Repository navigation
Conversation
… errors The Errors tab rendered every diagnostic in the same destructive red and summed them into a single hero count, with missing resources sorted above real errors. Missing models, node packs and media are recoverable setup states, not failures (DES-442), and the Figma gateway frames only cover the pre-first-run case, so the mixed state was unspecified. Give every ErrorGroup a derived severity and compose from it: - The hero splits into an error row and a setup row, each counting only its own severity. No number anywhere sums across severities. - Blocking error sections sort above missing sections. - Section badges and the sidebar tab follow the maximum severity. The tab differentiates by glyph as well as colour so severity is not carried by colour alone. - A validation error whose root cause is an already-tracked missing resource is absorbed into that amber card, which keeps its severity and its resolve actions and gains a compact 'Blocked last run' marker. Matching is by stable identity - node execution id, including the lifted source id for promoted subgraph widgets, plus the normalized widget value - never by message text. An unrelated validation error on the same node still renders its own card. - Missing resources continue not to block Run; covered by regression tests.
🎭 Playwright: ❌ 1762 passed, 4 failed · 1 flaky❌ Failed Tests📊 Browser Reports
🎨 Storybook: ✅ Built — View Storybook📦 Bundle: 8.19 MB gzip 🔴 +2.19 kBDetailsSummary
Category Glance App Entry Points — 3.66 kB (baseline 3.66 kB) • ⚪ 0 BMain entry bundles and manifests
Status: 1 added / 1 removed Graph Workspace — 1.3 MB (baseline 1.29 MB) • 🔴 +10.5 kBGraph editor runtime, canvas, workflow orchestration
Status: 1 added / 1 removed / 1 unchanged Views & Navigation — 112 kB (baseline 112 kB) • ⚪ 0 BTop-level views, pages, and routed surfaces
Status: 12 added / 12 removed / 4 unchanged Panels & Settings — 551 kB (baseline 551 kB) • ⚪ 0 BConfiguration panels, inspectors, and settings screens
Status: 11 added / 11 removed / 15 unchanged User & Accounts — 29.1 kB (baseline 29.1 kB) • ⚪ 0 BAuthentication, profile, and account management bundles
Status: 7 added / 7 removed / 3 unchanged Editors & Dialogs — 125 kB (baseline 125 kB) • ⚪ 0 BModals, dialogs, drawers, and in-app editors
Status: 6 added / 6 removed / 1 unchanged UI Components — 70.7 kB (baseline 70.7 kB) • ⚪ 0 BReusable component library chunks
Status: 6 added / 6 removed / 9 unchanged Data & Services — 3.43 MB (baseline 3.43 MB) • 🔴 +374 BStores, services, APIs, and repositories
Status: 14 added / 14 removed / 3 unchanged Utilities & Hooks — 386 kB (baseline 386 kB) • ⚪ 0 BHelpers, composables, and utility bundles
Status: 19 added / 19 removed / 17 unchanged Vendor & Third-Party — 15.7 MB (baseline 15.7 MB) • ⚪ 0 BExternal libraries and shared vendor chunks Status: 16 unchanged Other — 12.6 MB (baseline 12.6 MB) • 🔴 +523 BBundles that do not match a named category
Status: 77 added / 77 removed / 201 unchanged ⚡ Performance Report
Show regressions
All metrics
Historical variance (last 15 runs)
Trend (last 15 commits on main)
Raw data{
"timestamp": "2026-07-30T12:52:26.168Z",
"gitSha": "044f98e7e5a23e439196ef003d30c0826d28fbb2",
"branch": "jaeone94/errors-panel-mixed-severity",
"measurements": [
{
"name": "canvas-idle",
"durationMs": 2189.6840000000566,
"styleRecalcs": 10,
"styleRecalcDurationMs": 10.925,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 612.706,
"heapDeltaBytes": 5244032,
"heapUsedBytes": 73079860,
"domNodes": 20,
"jsHeapTotalBytes": 20049920,
"scriptDurationMs": 23.055,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "canvas-idle",
"durationMs": 2059.5240000000103,
"styleRecalcs": 11,
"styleRecalcDurationMs": 7.389000000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 502.19,
"heapDeltaBytes": -21359560,
"heapUsedBytes": 46448960,
"domNodes": -279,
"jsHeapTotalBytes": 20963328,
"scriptDurationMs": 15.132000000000001,
"eventListeners": -148,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "canvas-mouse-sweep",
"durationMs": 1857.9560000000583,
"styleRecalcs": 75,
"styleRecalcDurationMs": 40.160000000000004,
"layouts": 12,
"layoutDurationMs": 3.8829999999999996,
"taskDurationMs": 833.43,
"heapDeltaBytes": -417196,
"heapUsedBytes": 67380264,
"domNodes": 58,
"jsHeapTotalBytes": 20312064,
"scriptDurationMs": 114.307,
"eventListeners": 4,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "canvas-mouse-sweep",
"durationMs": 1840.041000000042,
"styleRecalcs": 72,
"styleRecalcDurationMs": 33.443,
"layouts": 12,
"layoutDurationMs": 3.526,
"taskDurationMs": 871.4329999999999,
"heapDeltaBytes": -15402308,
"heapUsedBytes": 52482484,
"domNodes": -281,
"jsHeapTotalBytes": 21225472,
"scriptDurationMs": 109.752,
"eventListeners": -148,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "canvas-zoom-sweep",
"durationMs": 1675.5080000000362,
"styleRecalcs": 31,
"styleRecalcDurationMs": 15.427999999999999,
"layouts": 6,
"layoutDurationMs": 0.662,
"taskDurationMs": 359.757,
"heapDeltaBytes": 7963984,
"heapUsedBytes": 75756580,
"domNodes": 77,
