`Tile` takes an optional link: Nodes and Flows running to /flows, Runs
running to /runs, Failures down to the list further down the same page.
Queue backlog and loop lag have no screen of their own and stay plain.
`Logo`'s responsive variant carried a second `<img>` for the collapsed
rail, but AppSidebar hides the whole logo when collapsed, so an ancestor
`display: none` meant it never rendered in any state.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KYM38KSb4V4v2T71eifnZv
**nginx served the bundle uncompressed and uncacheable.** The base image
ships gzip commented out and nothing set `Cache-Control`, so every load
carried the whole thing and every reload cost a 304 per asset. Measured on
the built image: the entry chunk 838 kB → 311 kB, the Monaco chunk 2.66 MB
→ 832 kB, and the ~50 content-hashed assets are now immutable for a year.
`index.html` is explicitly `no-cache`, since it is what names the rest.
**A widget that throws no longer blanks the screen.** There was one error
boundary in the app, on the root route, so anything that threw replaced
everything including the navigation — on `/view/{name}`, an unattended wall
panel with no way back. Each tile has its own boundary now, and the app
shell has one inside it so a screen that fails leaves the sidebar standing.
`react-error-boundary` was already a dependency and imported nowhere.
**`localStorage` cannot take the app down.** Reaching it raises where the
browser blocks site data, and `setItem` raises once the origin's quota is
full — which the flow editor's node clipboard, carrying whole Python
sources, can genuinely reach. Thrown from a key handler that escaped to
`window.onerror`, which the single root boundary then turned into a blank
page. `lib/safeStorage.ts` is the guarded pair the pre-paint theme script in
`index.html` was already using; a copy too large to store now says so.
Queries default to `staleTime: 5000` — below every poll interval on any
screen, so nothing polls less often than it did, but a route mounting twice
in a few seconds stops refetching everything it touches. Window-focus
refetching is off: the socket pushes what changes and a reconnect
invalidates what it feeds, so a focus event has nothing of its own to say.
Home alone reads about ten queries on every one of those.
Render cost, two that showed up in the audit:
- `LogsPanel` was rendered unconditionally by the dock and decided inside
itself whether to draw, so with the panel *shut* it still subscribed to
the log store and re-filtered five hundred lines per line a flow
published. It returns before any of that now.
- `HealthActivity` subscribed to the whole engine-event array and used one
number from it, so a flapping node re-rendered the component that draws
Home's two uPlot charts — each of which rebuilds its series on every
render by design. It subscribes to that number.
- the global search bucketed the index nine times per keystroke, once per
group. One pass, and each group offers at most twenty.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M6hPWS6YEbT1P8LxhhFb2T
Runs: a run can now be deleted (DELETE /runs/{id}, cancelling a live one
first), exported as csv from the screen's own filters, and its flow label
opens the flow. Its "Parameters" panel became "Inputs" and lists every
input the flow declares, marking the ones that took the flow's own value
rather than the run's — the comparison table resolves the same defaults
instead of printing "unset".
Home reads brain, dashboards, health, flows: the mosaic is one full-width
scrolling strip, and the flows list and the flow-activity rollups merged
into a single left-joined table so a flow's state and its numbers sit on
one row.
Charts take a drag to narrow the x window and a double click or tap to
come back out. UplotChart holds the scale and passes resetScales:false
while a window is held, which is what the old comment said made this
impossible.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019LrWVRguqbk33YzfEeUx5W
The tile only exists on an installation that has run something, which is the
same "nothing has been run yet" the Runs screen goes by. The count comes from
/runs/overview, already grouped by status in the database and already polled by
the Runs screen, so Home shares its cache entry rather than paging rows or
growing a second endpoint.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01K1moruzue2kTJd3uVisgNk
Failures were slot 1 of a ramp that carries identity by lightness alone, so at
a 2px stroke the line that matters read as one picture with avg and lag. It
takes --brand-secondary now — the same terracotta the brain graph above it
already spends on a flow that cannot run. A palette entry naming a token is
resolved off the chart's element, since a canvas cannot parse var().
