLength 1-15 (step 0.5),
labelSize 3-24 (step 0.5).
- GLOW_PAD 9 -> 6 (halo dominated small discs), REGION_LABEL_SIZE 18 -> 14.
- Tests updated for the new clamp bounds plus a low-end pass-through case
(403/403). The saved render-settings.json still shadows defaults by
design; the local copy was reset so the new defaults showed immediately.
Fixes the zoom frame-rate collapse and rebuilds draw() around a single
camera transform, per the user's Obsidian-style sizing proposal.
Bug fix first: the eased zoom leaked a queued rAF callback per frame --
advanceZoom applies each step through d3-zoom, whose zoom event called
markDirty() from inside the tick loop, scheduling an extra draw while the
loop scheduled its own. Draws multiplied per displayed frame; sustained
wheel ran at ~20fps against a 120Hz display. markDirty now skips
scheduling while inTick (verified: 62 frames/500ms during a 28x zoom
burst, matching pan).
World-space model: every draw call now runs in graph coordinates under
one setTransform(k). Sizes are defined as screen-px-at-fit and converted
once per geometry build (/ kFit), so nodes, node labels, arrowheads and
the hover halo scale with the graph as you zoom (0.25x painted area at
half zoom, exactly k^2) while edge widths and hull borders stay
viewer-relative hairlines (/ k). Cluster hull geometry (convex hull +
three offset fills + solid boundary -- dashes dropped for now) is built
as cached Path2Ds per data/visibility/settings change; camera moves do
zero geometry math, making zoom per-frame cost identical to pan. Label
wrapping runs once per build, not per frame. Region labels stay
viewer-relative (18px at any zoom) as UI chrome. Picking radius scales
with the world-sized discs.
Glide feel: ZOOM_EASE_MS 120 -> 45, convergence 0.1% -> 0.2%; a notch now
settles in ~260ms instead of ~750ms. Resize still paints synchronously.
Also fixes barrel-law lint violations from the settings wave (import
from @/src/core and @/src/server, not deep module paths).
Five refinements to the hand-rolled renderer, all user-directed:
1. Settings are SERVER-persisted (data/render-settings.json via new
GET/PUT /api/graph/render-settings) - one source of truth for every
browser, replacing localStorage. The client loads once after mount,
saves debounced on user changes, and never lets a slow GET clobber a
faster user (or mounting PUT defaults over the server copy). The
shape/defaults/sanitizer moved to src/core/render-settings.ts - pure,
shared by client, server store, and tests; every disk/network value
passes through sanitizeRenderSettings.
2. New "Label size" control (default 12px - the previous hardcoded size).
Font, wrap width, and line height all scale; verified live: +76k px
with the label gate open.
3. "Labels until zoom" slider reworked: log-scaled over 0.0625-4 so the
useful band (0.25-0.5 camera-ratio units) sits at 33%/50% of the
track instead of crammed at the far left of the old linear 0.25-4
slider. Display units are normalized: the default 0.5 reads as "1x",
0.25 as "0.5x". The underlying labelZoom semantics are unchanged,
so existing values survive.
4. Smooth wheel zoom: d3-zoom instant per-notch jumps (x1.2, bang) are
replaced by an eased glide - wheel events accumulate a cursor-anchored
target scale and each tick eases the camera toward it (exponential,
120ms constant, frame-rate independent). Pan/pinch stay d3-owned; a
pan cancels a glide in progress. Verified live: 8-frame glide.
5. Resize blanking fixed (graph vanished for the duration of a window
drag, reappearing ~300ms after). Cause: every resize event
reallocated the canvas backing store - which CLEARS it - and the
redraw was deferred to rAF, which live resizes starve via relayout,
so the canvas sat wiped for the whole drag. Fix: the ResizeObserver
callback now redraws SYNCHRONOUSLY after reallocating (verified via
a probe RO firing in the same delivery batch: canvas fully painted).
kFit (the label gate 1x reference) is also recomputed on resize from
the now-cached world bbox - the camera itself still persists.
tests/render-settings.test.ts: 10 tests pinning the sanitizer (garbage/
partial/clamps, legacy no-labelSize records) and the slider mapping
(default dead-center at 500, monotonic, exact round-trips, display
units). 402/402 green; typecheck, lint, build pass.