"jsHeapTotalBytes": 20574208,
"scriptDurationMs": 17.258000000000003,
"eventListeners": 19,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "canvas-zoom-sweep",
"durationMs": 1739.7060000000693,
"styleRecalcs": 31,
"styleRecalcDurationMs": 15.884999999999996,
"layouts": 6,
"layoutDurationMs": 0.5950000000000001,
"taskDurationMs": 371.644,
"heapDeltaBytes": 7981792,
"heapUsedBytes": 75816464,
"domNodes": 77,
"jsHeapTotalBytes": 19787776,
"scriptDurationMs": 18.249000000000002,
"eventListeners": 19,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "dom-widget-clipping",
"durationMs": 607.3300000000472,
"styleRecalcs": 12,
"styleRecalcDurationMs": 9.114,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 394.59299999999996,
"heapDeltaBytes": -12550712,
"heapUsedBytes": 55346512,
"domNodes": 20,
"jsHeapTotalBytes": 22147072,
"scriptDurationMs": 59.768,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "dom-widget-clipping",
"durationMs": 609.230000000025,
"styleRecalcs": 12,
"styleRecalcDurationMs": 7.403,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 366.14000000000004,
"heapDeltaBytes": -12067928,
"heapUsedBytes": 55727688,
"domNodes": 20,
"jsHeapTotalBytes": 21884928,
"scriptDurationMs": 54.086999999999996,
"eventListeners": 0,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "large-graph-idle",
"durationMs": 2028.5959999999932,
"styleRecalcs": 9,
"styleRecalcDurationMs": 7.082999999999999,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 671.257,
"heapDeltaBytes": 8346708,
"heapUsedBytes": 67396800,
"domNodes": -281,
"jsHeapTotalBytes": 4517888,
"scriptDurationMs": 93.67099999999999,
"eventListeners": -144,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "large-graph-idle",
"durationMs": 2027.152000000001,
"styleRecalcs": 9,
"styleRecalcDurationMs": 7.318000000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 654.6890000000001,
"heapDeltaBytes": 8553992,
"heapUsedBytes": 67445744,
"domNodes": -282,
"jsHeapTotalBytes": 4780032,
"scriptDurationMs": 90.26299999999999,
"eventListeners": -144,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "large-graph-pan",
"durationMs": 2172.2140000000536,
"styleRecalcs": 68,
"styleRecalcDurationMs": 14.203000000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 1208.3129999999999,
"heapDeltaBytes": 3370844,
"heapUsedBytes": 63156876,
"domNodes": -282,
"jsHeapTotalBytes": 5267456,
"scriptDurationMs": 396.087,
"eventListeners": -142,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333335,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "large-graph-pan",
"durationMs": 2186.427999999978,
"styleRecalcs": 69,
"styleRecalcDurationMs": 14.558000000000002,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 1208.3249999999998,
"heapDeltaBytes": 5890832,
"heapUsedBytes": 66335600,
"domNodes": -284,
"jsHeapTotalBytes": 5267456,
"scriptDurationMs": 397.304,
"eventListeners": -144,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "large-graph-zoom",
"durationMs": 3196.4080000000195,
"styleRecalcs": 66,
"styleRecalcDurationMs": 16.569,
"layouts": 60,
"layoutDurationMs": 8.101,
"taskDurationMs": 1519.4540000000002,
"heapDeltaBytes": -3150560,
"heapUsedBytes": 58222772,
"domNodes": -285,
"jsHeapTotalBytes": 7925760,
"scriptDurationMs": 527.957,
"eventListeners": -148,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "large-graph-zoom",
"durationMs": 3136.634000000072,
"styleRecalcs": 64,
"styleRecalcDurationMs": 13.765000000000002,
"layouts": 60,
"layoutDurationMs": 7.672,
"taskDurationMs": 1496.222,
"heapDeltaBytes": -5249672,
"heapUsedBytes": 56578652,
"domNodes": -289,
"jsHeapTotalBytes": 7925760,
"scriptDurationMs": 538.644,
"eventListeners": -148,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "minimap-idle",
"durationMs": 2008.2569999999578,
"styleRecalcs": 8,
"styleRecalcDurationMs": 6.0420000000000025,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 652.525,
"heapDeltaBytes": 7590996,
"heapUsedBytes": 67874732,
"domNodes": -283,
"jsHeapTotalBytes": 4780032,
"scriptDurationMs": 88.376,
"eventListeners": -144,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "minimap-idle",
"durationMs": 2026.6599999999926,
"styleRecalcs": 9,
"styleRecalcDurationMs": 7.171,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 658.892,
"heapDeltaBytes": 7579412,
"heapUsedBytes": 68251676,
"domNodes": -282,
"jsHeapTotalBytes": 4780032,
"scriptDurationMs": 90.74300000000001,
"eventListeners": -144,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.699999999999818
},
{
"name": "subgraph-dom-widget-clipping",
"durationMs": 546.8830000000935,
"styleRecalcs": 46,
"styleRecalcDurationMs": 9.336999999999998,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 380.523,
"heapDeltaBytes": -11851592,
"heapUsedBytes": 56187724,
"domNodes": 18,
"jsHeapTotalBytes": 22671360,
"scriptDurationMs": 116.02699999999999,
"eventListeners": 8,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "subgraph-dom-widget-clipping",
"durationMs": 603.2149999999774,
"styleRecalcs": 47,
"styleRecalcDurationMs": 10.14,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 396.479,