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The offline pill was drawn inside a `flex h-0` container with align-items at
its default stretch, so the pill was stretched to a zero-height box while still
carrying its padding — the "box too small" it was reported as. It is a
persistent notification now, keyed on one id so a poll that keeps confirming
"still offline" replaces the card instead of restacking it.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The newest metric bucket is upserted every flush while its minute runs, and
both rollup endpoints summed it in — so every curve on Home ended on a fall
that was only the clock. The timeseries now stops at the last closed bin, and
the flow rollups' window ends on the last closed minute, so all sixty slices
are whole ones.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
`RedisWorkQueue.stats` read XPENDING, which counts entries delivered to a
consumer and not yet acknowledged — work in progress. Entries sitting in the
stream undelivered were counted nowhere, so an engine hours behind reported
itself idle: on the house, `pending: 4` while the group's lag was 1554.
The group's own `lag` is the missing number. `backlog` now carries it on both
queues (`len(_items)` in memory), leads the health tile, and a sustained one
publishes `engine_degraded` from the timer thread — named with the flow most
of the waiting work belongs to, sampled from the undelivered tail, since that
is the actionable half. It is a summary problem rather than a /utils/health
503: a backlog should not restart the container.
Also drops the keyspace `scan_iter` `stats()` did per poll to count parked
items — it walked every state and idempotency key twice per ten seconds — for
a set the park/unpark path maintains.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BpfSinyCBfjuieikyfMPbf
/observability/timeseries and /flows read every metric_minute row in the window
and folded them in Python, so the 7d preset pulled a week of rows on each 30 s
poll. date_bin() does the binning now — the row count drops to the slices asked
for, and to flows × 60 for the sparklines. A window of zero hours used to divide
by nothing and answer 500; windows are clamped to an hour at the low end and to
the retention period at the high end, past which there is nothing to find.
/observability/runs returns {data, count} rather than a bare list, so a minute
busier than the 200-row cap says so instead of quietly showing its newest 200.
The count is only queried when the page comes back full, which keeps the poll
from handing back what the fold just saved.
failures_24h leaves the summary — the Home tile counts errors over the selected
window from the rollups, and nothing had read the field since.
Deleting a flow now takes its Run rows and their nodes, metrics and artifacts
with it. This lives in the route rather than in forget_flow because renaming a
flow calls that too, and a rename must keep its history. The observability
rollups stay: they are the record of what ran, and retention already prunes them.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Uq8mtNb97A7praJLyeEYgs
Give the Home flow-activity name cell a `sm:min-w-32` floor beside its
`max-w-0` cap, so an ordinary name reads in full where there is room
while the phone keeps today's truncation. Name the bar widget's nested
reading by its port rather than its qualified message. Correct
`MarkdownWidget`'s docstring to the subset it actually renders.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SKL7sUgNWhukDEz95vSMQv
Reaching Home from another page builds its charts in the same commit their
theme effect first fires in, with the readings already cached. uPlot ranges its
scales in a microtask, so `redraw()` re-set the x scale from the chart's own —
still empty — bounds before that ran, and the pending range taken from the data
was lost: axes without ticks, no lines, and no way back but the range control,
which rebuilds the chart. Redrawing without the paths is all a colour swap
needs and leaves the scales alone.
The cards also carried "Nothing has run yet." while the first readings were
still on their way; they now carry the skeleton the rest of Home uses.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SKL7sUgNWhukDEz95vSMQv
The socket belonged to whichever hook instance ran its effect first. Passive
effects run children before parents, so on Home that was the brain graph rather
than the shell: navigating to a sibling route unmounted the graph, which closed
the socket, while the shell kept the reference count above zero. From there the
page was deaf for the rest of its life, with nothing left to reconnect it.
A module-level connection with a real refcount replaces it — connect on the
first subscriber, disconnect on the last — and the hook is a thin subscription
with the same signature, correct under StrictMode's mount/unmount/mount.
A 1008 now reconnects instead of returning silently: the token is read afresh
per attempt, and three consecutive rejections fall through to the caller's auth
handler so a revoked session surfaces rather than spins.