Core-survival labelings (vision section 1, DECIDED 2026-09-13):
ClusterLabeling records in a new append-only ledger
(data/graph/cluster-labels.jsonl), immutable, id = clu_ + sha256 of the
sorted core so re-naming races and re-runs dedupe at the store seam.
matchClusterLabelings (core/clusters.ts, pure): a community keeps a
record's label while at least half its core survives in it; both sides
single-successor (a split keeps one name; greedy by overlap, ties to
the newer record).
The namer (src/agent/subagents/cluster-namer.ts): ONE singular cheap LLM
call (new 'labeler' role, glm-flash/low) names every new/changed
community at once, given member names + statements (capped at 12
members per cluster). refreshClusterLabels runs POST-EXTRACTION
(user-decided) from both front doors, fire-and-forget — naming failure
never fails the run; refreshes serialize behind a chain so parallel
imports never mint duplicate labels. import-notes.ts awaits
clusterNamingSettled before exit.
Payload: GraphSnapshot.clusters (>= MIN_CLUSTER_SIZE 3) joins detection
with labelings at read time — claimed clusters carry the record's label
and id; unlabeled ones carry a member-hash id and null label. Detection
is deterministic so no memo in v1 (ms at current scale; revisit at
vault scale).
Tests (372/372): matcher half-rule/split/merge/greedy/tie cases; naming
end-to-end through importNotes (one call, one record, claim on
re-refresh), naming failure non-fatal; snapshot derivation with
labeling join; 11 test pipelines now pin an explicit labeler seam.
detectCommunities(nodes, edges) in src/core/clusters.ts (new dep
graphology-communities-louvain): Louvain modularity over the undirected,
unweighted follows+supports structure (user-decided 2026-09-13). The
random-walk traversal runs on an inlined seeded mulberry32 so the same
graph always yields the same partition; community indices are still
unstable across edits, which is exactly why labelings (phase B) key to
sorted MEMBER SETS. follows+supports between the same pair collapse to
one edge; self-loops and dangling edges are skipped defensively; no
min-size filter here (the >=3 render threshold is a presentation rule).
Tests: two-cliques-plus-bridge split, determinism, singleton/empty
graphs, pair collapse, defensive skips, member sorting. 363/363.
Real-data smoke over the current notes graph: 338 nodes / 211 edges ->
138 communities, modularity 0.95, and the big ones are exactly the
learn-session topics (databases, differential forms, sales, servers,
vim, copulas...). 32 communities clear the >=3 render threshold.
The same extractor→auditor→merger pipeline now runs over raw personal
notes in arbitrary formats (vision §1, DECIDED 2026-09-13):
- NotesSource records (data/graph/sources.json, deterministic id from
path hash → idempotent re-imports) play the chatId role in mappings,
checkpoints, and observation events; usage lands on the source record
(appendSourceUsage, idempotent per startedAt, kept across upserts).
- mergeFindings is chat-store-free: callers pass MergeSource
{messages, foldedAt} — sessions pass the chat record, notes pass
messages:null (no quiz-event rebuild; a deterministic rebuild needs
structured quiz tool parts).
- Second extractor prompt (NOTES_EXTRACTOR_INSTRUCTIONS) for general-
purpose extraction; understanding judges the AUTHOR's grasp. Auditor
wording input-agnostic. extract/audit share one generic runSubmitAgent.
- graphSnapshot.fileSources (srcId→filePath); notes sources render as
static path entries in the detail card (FileTextIcon, never a link).
- One shared pipeline instance (exported from chat-handler): same
single-flight registry, live status chips (label = file path), usage
accounting, and true abort for both front doors.
- Transport: POST /api/graph/import (multipart, 3-wide worker pool) +
Import notes… button in the graph view; scripts/import-notes.ts for
batch runs (concurrency 4).
- real-data-extraction/ gitignored (raw import sources never committed).
355/355 tests; typecheck, lint, build green.
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.