"heapDeltaBytes": -11767840,
"heapUsedBytes": 56203580,
"domNodes": 20,
"jsHeapTotalBytes": 22147072,
"scriptDurationMs": 123.66700000000002,
"eventListeners": 6,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333335,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "subgraph-idle",
"durationMs": 2004.7149999999192,
"styleRecalcs": 10,
"styleRecalcDurationMs": 8.818999999999999,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 466.62899999999996,
"heapDeltaBytes": 4397536,
"heapUsedBytes": 72332648,
"domNodes": 20,
"jsHeapTotalBytes": 20574208,
"scriptDurationMs": 13.961,
"eventListeners": 4,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "subgraph-idle",
"durationMs": 1998.9219999999932,
"styleRecalcs": 10,
"styleRecalcDurationMs": 8.131999999999998,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 445.53400000000005,
"heapDeltaBytes": 4332664,
"heapUsedBytes": 72476004,
"domNodes": 20,
"jsHeapTotalBytes": 20836352,
"scriptDurationMs": 12.372000000000002,
"eventListeners": 4,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333335,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-mouse-sweep",
"durationMs": 1690.1719999999614,
"styleRecalcs": 78,
"styleRecalcDurationMs": 33.193,
"layouts": 16,
"layoutDurationMs": 4.3149999999999995,
"taskDurationMs": 781.438,
"heapDeltaBytes": -20562916,
"heapUsedBytes": 47380916,
"domNodes": -281,
"jsHeapTotalBytes": 20701184,
"scriptDurationMs": 85.33399999999999,
"eventListeners": -148,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "subgraph-mouse-sweep",
"durationMs": 1713.4149999999408,
"styleRecalcs": 76,
"styleRecalcDurationMs": 33.124,
"layouts": 16,
"layoutDurationMs": 3.888,
"taskDurationMs": 756.597,
"heapDeltaBytes": -19667420,
"heapUsedBytes": 48183744,
"domNodes": -279,
"jsHeapTotalBytes": 20963328,
"scriptDurationMs": 82.26,
"eventListeners": -148,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "subgraph-transition-enter",
"durationMs": 1367.9639999999154,
"styleRecalcs": 19,
"styleRecalcDurationMs": 28.829999999999995,
"layouts": 15,
"layoutDurationMs": 11.268,
"taskDurationMs": 894.3670000000002,
"heapDeltaBytes": 29109604,
"heapUsedBytes": 98086444,
"domNodes": 13673,
"jsHeapTotalBytes": 16252928,
"scriptDurationMs": 40.491,
"eventListeners": 2371,
"totalBlockingTimeMs": 121,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.800000000000182
},
{
"name": "viewport-pan-sweep",
"durationMs": 8215.38499999997,
"styleRecalcs": 249,
"styleRecalcDurationMs": 39.307,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 4167.054,
"heapDeltaBytes": 10745152,
"heapUsedBytes": 69855852,
"domNodes": -282,
"jsHeapTotalBytes": 7364608,
"scriptDurationMs": 1264.994,
"eventListeners": -128,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "viewport-pan-sweep",
"durationMs": 8189.01500000004,
"styleRecalcs": 249,
"styleRecalcDurationMs": 39.42900000000001,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 4228.135,
"heapDeltaBytes": 10047024,
"heapUsedBytes": 69813984,
"domNodes": -279,
"jsHeapTotalBytes": 6807552,
"scriptDurationMs": 1283.896,
"eventListeners": -128,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.66333333333332,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "vue-large-graph-idle",
"durationMs": 17147.937999999955,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 17133.131,
"heapDeltaBytes": -39953384,
"heapUsedBytes": 166858548,
"domNodes": -8313,
"jsHeapTotalBytes": -12521472,
"scriptDurationMs": 536.7239999999999,
"eventListeners": -16389,
"totalBlockingTimeMs": 0,
"frameDurationMs": 17.776666666666763,
"p95FrameDurationMs": 16.80000000000291
},
{
"name": "vue-large-graph-idle",
"durationMs": 17462.24399999994,
"styleRecalcs": 0,
"styleRecalcDurationMs": 0,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 17427.893,
"heapDeltaBytes": -31522404,
"heapUsedBytes": 171498560,
"domNodes": -8312,
"jsHeapTotalBytes": -7802880,
"scriptDurationMs": 541.874,
"eventListeners": -16387,
"totalBlockingTimeMs": 0,
"frameDurationMs": 17.780000000000047,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-large-graph-pan",
"durationMs": 20870.30799999991,
"styleRecalcs": 145,
"styleRecalcDurationMs": 18.728999999999996,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 20846.786,
"heapDeltaBytes": -30778452,
"heapUsedBytes": 177245204,
"domNodes": -8312,
"jsHeapTotalBytes": -13119488,
"scriptDurationMs": 811.185,
"eventListeners": -16383,
"totalBlockingTimeMs": 218,
"frameDurationMs": 17.779999999999927,
"p95FrameDurationMs": 16.799999999999272
},
{
"name": "vue-large-graph-pan",
"durationMs": 21423.107000000073,
"styleRecalcs": 150,
"styleRecalcDurationMs": 19.16900000000002,
"layouts": 0,
"layoutDurationMs": 0,
"taskDurationMs": 21400.05,
"heapDeltaBytes": -66477180,
"heapUsedBytes": 154393632,
"domNodes": -8312,
"jsHeapTotalBytes": -14622720,
"scriptDurationMs": 912.729,
"eventListeners": -16385,
"totalBlockingTimeMs": 285,
"frameDurationMs": 18.333333333333332,
"p95FrameDurationMs": 16.80000000000291
},
{
"name": "workflow-execution",
"durationMs": 450.54400000003625,
"styleRecalcs": 14,