The snapshot's emit counts are read into a store of their own, apart from the
live count, so a graph that connects into a busy engine is drawn as busy without
every neuron claiming it just fired. The neuron and edge pulses now key off a
change seen while they were mounted, so returning to Home no longer replays
every emission of the session.
Home gets a live indicator for the case none of this can fix: quiet while the
socket is up, and named in words when it is down, since HTTP polling keeps the
rest of the page looking current.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HTsT1isxUjw5gtkJk8WhuA
The "running" tile counted paused and invalid flows as running, so it read
"5/5" beside "1 flow(s) cannot run". Its note is now additive rather than a
precedence chain, so a quarantine no longer hides the invalid count.
Adds dur() beside si(): a ms reading picks its own unit, so a slow run reads
"1.24 s" instead of "1.2k ms".
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HTsT1isxUjw5gtkJk8WhuA
Two things the first pass got wrong.
The Changes and "Given up on" lists had no cap, so they ran as long as the
audit trail did — and worse, a truncating flex row still offers its whole
unwrapped line as a min-content contribution, so the card sized itself to the
longest entry and took the page sideways with it. Both now use the same
capped, scrolling box the runs and failures lists already had, which fixes the
length and the width together.
Running downwards, a rank of eight nodes was two thousand pixels across
because a node box is landscape: siblings cost a rank four times as much
across as they do down it. A rank wider than two now wraps onto the ranks
below, settling over a few passes since wrapping one rank pushes what was
under it up into the room that freed. The value chip on an edge no longer
reserves its width there either — that width is spent across the screen rather
than along the flow, and the rank gap it would widen is already wider than the
chip is tall. The same flow that laid out 2040x216 is now 803x722; a chain and
a diamond are unchanged, and so is every desktop layout.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VDSXaRhvqHYNevgDGmNAto
The canvas lays itself out: a layered graph, left to right on a desktop and
top to bottom on a phone, with room reserved for the value each edge carries.
Nodes cannot be dragged and `NodeDef.position` is gone from the document —
a graph nobody can arrange is one worth keeping small, which is what keeps
flows atomic. Endpoints join the same layout, so their lanes and the
localStorage that remembered where they were dragged go too.
Mobile, per the new Responsive section of DESIGN-GUIDELINES.md: the dock caps
its width and wraps instead of running off the screen, the dashboard stacks
into one column rather than shrinking a wall panel to a fifth of its size, and
Home stops widening its grid track past the viewport. A Playwright project at
a phone's width fails the build when a screen no longer fits.
Along the way: publish is the checkmark that was already there rather than a
button that appears and disappears, with discard beside it on both the flow
and the dashboard; the brain reveals a neuron's name on the first tap; and the
port sparklines get room to breathe.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VDSXaRhvqHYNevgDGmNAto
The Home table drew the one sparkline without its live dot, on the grounds
that a polled rollup's right edge is the last complete slice rather than this
instant. Identical curves reading differently is the worse trade, so the dot
is back everywhere and the cell keeps room for the half that hangs past the
curve.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XC2jX6Hdj7pxGGKzBTrbqB
The health screen was fixed at 24 hours everywhere except its lists, which
were fixed at nothing: `failuresQueryOptions` read the newest 100 rows and
`HealthActivity` filtered them client-side, so on a busy engine the failures
list covered whatever few minutes 100 rows happened to span while the chart
beside it spanned a day — and pinning an older minute showed an empty card.
One `RangePicker` now sits on the Health heading and governs the whole block:
the tiles, the flow table, both charts and both lists. Presets are 1h / 6h /
24h / 7d — the collector prunes at `OBS_RETENTION_DAYS` (30), so a week is
behind the last one. The longer windows ask for coarser buckets, since a week
of minute rollups is ten thousand points nobody can see.
Failures get the escape hatch the runs already had: a pinned minute is asked
for with `since`/`until` rather than filtered out of what is held, and the
list itself is bound to the selected range. `RUN_DEPTH`/`EVENT_DEPTH` become
one `LIST_DEPTH`, which now buys coverage of the window on screen instead of
a fixed newest-N — narrowing the range is what makes the same rows reach the
whole of it. The "Failures (24h)" tile counts errors over the selected window
from the rollups the table is drawn from, so tile, column and chart agree.