The architectural answer to the validation-crash question: validation IS
first at the model-facing boundary (pi validates arguments before
execute; failures return to the model same-turn — all three live
incidents retried successfully). But the same bytes also exist UNVALIDATED
by design in two other places: the live wire streams raw tool-input
deltas DURING generation, and the transcript keeps the raw input of
REJECTED calls (honesty + the replay invariant). The consumers were
supposed to trust part STATES — but that trust was enforced by
TypeScript casts ('part.input as X'), which are compile-time lies with
no runtime check behind them. That is the hole both crashes came
through, and it should not have been possible.
The structural rule now: model-authored INPUT is consumed only through
parsers running the SAME zod schema the model faced — boundary-valid ⇔
consumer-parseable, mechanically. Server/learner-authored OUTPUT stays
trusted (our code authors it).
- src/core/tool-parts.ts (new): parseQuizInput (display parse over the
shared quizInputBase — strips the teacher-only nodeId, accepts both
prober and teacher boundary shapes), parseAskInput, parseProbeDoneInput.
- contracts/probe.ts: probeDoneInputSchema is now core-owned zod — the
probe_done tool declares it (single source of truth), the seam runs
the literal same schema.
- question-dock.tsx: the dock parses its input at entry; unparsable
renders the quiet malformed-question fallback, never a crash.
- probe-done-card.tsx: the input preview goes through the parse; the
old state reasoning and shape guards collapse into it.
Pinned in tests/tool-parts.test.ts: both exact live crash shapes
(stringified quiz options, stringified assessment), boundary-parity
cases for all three schemas, nodeId stripping, mid-stream partials;
render-cards pins the dock fallback markup.
Gates: tsc, lint, build, 328/328.
Two live bugs from the modal-realism stress runs:
1. Multi-select ask: picking the custom 'Other…' row cleared every
already-picked option, and picking any option cleared a typed custom
answer — single-select displacement logic applied to both modes. The
dock's selection is now a pure state machine in question-core
(toggleOption/toggleOther over DockSelection, unit-tested): single
displaces (unchanged), multi coexists — neither channel clears the
other; the typed text always survives.
2. ProbeDoneCard crashed on a REJECTED probe_done call: the model
stringified 'assessment' (the same double-encoding habit as the
options bug), zod rejected the call, the model retried — but the
card's input-preview fallback trusted the errored part's unvalidated
input shape and died on assessment.strands.map. The preview now reads
input only while the call is pending or streaming (guarded with
Array.isArray); an output-error part renders a quiet
malformed-and-retried note instead of the garbage.
Pinned: question-core.test.ts (3 dock-selection machine cases),
render-cards.test.tsx (4 probe-done card cases incl. the exact live
stringified shape, plus the multi-select ask markup pin).
Gates: tsc, lint, build, 315/315.
Live observation (modal-realism run): the teacher batched two quizzes in
one turn; the dock showed the SECOND question first, so the answering
order was reversed vs the saved record and the transcript cards.
Cause: findPending scanned parts BACKWARD within a message, so with two
input-available quizzes it picked the last one first. No overwriting was
involved — parts are created in call order on both wire and record; the
scan direction alone reversed the interaction order. (pi executes a
parallel batch with starts + result messages in call order; the
execution-end order is completion-racy, which also allowed a millisecond
dock-subject flip mid-turn.)
Fix A — src/core/pending.ts: parts now scan FORWARD within a message
(messages stay newest-first): the FIRST unanswered question is the dock
subject; the second appears only after the first is answered.
Fix B — src/agent/lh-tool.ts: question tools opt into sequential
execution (pi runs the batch in call order when any called tool opts
in), so tool-input-available chunks always arrive in call order and the
dock subject can never flip mid-turn. Question executes are instant.
Pinned in chat-flow.test.ts: a parallel two-quiz batch asserts wire
order (A's available chunk before B's), part persistence order, and the
dock selection sequence (A first, then B after A is answered).
Gates: tsc, lint, build, 307/307.
The Vercel AI SDK survives only as Band D (useChat client + UIMessage
record types); @openrouter/ai-sdk-provider is uninstalled.