"styleRecalcDurationMs": 19.43,
"layouts": 3,
"layoutDurationMs": 0.7489999999999999,
"taskDurationMs": 111.19000000000001,
"heapDeltaBytes": -15779416,
"heapUsedBytes": 51184412,
"domNodes": 130,
"jsHeapTotalBytes": 7729152,
"scriptDurationMs": 9.499,
"eventListeners": 65,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
},
{
"name": "workflow-execution",
"durationMs": 470.1039999999921,
"styleRecalcs": 18,
"styleRecalcDurationMs": 19.851999999999997,
"layouts": 3,
"layoutDurationMs": 1.36,
"taskDurationMs": 110.07000000000001,
"heapDeltaBytes": -15995812,
"heapUsedBytes": 51028816,
"domNodes": 141,
"jsHeapTotalBytes": 7991296,
"scriptDurationMs": 8.751,
"eventListeners": 65,
"totalBlockingTimeMs": 0,
"frameDurationMs": 16.666666666666668,
"p95FrameDurationMs": 16.700000000000728
}
]
} |
📝 WalkthroughWalkthroughThis change classifies panel errors as blocking or setup-related, absorbs matching missing-resource validations into missing-resource groups, tracks blocked runs, updates error-panel presentation and accessibility, and clears resolved missing-node prompt errors. ChangesError severity and absorption
Panel presentation and cleanup
Regression coverage
Estimated code review effort: 4 (Complex) | ~45 minutes Possibly related PRs
Suggested labels: Suggested reviewers: 🚥 Pre-merge checks | ✅ 6✅ Passed checks (6 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Codecov Report❌ Patch coverage is
@@ Coverage Diff @@
## main #14252 +/- ##
==========================================
- Coverage 79.43% 79.21% -0.23%
==========================================
Files 1755 1764 +9
Lines 117569 119732 +2163
Branches 38946 40533 +1587
==========================================
+ Hits 93390 94843 +1453
- Misses 23586 24265 +679
- Partials 593 624 +31
Flags with carried forward coverage won't be shown. Click here to find out more.
... and 74 files with indirect coverage changes 🚀 New features to boost your workflow:
|
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/components/rightSidePanel/errors/ErrorGroupList.vue`:
- Around line 32-46: Replace the static -hover design tokens with their
equivalent non-hover destructive semantic tokens: update
text-destructive-background-hover in ErrorGroupList.vue lines 32-46 for the hero
error count, and bg-destructive-background-hover in ErrorCardSection.vue lines
12-20 for the severity badge fill. Preserve the existing styling and layout.
In `@src/components/rightSidePanel/errors/missingResourceAbsorption.ts`:
- Around line 44-82: Extract the shared validation and widget/received-value
matching logic from matchesMissingModel and matchesMissingMedia into a helper
that accepts candidateNodeIds and the candidate’s missing-state data. Update
both functions to supply their respective node ID lists—[sourceExecutionId,
nodeId] for models and [nodeId] for media—while preserving their existing
matching behavior.
In `@src/components/rightSidePanel/errors/TabErrors.test.ts`:
- Around line 493-502: Remove the negative style assertions for rounded-lg,
bg-warning-background, and text-charcoal-800 from the missingHero count test.
Keep the positive class assertions and the existing “Setup pending” text
assertion that verify the intended severity behavior.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: e2a49adc-3557-4032-8363-3bd9a8ec53ca
📒 Files selected for processing (14)
src/components/actionbar/ComfyRunButton/ComfyQueueButton.test.tssrc/components/error/ErrorOverlay.test.tssrc/components/error/useErrorOverlayState.test.tssrc/components/rightSidePanel/RightSidePanel.vuesrc/components/rightSidePanel/errors/ErrorCardSection.vuesrc/components/rightSidePanel/errors/ErrorGroupList.vuesrc/components/rightSidePanel/errors/TabErrors.test.tssrc/components/rightSidePanel/errors/missingResourceAbsorption.test.tssrc/components/rightSidePanel/errors/missingResourceAbsorption.tssrc/components/rightSidePanel/errors/types.tssrc/components/rightSidePanel/errors/useErrorGroups.test.tssrc/components/rightSidePanel/errors/useErrorGroups.tssrc/components/rightSidePanel/errors/useHasBlockingError.tssrc/locales/en/main.json
| <div data-testid="errors-summary-hero" class="bg-base-foreground/5"> | ||
| <div | ||
| v-if="errorCount > 0" | ||
| data-testid="errors-summary-hero-error" | ||
| class="flex items-center gap-2 p-2" | ||
| > | ||
| {{ totalErrorCount }} | ||
| </span> | ||
| <span | ||
| aria-hidden="true" | ||
| class="h-9 w-px shrink-0 bg-interface-stroke" | ||
| /> | ||
| <div class="flex min-w-0 flex-1 flex-col gap-1 px-2"> | ||
| <span class="text-xs/tight font-semibold text-base-foreground"> | ||
| {{ t('rightSidePanel.errorsDetected', totalErrorCount) }} | ||
| <span | ||
| class="flex h-12 min-w-9 shrink-0 items-center justify-center px-1 text-[2rem]/none font-extrabold text-destructive-background-hover tabular-nums" | ||
| > | ||
| {{ errorCount }} | ||
| </span> | ||
| <span class="text-xs/tight text-muted-foreground"> | ||
| {{ t('rightSidePanel.resolveBeforeRun') }} | ||
| <span | ||
| aria-hidden="true" | ||
| class="h-9 w-px shrink-0 bg-interface-stroke" | ||
| /> |
There was a problem hiding this comment.
📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win
-hover-suffixed design tokens used as static (non-hover) colors in two places. Both sites apply a token that's meant for hover/selected states as a permanent base color, which the repo's design-standards guidance explicitly says to avoid ("skip -hover/-selected suffixed tokens and derive hover/selected styles programmatically").
src/components/rightSidePanel/errors/ErrorGroupList.vue#L32-L46: replacetext-destructive-background-hover(hero error count, line ~39) with the non-hover destructive semantic token, or a distinct "strong emphasis" token if one exists.src/components/rightSidePanel/errors/ErrorCardSection.vue#L12-L20: replacebg-destructive-background-hover(error-severity badge fill) with the equivalent non-hover token.
📍 Affects 2 files
src/components/rightSidePanel/errors/ErrorGroupList.vue#L32-L46(this comment)src/components/rightSidePanel/errors/ErrorCardSection.vue#L12-L20
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@src/components/rightSidePanel/errors/ErrorGroupList.vue` around lines 32 -
46, Replace the static -hover design tokens with their equivalent non-hover
destructive semantic tokens: update text-destructive-background-hover in
ErrorGroupList.vue lines 32-46 for the hero error count, and
bg-destructive-background-hover in ErrorCardSection.vue lines 12-20 for the
severity badge fill. Preserve the existing styling and layout.
Source: Path instructions
R4 absorbed validation errors for missing models and media, but not for missing node packs, so running with an uninstalled pack still produced a red "Missing node type" card alongside the amber Missing Node Packs section - the same problem stated twice, which is exactly the duplication the rule exists to prevent. A node pack cannot go through the node-scoped resource matcher: the backend rejects it at prompt level, so the error carries no node id, no input name and no pack id (the pack is inferred client-side from the node type name, and can resolve to null). Match it at the type level instead: absorb when the prompt error is missing_node_type and missing node types are tracked. The second condition matters - if the re-scan found nothing, there is no amber section to carry the meaning, so the red card must stay. Both the panel and the sidebar tab read the same predicate, so the tab cannot turn red for an error the panel has absorbed. Two consequences ship with it, because the feature is wrong without them: - Clear the prompt error once the missing node types are resolved. It reads live state, so without this, clicking Replace All or deleting the last missing node un-absorbed the error and escalated the panel from amber to red right after the user fixed the only problem. Guarded on the non-empty to empty transition: in the desync case where the backend reports a missing type the re-scan cannot find, the red card is the only surface, and deleting any unrelated node must not wipe it. - Shorten the blocked-last-run marker as the header narrows. The node sections carry wide Install All / Replace All buttons, so a full-width marker squeezed the section title until it truncated at the default panel width. The marker now yields instead of the title: full text when the header is wide, "Blocked" when tighter, the dot alone when narrow, with a screen-reader label that stays constant across all three. Container queries only - no logic, following the existing idiom in KeybindingList.vue.
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (3)
src/components/rightSidePanel/errors/ErrorGroupList.vue (1)
136-154: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winBelow 332px the marker degrades to an unlabeled dot for sighted users.
Both text variants are hidden under the 332px container breakpoint, leaving only the
aria-hiddendot. Screen-reader users still get thesr-onlytext, but sighted users at narrow panel widths see a red dot with no affordance explaining it. Add atitle(or tooltip) on the wrapper so the meaning survives the narrowest band.🛡️ Proposed fix
data-testid="blocked-last-run-indicator" + :title="t('rightSidePanel.blockedLastRun')" class="flex shrink-0 items-center gap-1.5 text-xs text-muted-foreground"🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/components/rightSidePanel/errors/ErrorGroupList.vue` around lines 136 - 154, Add a visible hover affordance to the blocked-last-run marker wrapper in ErrorGroupList, using the full blockedLastRun translation as its title or tooltip. Ensure the label remains available below the 332px breakpoint while preserving the existing responsive text and screen-reader content.src/components/rightSidePanel/errors/ErrorCardSection.vue (1)
18-21: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winBadge foregrounds bypass semantic theme tokens.
text-whiteandtext-charcoal-800are fixed palette values, so the badge text will not track theme changes and contrast onbg-warning-backgroundis unverified across themes. Map these to the semantic foreground tokens that pair with the destructive/warning backgrounds.As per path instructions,
docs/guidance/design-standards.mdrequires mapping Figma tokens to Tailwind semantic tokens (no hardcoded values).🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/components/rightSidePanel/errors/ErrorCardSection.vue` around lines 18 - 21, Update the severity badge classes in ErrorCardSection’s class selection to replace the hardcoded text-white and text-charcoal-800 colors with the semantic foreground tokens paired with the destructive and warning backgrounds. Preserve the existing severity-specific background classes and follow the semantic token mappings defined in the design standards.Source: Path instructions
src/components/rightSidePanel/errors/useErrorGroups.test.ts (1)
331-391: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winTwo new tests can pass vacuously. Both assertions are purely relational over derived values and never pin an expected outcome, so a regression that empties or uniformly flips the derived state still satisfies them.