The node-panel and edge trend curves get no picker. They are a Redis ring of
the last 120 values per message with no window to ask for, so hovering one
reveals what it actually shows — how many readings, and the span they cover.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XC2jX6Hdj7pxGGKzBTrbqB
The trend column carried `w-full`, which made it swallow every spare pixel:
the five columns beside it huddled in the left 40% with a dead gap before the
curve. It takes a bounded third now, so the auto layout spreads the rest over
the columns that have content to show, and a long flow name still widens its
own column into the card's scroll.
`si` reaches the rollup counts it was written for — 179240 executions reads as
"179k", and the failure badge and the failures tile follow. The plain window
leaves the small ones alone: 615 is still 615. The standing counts (nodes,
flows running, queue depth) stay exact, as does everything acted on digit by
digit.
The fade ran a full 40 of the 100-unit box. In a table cell three times the
width of a panel that washed out readings still worth seeing, so it is 25 —
a hint at the left end rather than a quarter of the series.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XC2jX6Hdj7pxGGKzBTrbqB
The trend curve was two components: a rich one in the node panel and on edges,
and a bare line in the flow table. It is one `Sparkline` now, taking the colour
token, the height and whether the live dot and the readout show. The flow table
draws its rollup in the chart ramp with no dot — the rollups are polled, so the
right edge is the last completed slice rather than this instant — and keeps its
"nothing yet" state, as the panel keeps its three distinct silences.
All of them fade out to the left, through an SVG mask over the curve and its
area. The dot sits outside the mask: the newest reading is the one thing that
must stay solid.
`si` replaces `compact` and the ad-hoc "k" the uPlot axis carried. It prefixes
k/M/G and m/µ, but only outside 0.01–1000, where the plain number is already
the shortest thing to read and a written unit ("0.4 ms") stays honest. The
sparkline readout asks for four digits, so two neighbouring readings never
collapse into one string. Exact counts and anything the user acts on — a
payload, a form field, the edge inspector's value — are left unrounded.
`src/lib/utils.check.ts` asserts the rounding cases.
The activity table now spans its card: the flow name anchors the left, the four
numbers read down their own centre, and the trend takes the slack on the right.
Panel rhythm steps up one notch, gap-5 to gap-6 outside and gap-2 to gap-3
within a section.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XC2jX6Hdj7pxGGKzBTrbqB
The force layout settled roughly square, so in a wide band the graph sat as a
small island. Vertical centring now pulls nearly three times as hard as
horizontal and the seed sits on an ellipse, which settles at about 2.2:1
without stretching the circles or the edges.
The graph also slid in on load: the rebuild fit ran with a 300ms duration, so
React Flow animated the pan from its default viewport to the content. The fit
is instant now, and the canvas grows in from the centre with `scaleIn` once it
has a layout.
The pinned chart header said "Esc clears", which a phone cannot act on. It
carries the same "× Clear" button as the list header instead; Escape still
works.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017MeiWk3Yq12n2pTvnQWYvt
/observability/runs gains since/until, so the throughput chart's pin asks
the server for its minute instead of filtering a fixed recent list. This
engine writes ~60 runs a minute, so any minute but the newest read empty.
since is inclusive and until exclusive, matching the minute buckets the
charts are drawn from. Hover stays the client-side preview it was:
scrubbing a day would otherwise be a request per minute rested on.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017MeiWk3Yq12n2pTvnQWYvt
Home is the one overview now: the brain graph flat across the top, the flow
switches, then the health sections. The /brain and /health routes and their
sidebar entries are gone.
- charts report their cursor and clicks, so hovering one filters the list
beside it to that minute and a click pins it until Escape or Clear
- chart values round to about three significant digits, the legend mounts
under the plot so it can wrap without leaving the card, and axis ticks
shorten past a thousand
- the embedded brain leaves the wheel to the page rather than zooming
- number fields no longer draw their up/down spinner (NOTEPAD)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017MeiWk3Yq12n2pTvnQWYvt
stroblmeandClaude Opus 5