- src/agent/models.ts: PiEngine {model, streamFn} via createModels +
openrouterProvider subpath; thinkingLevelFor maps config effort none->off
- src/agent/lh-tool.ts: neutral LhTool (zod authoring, optional execute,
clientQuestion flag); toPiTool converts via z.toJSONSchema and re-runs
zod inside execute so refinements ride the error-result retry path
- src/server/pi/record-adapter.ts: UIMessage records -> pi Messages
(dangling tool calls get synthesized interrupted results)
- src/server/pi/ui-stream.ts: SSE bridge emitting AI-SDK-shaped chunks so
the useChat client parses unchanged (continuation id rule, question
tools never emit output, tool-input-error for bad authoring)
- teacher-agent.ts: stateless per-turn pi Agents; shouldStopAfterTurn
stops only when every question-tool call resolved cleanly (errors retry)
- chat-handler.ts: buildAgent(outgoing) factory seam (stateless doctrine);
assertValidUiMessages owns the retired SDK validation gate
- sub-agents via shared runSubagent (pi-run.ts); extraction stages stop
after a clean submit; visual-maker keeps its hand-rolled loop over
direct streamSimple calls; aborted model calls bill no usage
- tests: all model seams ported to pi-ai faux provider (scripted
FauxResponseSteps, agent factories); handleChatRequest second param
is now an agent factory
Gates: tsc, lint, build, 305/305. Live GLM smoke (disposable driver,
deleted): prober streaming + persistence, question seam, turn-2
continuation same message id, per-turn real token usage - ALL PASS.
Data wiped at the pivot (chats/graph/jobs, .gitkeeps kept).
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.
Emission runs at persistTurn — after the whole turn is composed — so
appending the block buried it under the same turn's teaching prose and
quiz calls (user-observed in chat_mtwgb9lj: the DAG appeared only after
the first check). Splice it in immediately after the (last) successful
plan_init part instead. Two positional tests added.
Diagnosis (2026-09-11 user testing, minimal-effort GLM): sessions stuck on
'plan' forever. Two causes: (1) the three-phase view read the stored tree's
currentNodeId, which nothing ever writes post author-once (2026-09-03) — a
structural dead seam; (2) the teacher stochastically left quiz.nodeId
untagged, and done-by-quiz-result semantics would have falsely completed
nodes anyway.
The redesign (docs/probe-teach-split.md) abolishes derived occupancy in
favor of explicit events:
- Two agents over one transcript: createProberAgent (quiz without a
nodeId field, ask, probe_done) and createTeacherAgent (full toolset,
plan_init onward). The chat-handler construction branch keys on a
probe_done part in the transcript. MODELS.prober added.
- probe_done: server-side seam tool — carries the learner assessment
(strands + free-form edge read), refuses while a question is open or
the probe is already concluded, and queues a handoff steer through the
generalized wake pipeline (steers now dedupe by id, not jobId; job
finalization only for job steers). The teacher auto-wakes and authors
the plan; the learner sees the assessment via the probe-done card.
- phases: probe ends at a concluded probe_done, plan ends at a
successful plan_init — strictly sequential, no skipping, no legacy
compatibility (data wiped at the pivot per doctrine).
- Plan map = position, not completion: derivePlanStatuses reads only
tagged quiz CALLS (state/outcome feed nothing); current = latest
tagged node with consecutive duplicates collapsed, visited = earlier
distinct stays (a revisit stays in both — current wins at render).
PlanTree is structure-only; currentNodeId/doneNodeIds and the dead
mutation helpers are gone. No fallback current.
- Prompts split per agent (prompts/prober/, prompts/teacher/ over the
trimmed shared base); graph-context injection removed entirely
(shelved, docs/probe-teach-split.md §5); knowledge-graph context no
longer reaches either agent.
- Plan blocks are full-only snapshots at plan_init (neighborhood
machinery retired; the persistent DAG window is the future UI wave);
usage capture unchanged.
Gates: tsc clean, 294/294 tests (phases/plan-tree/plan-blocks/question-
tools rewritten to the new semantics; probe-done + handoff-wake tests
added), lint, build. Live GLM smoke PASS: prober concluded with an honest
assessment, teacher auto-woke, authored a 10-node plan with the block
attached, then tagged checks on slope-instinct → secant derived current/
visited exactly as designed; grading honest both ways (driver off-by-one
proved a wrong answer grades false).