src/components/rightSidePanel/errors/useErrorGroups.test.ts#L331-L391: replaceexpectSeveritySignalsToMatch()with a helper taking an explicit expected boolean, and assert bothhasBlockingError.valueand the group-severity predicate against it at each phase.src/components/rightSidePanel/errors/useErrorGroups.test.ts#L518-L527: asserterrorIndicesandmissingIndicesare both non-empty before theMath.max/Math.mincomparison, since spreading an empty array yields-Infinity/Infinity.As per path instructions,
docs/guidance/vitest.mdrequires behavioral assertions rather than change-detector tests that only assert defaults.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/components/rightSidePanel/errors/useErrorGroups.test.ts` around lines 331 - 391, Strengthen both tests in src/components/rightSidePanel/errors/useErrorGroups.test.ts:331-391 and :518-527. In the test around keeps blocking severity aligned with rendered groups, change expectSeveritySignalsToMatch to accept an explicit expected boolean and assert both hasBlockingError.value and the group-severity predicate equal it at every phase. In the errorIndices/missingIndices test at :518-527, assert both arrays are non-empty before the Math.max/Math.min comparison so the test verifies behavioral data rather than vacuous extrema.Source: Path instructions
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/components/rightSidePanel/errors/ErrorGroupList.vue`:
- Around line 126-132: Move the optional blockedLastRun boolean from the four
missing-severity variants onto ErrorGroupBase, removing the duplicated
declarations from those variants. Then simplify the ErrorGroupList.vue condition
to check only group.blockedLastRun, preserving the existing truthy behavior
while eliminating the explicit type enumeration.
---
Outside diff comments:
In `@src/components/rightSidePanel/errors/ErrorCardSection.vue`:
- Around line 18-21: Update the severity badge classes in ErrorCardSection’s
class selection to replace the hardcoded text-white and text-charcoal-800 colors
with the semantic foreground tokens paired with the destructive and warning
backgrounds. Preserve the existing severity-specific background classes and
follow the semantic token mappings defined in the design standards.
In `@src/components/rightSidePanel/errors/ErrorGroupList.vue`:
- Around line 136-154: Add a visible hover affordance to the blocked-last-run
marker wrapper in ErrorGroupList, using the full blockedLastRun translation as
its title or tooltip. Ensure the label remains available below the 332px
breakpoint while preserving the existing responsive text and screen-reader
content.
In `@src/components/rightSidePanel/errors/useErrorGroups.test.ts`:
- Around line 331-391: Strengthen both tests in
src/components/rightSidePanel/errors/useErrorGroups.test.ts:331-391 and
:518-527. In the test around keeps blocking severity aligned with rendered
groups, change expectSeveritySignalsToMatch to accept an explicit expected
boolean and assert both hasBlockingError.value and the group-severity predicate
equal it at every phase. In the errorIndices/missingIndices test at :518-527,
assert both arrays are non-empty before the Math.max/Math.min comparison so the
test verifies behavioral data rather than vacuous extrema.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 032e5470-d408-4558-a4cd-b528914dd12c
📒 Files selected for processing (15)
src/components/rightSidePanel/errors/ErrorCardSection.vuesrc/components/rightSidePanel/errors/ErrorGroupList.vuesrc/components/rightSidePanel/errors/TabErrors.test.tssrc/components/rightSidePanel/errors/missingResourceAbsorption.tssrc/components/rightSidePanel/errors/types.tssrc/components/rightSidePanel/errors/useErrorGroups.test.tssrc/components/rightSidePanel/errors/useErrorGroups.tssrc/components/rightSidePanel/errors/useHasBlockingError.tssrc/composables/graph/useErrorClearingHooks.test.tssrc/composables/graph/useErrorClearingHooks.tssrc/locales/en/main.jsonsrc/platform/nodeReplacement/useNodeReplacement.test.tssrc/platform/nodeReplacement/useNodeReplacement.tssrc/stores/executionErrorStore.tssrc/utils/executionErrorUtil.ts
…ed marker The blocked-last-run marker was briefly made responsive with container queries so that it, rather than the section title, gave way when the header ran out of room. Two things were wrong with that. It keyed off the wrong thing. A container query reports the container's own size, and every section header is the panel width minus padding, so a single threshold applied to all four sections. Sections with a narrow action button - Missing Models, Missing Inputs - shortened their marker at widths where the full text still fit. Per-section thresholds would have fixed that, but they depend on measured text widths, and the section title currently renders in Arial rather than Inter because it sits in a button and the app loads no Tailwind preflight. Correcting that font later would silently invalidate every threshold. Flex shrink cannot substitute for it either: the title button is flex-1, so its basis is 0 and it never overflows - it simply grows less. Shrink factors on the marker are inert. Keep the marker at its full text and let the title truncate. A truncated section title is still identifiable from the section's own contents and description, which is not true of a truncated status marker.
temp/ is gitignored; these two files were tracked before the ignore applied and would otherwise ship in the PR diff.
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/components/rightSidePanel/errors/errorSeverityClassification.test.ts`:
- Line 1: Replace createPinia with createTestingPinia in
errorSeverityClassification.test.ts and initialize it with stubActions: false,
updating the import and setup calls while preserving the existing test behavior.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 678e29da-ef46-4eea-ad42-f3394580b13a
📒 Files selected for processing (18)
browser_tests/tests/propertiesPanel/errorsTabMissingModels.spec.tssrc/components/error/ErrorOverlay.test.tssrc/components/error/useErrorOverlayState.test.tssrc/components/rightSidePanel/errors/ErrorGroupList.vuesrc/components/rightSidePanel/errors/TabErrors.test.tssrc/components/rightSidePanel/errors/errorSeverityClassification.test.tssrc/components/rightSidePanel/errors/errorSeverityClassification.tssrc/components/rightSidePanel/errors/missingResourceAbsorption.test.tssrc/components/rightSidePanel/errors/missingResourceAbsorption.tssrc/components/rightSidePanel/errors/types.tssrc/components/rightSidePanel/errors/useErrorClassification.tssrc/components/rightSidePanel/errors/useErrorGroups.test.tssrc/components/rightSidePanel/errors/useErrorGroups.tssrc/components/rightSidePanel/errors/useHasBlockingError.tssrc/composables/graph/useErrorClearingHooks.test.tssrc/platform/nodeReplacement/missingNodeScan.test.tssrc/platform/nodeReplacement/useNodeReplacement.tssrc/stores/executionErrorStore.ts
| @@ -0,0 +1,144 @@ | |||
| import { createPinia, setActivePinia } from 'pinia' | |||
There was a problem hiding this comment.
📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win
Use createTestingPinia instead of createPinia.
The sibling suite added in this PR (missingResourceAbsorption.test.ts line 31) already uses createTestingPinia({ stubActions: false }); align this file for consistency.
♻️ Proposed change
-import { createPinia, setActivePinia } from 'pinia'
+import { createTestingPinia } from '`@pinia/testing`'
+import { setActivePinia } from 'pinia'
import { beforeEach, describe, expect, it } from 'vitest' beforeEach(() => {
- setActivePinia(createPinia())
+ setActivePinia(createTestingPinia({ stubActions: false }))
})As per path instructions: "Pinia tests: use createTestingPinia (not createPinia) and prefer stubActions: false when testing real action behavior."
Also applies to: 16-18
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@src/components/rightSidePanel/errors/errorSeverityClassification.test.ts` at
line 1, Replace createPinia with createTestingPinia in
errorSeverityClassification.test.ts and initialize it with stubActions: false,
updating the import and setup calls while preserving the existing test behavior.
Source: Path instructions
|
Superseded by two stacked PRs. The change was large enough that the two concerns should be reviewed separately:
The split was verified by recombination: #14440's tree is byte-identical to this branch's tip ( All review feedback from this PR is carried into the two successors. |
## ELI5 Our warning color (gold) is bright enough to read on dark backgrounds, but on the light theme it's like yellow chalk on a whiteboard — technically there, hard to see. This adds one darker shade of the same gold to the palette so light-theme warnings can meet contrast guidelines. ## Summary - Adds `--color-gold-700: #c77700` to the design-system primitives, extending the existing gold ramp (400/500/600). - Same hue family as the current golds; measured for the light-theme `--warning-background` slot, where gold-400 sits at 1.48:1 against the hero band (WCAG wants 3:1 at that size). ## Context Identified during the mixed-severity errors panel review (Comfy-Org#14252): dark theme measures 7.28:1 but light theme fails contrast. Agreed there that the token lands via the design-system side and Comfy-Org#14252's follow-up points `--warning-background` at it for light theme, so this PR intentionally adds the primitive only — no consumers change and there is no visual change yet. Note Tailwind v4 may tree-shake the unused variable from built CSS until it's referenced; the source token is what the follow-up wires up. Part of DES-588. ## Test plan - [ ] `pnpm format:check` and stylelint pass (pre-commit) - [ ] No visual change expected until `--warning-background` (light) is repointed in the Comfy-Org#14252 follow-up - [ ] Not user-visible on its own, so no feature flag needed
…ors (Comfy-Org#14439) Missing resources (node packs, models, media) currently render as blocking errors, identical in weight to a runtime failure. They are not the same thing: a missing model is resolved by installing it, a failed sampler is not. This PR introduces the severity distinction and its presentation, following the updated Figma spec (`Errors-ComfyUI`, node `3052-8805`). It does not change which error cards are produced. - `ErrorGroupSeverity` on `ErrorGroupBase`, assigned at each group builder - One unified summary hero: coral count + "Errors detected" (errors only), gold count + "Setup required" (missing only), neutral total + "Issues detected" (mixed) - In the mixed case, outline-only per-severity filter chips under the hero. A pressed chip shows only that severity; it releases automatically — never auto-selects — when the hidden severity gains new entries, when the filtered severity empties, or when the canvas selection changes to include the hidden severity - Gold palette for missing-resource section badges, and a gold `triangle-alert` tab icon (with an accessible label) when nothing is blocking - Execution error groups sort above missing-resource groups - While the Issues tab is only retained by a pending error scan (no issues present), it renders no severity icon - Context strip wording follows severity ("items" for setup-only workflows) - The Errors tab is renamed to Issues (folds Comfy-Org#14253 in: the tab now lists two severities side by side, and the selection-emphasis strings previously mislabeled missing-only nodes as errors) - Light theme uses the DES-588 AA amber: gold-800 `#8c6a38` behind a `warning-foreground` semantic for chip/hero/icon text, and `warning-on-background` (charcoal-800) for badge text on gold; dark theme renders unchanged `main` already renders both the missing-resource card and the duplicate validation card under one red hero, so this PR classifies an existing duplicate rather than creating one. The duplicate itself is removed in the follow-up. Follow-up: Comfy-Org#14440 absorbs the duplicate validation errors into the missing-resource cards, adds the "Blocked last run" marker, and restyles it to the new spec. Notes for reviewers: - `TabErrors.test.ts` retains the test name `uses error severity for an unabsorbed node validation error`. Absorption does not exist yet in this PR — the name is kept so the follow-up does not have to churn it back. - Light-theme amber text follows the DES-588 value; the dark-theme badge keeps the Figma-specified white-on-gold (2.16:1 — flagged to design). - Part of the errors-tab severity work split out of Comfy-Org#14252 ## Screenshots <img width="618" height="497" alt="스크린샷 2026-08-12 오전 9 59 51" src="https://github.com/user-attachments/assets/57385397-b217-471c-8454-904f98f2267d" /> <img width="620" height="310" alt="스크린샷 2026-08-12 오전 9 59 59" src="https://github.com/user-attachments/assets/97056832-bc46-47df-ac73-fdd9b44588bf" /> <img width="618" height="346" alt="스크린샷 2026-08-12 오전 10 00 07" src="https://github.com/user-attachments/assets/f2c21e85-b7f8-400d-a172-f2ac14d446cc" />