Live smoke exposed two bugs in the turn usage log:
- onFinish messages carry no ids (withMessageIds fills copies inside
persistTurn), so fresh turns keyed messageId "" — provenance lost.
- The SDK coalesces continuation turns into ONE response message (a whole
probe conversation lands in a single assistant message), so per-messageId
idempotency swallowed every continuation turn's usage.
TurnUsage gains id: string (unique per request, the append idempotency
key); messageId stays as transcript provenance, taken from the id-filled
list. appendTurnUsage dedupes by id; continuation turns with the same
messageId each keep their usage.
- src/core/plan-blocks.ts: pure emission builder — full block on successful
plan_init, neighborhood block when the derived plan state moved since the
newest existing block; the as-of state reads from the block's frozen
snapshot (transcript quiz parts mutate in place on answer, so
re-derivation is post-mutation blind). Idempotent by construction.
- turn-service persistTurn enriches with withPlanBlocks at the canonical
write; chat.tsx adopts the record once the server agrees the turn is
done (endTurn flips only after persist).
- chat-handler: per-request usage collector via agent-level onStepFinish
(StepResult.usage), appended at onFinish to ChatRecord.turns via
appendTurnUsage, idempotent per response message id.
- components/plan-block.tsx: hand-built SVG (user decision) — deterministic
layered layout (Kahn topo + longest-path depth, rows centered, authoring
tie-break), bezier edges, the retired panel's status color language.
- PlanBlock renders data-plan-block parts AS CARRIED (frozen snapshots,
never re-derived): kind full = borderless, viewBox = layout extent so
nodes hug the column edges, natural height; kind neighborhood = own
small labeled 'Plan fragment' container with the current node glowing
(pulsing primary halo).
- PlanPopup: full-screen overlay with the LIVE plan — latest tree from the
transcript, statuses derived at click time via derivePlanStatuses
(contracts/plan.ts). Esc / backdrop / X close; drag pan, wheel
zoom-to-cursor, zoom buttons; fit-floor semantics harvested from the
retired tree panel (zoom floor = the fit, pan locks at it).
- chat.tsx: data-plan-block case in the part switch + popup mount.
- src/core/ui-message.ts: TeacherUIDataParts gains "plan-block": PlanBlockData
(additive, type-only — lane touch flagged to master).
- tests/render-plan-block.test.tsx: pure layout (layering, centering,
dangling-edge tolerance), wrap, frozen-snapshot render, live derivation.
Extraction is real spend and background work; the dock needs to see
and stop it. User decisions: persist all cost data (chat record),
per-stage usage granularity, and TRUE cancellation for abort.
USAGE CAPTURE (live + persisted):
- extraction.ts: per-call onStepFinish captures each billed step's
usage (SDK LanguageModelUsage -> CallUsage: input/output/cache
read+write/reasoning) into the registry's live per-stage usage
(extractor/auditor; the merger is deterministic — no model call).
- /api/graph/status runs now carry usage (model + steps per stage);
cost/context folds are the consumer's (core/prices.ts — nothing
stores dollars).
- ChatRecord.extractions: append-only per-run usage log (mirrors the
turns log): startedAt key, at, durationMs, merge stats when the
merger ran, stages. Persisted on success AND failure — a failed run
still spent money. Idempotent per startedAt; the list payload
(ChatSummary) never leaks it.
- EmissionStats moved to contracts/graph.ts (its contract home);
core/graph.ts re-exports for existing importers.
- Verified end-to-end against the real pipeline (backfill one chat):
per-step token counts, reasoning tokens, and the auditor's growing
prompt-cache reads captured and persisted.
TRUE ABORT (the dock's abort button, user decision):
- RunningExtraction owns one AbortController per run, shared across
phases; abortExtraction(chatId) fires it and returns whether a run
was aborted (idempotent no-op otherwise). The pending model call
receives the signal — the provider cancels the request, spend
stops. throwIfAborted() at phase boundaries ends runs aborted in
sync gaps.
- Abort is a deliberate stop, not a failure: same ledger discipline
(nothing merged, nothing checkpointed, partial usage persisted) but
the run reports zero emissions instead of throwing. Exported via
src/server/observability.ts for master's /api/activity/abort route.
- Tests: mid-run abort (hanging mock that rejects on the signal),
live-accumulation + persisted-record assertions, store-seam
idempotency, and the summary-payload leak guard.
Wave 4 (agent-side observability + control), reconciled with the dock
contract (master relay 2026-09):
- abortJob in src/server/jobs.ts, exported through the server barrel for
POST /api/activity/abort (master's thin route): fires the job's
AbortController (real kill — the in-flight model call aborts, not just
the next loop check), drops any queued result steer, flips the record
to error "Aborted by learner" + activity "aborted" synchronously, and
queues a one-line abort steer (buildAbortSteer) through the same wake
pipeline so the teacher's promise doesn't hang in the lesson. Shares
the result steer's deterministic id space, so abort + racing result
can never both deliver (steerDelivered dedupe).
- Abort registry (src/agent/subagents/aborts.ts) on globalThis, mirroring
the wake pipeline's module-state pattern; runner registers/unregisters
per run, so the route's module instance sees the loop's controller.
- Runner usage capture: LoopContext (abortSignal + onUsage) threads into
all three kind loops; per-step usage folds via accumulateJobUsage into
JobRecord.usage (model from config MODELS, absent-never-zero), flushed
in runJob's finally so an aborted job's money record survives the kill.
- Contracts: JobUsage token fields optional + lastStep (most recent
call's accounting, for the dock's per-blob context line vs
CONTEXT_LIMITS); SubagentResultMeta.aborted; AbortSteerMessage;
buildAbortSteer; SteerMessage union generalized.
- demo-builder/visual-maker accept abortSignal + onUsage (visual folds
generateText usage per iteration).
- stillRunning/finishJob/jobContext guards: an aborted job is never
overwritten by the runner (terminal no matter what the wake does);
failJob and the pump's finalizeJob keep the flip.
- UI: abort steer renders a quiet aborted chip (job-aborted-badge).
- Prompt 20-jobs.md: one line about aborted jobs.
Tests: accumulateJobUsage folds (undefined-preservation, lastStep),
runner usage capture, real-kill abort (sync flip, silent unwind,
abort steer wakes), usage-so-far survives the kill, aborting a
delivering job drops the queued result steer, outcome reasons,
abort-chip render. 330/330, typecheck/lint/build green.
The 2026-09 observability wave's type-level foundation, landed before the
streams so nothing renders against guessed shapes:
- contracts/chat.ts: CallUsage + TurnUsage (per-model-call token accounting,
one entry per teacher/wake turn keyed by messageId) and ChatRecord.turns —
a DELIBERATELY parallel append-only usage log, not per-message metadata:
the client re-POSTs its message list every turn and its copies never
carry server-stamped fields, so anything inside messages gets clobbered
by the next persist (the wave-3 merge-rule lesson).
- contracts/chat.ts: PlanBlockData — the persisted data-plan-block part
(full DAG on plan_init, one-level neighborhood on plan updates, snapshot
semantics so history renders what was true then).
- contracts/jobs.ts: JobUsage + JobRecord.usage — a sub-agent run's
accumulated accounting, written by the runner.
- core/ports.ts + store/chats.ts: ChatStore.appendTurnUsage — idempotent
per messageId (abort replays / wake retries append once); listChats no
longer leaks turns into the list payload.
- core/prices.ts: PRICES + CONTEXT_LIMITS (generated from OpenRouter's
published pricing via scripts/sync-pricing.ts) + pure folds: callCost,
turnCost, sessionCost, contextWindow, jobCost. Unknown model or absent
fields render as absent, never zero.
- server/observability.ts: barrel-approved surface for the transport layer
(fixes the import-law violation in the graph status route that slipped
through the wave-3 merge verification).
Capture wiring (agent onStepFinish -> persist seam, plan-block emission)
lands next as master seam work; the streams start against these shapes.
amosblomqvist