…e cards (Comfy-Org#14440) Builds on Comfy-Org#14439, which introduced the severity distinction. This PR removes the duplicate reporting that severity alone only reclassified. When a resource already tracked as missing also fails backend validation, the panel can show the same problem twice. This absorbs the validation failure into the existing missing-resource warning. - Absorption for missing models and media, matched on node identity and input name, including every active consumer of inputs promoted across a subgraph boundary - Prompt-level `missing_node_type` errors absorb into the missing-node warning, including replaceable nodes - "Blocked last run" marker so an amber card still says the run was blocked - Severity classification extracted so the panel body and the tab icon cannot disagree; the old aggregate totalErrorCount and constituent counters are removed - Missing-node state survives submission, and the resolved prompt error is cleared through a single store transition rather than hand-placed call sites Model and media verification retire their previously absorbed validation errors independently. A pending or failed verification preserves that resource kind's errors while another kind can finish. Deferred confirmations update the panel without waiting for another workflow reload. Only confirmed missing resources absorb matching validation errors; uncertain matches stay blocking. Candidates whose widget was removed or whose selection changed during verification are discarded. Promoted models retain their stored owner's metadata while tracking all active consumers for matching and scope checks. Errors with unnormalisable node IDs remain visible and count toward affected nodes, without exposing an invalid navigation target. The Blocked last run marker remains per resource kind. Finer attribution to individual cards or mixed swap-node/missing-pack groups, and consolidation of the two media consumer traversals, are follow-ups. - Part of the errors-tab severity work split out of Comfy-Org#14252 ## Screenshots Before <img width="618" height="497" alt="스크린샷 2026-08-12 오전 9 59 51" src="https://github.com/user-attachments/assets/16832e31-6e9f-4e5c-860f-5a1ac67add8a" /> After <img width="519" height="400" alt="스크린샷 2026-08-12 오전 11 35 21" src="https://github.com/user-attachments/assets/fee8addc-f290-43af-9049-3902807df6b6" />
Summary
Missing models, node packs and media are recoverable setup states, not failures (DES-442), so they render in amber rather than red — and this defines how the panel behaves when amber setup items and real blocking errors are present at the same time, which the Figma gateway frames never mapped because they only cover the pre-first-run case.
Changes
ErrorGroupgains a derivedseverity('error' | 'missing'), and presentation composes from it:allErrorGroupsandselectionScopedGroups.severityis a required field onErrorGroupand a required prop onErrorCardSection. All construction sites and test mocks are updated.Review Focus
Absorption matching (the critical part).
missingResourceAbsorption.tsis a pure module matched by stable identity, never by error-message text:[error node id, error.extra_info.source_execution_id]against[candidate.sourceExecutionId, candidate.nodeId]. Both halves are needed:liftNodeErrorsToBoundaryre-keys input-level errors onto the host node, while the candidate holds the interior id, so a model promoted out of a subgraph would otherwise never match and would render a duplicate red card.received_valueagainst the normalizedcandidate.name.candidate.directoryis deliberately not part of the comparison — it is the model category (checkpoints,loras) fromgetCategoryForNodeType(), not the parent folder of the widget value, so comparing them can never succeed.useHasBlockingError. The tab must not go red for an error the panel has absorbed, so it derives from the same absorption verdict rather than raw store booleans. It intentionally re-implements the iteration rather than unifying the stores (out of scope); a drift-alarm test pins it againstallErrorGroups.Node packs absorb through a different match. A missing pack is rejected at prompt level, so the error carries no node id, no input name and no pack id — the pack is inferred client-side from the node type name and can resolve to
null. It is therefore matched at the type level: absorb when the prompt error ismissing_node_typeand missing node types are tracked. The second condition is load-bearing — if the re-scan found nothing (a backend/frontend desync), there is no amber section to carry the meaning, so the red card must stay.The stale prompt error is cleared on resolution. Absorption reads live state, so without this, clicking Replace All or deleting the last missing node un-absorbed the error and escalated the panel from amber to red right after the user fixed the only problem. It is guarded on the non-empty → empty transition, so in the desync case above, deleting an unrelated node does not wipe the red card that is the only surface for that failure.
The section title truncates, and the marker does not. The node sections carry wide
Install All/Replace Allbuttons, so at narrow panel widths the header cannot fit both the full title and the marker. Container-query tiers were tried and removed: a container query reports the container's own width, which is the panel width for every section, so sections with a narrow action button shortened their marker while they still had room. Per-section thresholds would depend on measured text widths, and the section title currently renders in Arial rather than Inter because it sits in a button and the app loads no Tailwind preflight — correcting that later would silently invalidate them. Flex shrink is not an alternative either: the title button isflex-1, so its basis is 0, it never overflows, and shrink factors on the marker are inert. A truncated section title stays identifiable from the section's own contents and description; a truncated status marker does not.Two deliberate deviations awaiting design sign-off:
triangle-alert(matchingComfyQueueButton) in addition to the gold color. Color alone would not satisfy WCAG 1.4.1.#a88043→ 3.24:1,#8c6a38→ 4.47:1. Dark needs no change.Every blocker fix in this work was verified by mutation: the production line reverted, the test observed failing, restored, observed passing.