## Summary
Hide the template selector when a first-time cloud user accepts a shared
workflow from a share link, so the shared workflow opens without the
onboarding template dialog lingering.
## Changes
- **What**: Added shared-workflow loader behavior to close the global
template selector on accept actions (`copy-and-open` and `open-only`)
while keeping cancel behavior unchanged.
- **What**: Added targeted unit tests covering hide-on-accept and
no-hide-on-cancel behavior in the shared workflow URL loader.
## Review Focus
Confirm that share-link accept paths now dismiss the template selector
and that cancel still leaves it available.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9913-fix-hide-template-selector-after-shared-workflow-accept-3236d73d365081099c04e350d499fad2)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
fix: restore native copy/paste events for image paste support (#9914)
## Summary
- Remove Ctrl+C and Ctrl+V keybindings from the keybinding service
defaults so native browser copy/paste events fire
- This restores image paste into LoadImage nodes, which broke after
#9459
## Problem
PR #9459 moved Ctrl+C/V into the keybinding service, which calls
`event.preventDefault()` on keydown. This prevents the browser `paste`
event from firing, so `usePaste` (which detects images in the clipboard)
never runs. The `PasteFromClipboard` command only reads from
localStorage, completely bypassing image detection.
**Repro:** Copy a node → copy an image externally → try to paste the
image into a LoadImage node → gets old node data from localStorage
instead.
## Fix
Remove Ctrl+C and Ctrl+V from `CORE_KEYBINDINGS` in `defaults.ts`. The
native browser events now fire as before, and `useCopy`/`usePaste`
handle them correctly. Ctrl+Shift+V, Ctrl+A, Delete, and Backspace
keybindings remain in the keybinding service.
Fixes#9459 (regression)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9914-fix-restore-native-copy-paste-events-for-image-paste-support-3236d73d365081c7ac53f983f316e10f)
by [Unito](https://www.unito.io)
fix: clear stale widget slotMetadata on link disconnect (#9885)
## Summary
Fixes text field becoming non-editable when a previously linked input is
removed from a custom node.
## Problem
When a widget's input was promoted to a slot, connected via a link, and
then the input was removed (e.g., by updating the custom node
definition), the widget retained stale `slotMetadata` with `linked:
true`. This prevented the widget from being editable.
## Solution
In `refreshNodeSlots`, removed the `if (slotInfo)` guard so
`widget.slotMetadata` is always assigned — either to valid metadata or
`undefined`. This ensures stale linked state is cleared when inputs no
longer match widgets.
## Acceptance Criteria
1. Text field remains editable after promote→connect→disconnect cycle
2. Text field returns to editable state when noodle disconnected
3. No mode switching needed to restore editability
## Testing
- Added regression test: "clears stale slotMetadata when input no longer
matches widget"
- All existing tests pass (18/18 in affected file)
---
**Note: This PR currently contains only the RED (failing test) commit
for TDD verification. The GREEN (fix) commit will be pushed after CI
confirms the test failure.**
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9885-fix-clear-stale-widget-slotMetadata-on-link-disconnect-3226d73d365081269319c027b42d9f6b)
by [Unito](https://www.unito.io)
fix: stabilize subgraph promoted widget identity and rendering (#9896)
## Summary
Fix subgraph promoted widget identity/rendering so on-node widgets stay
correct through configure/hydration churn, duplicate names, and
linked+independent coexistence.
## Changes
- **Subgraph promotion reconciliation**: stabilize linked-entry identity
by subgraph slot id, preserve deterministic linked representative
selection, and prune stale alias/fallback entries without dropping
legitimate independent promotions.
- **Promoted view resolution**: bind slot mapping by promoted view
object identity (`getSlotFromWidget` / `getWidgetFromSlot`) to avoid
same-name collisions.
- **On-node widget rendering**: harden `NodeWidgets` identity and dedup
to avoid visual aliasing, prefer visible duplicates over hidden stale
entries, include type/source execution identity, and avoid collapsing
transient unresolved entries.
- **Mapping correctness**: update `useGraphNodeManager` promoted source
mapping to resolve by input target only when the promoted view is
actually bound to that input.
- **Subgraph input uniqueness**: ensure empty-slot promotion creates
unique input names (`seed`, `seed_1`, etc.) for same-name multi-source
promotions.
- **Safety fix**: guard against undefined canvas in slot-link
interaction.
- **Tests/fixtures**: add focused regressions for fixture path
`subgraph_complex_promotion_1`, linked+independent same-name cases,
duplicate-name identity mapping, dedup behavior, and input-name
uniqueness.
## Review Focus
Validate behavior around transient configure/hydration states (`-1` id
to concrete id), duplicate-name promotions, linked representative
recovery, and that dedup never hides legitimate widgets while still
removing true duplicates.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9896-fix-stabilize-subgraph-promoted-widget-identity-and-rendering-3226d73d365081c8a1e8d0a5a22e826d)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
1.42.5 (#9906)
Patch version increment to 1.42.5
**Base branch:** `main`
---------
Co-authored-by: christian-byrne <72887196+christian-byrne@users.noreply.github.com>
Co-authored-by: github-actions <github-actions@github.com>
fix: skip redundant appScalePercentage updates during zoom/pan (#9403)
## What
Add equality check before updating `appScalePercentage` reactive ref.
## Why
Firefox profiler shows 586 `setElementText` markers from continuous text
interpolation updates during zoom/pan. The rounded percentage value
often doesn't change between events.
## How
Extract `updateAppScalePercentage()` helper with equality guard —
compares new rounded value to current before assigning to the ref.
## Perf Impact
Expected: eliminates ~90% of `setElementText` markers during zoom/pan
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9403-fix-skip-redundant-appScalePercentage-updates-during-zoom-pan-31a6d73d3650812db8f2d68ac73c95b0)
by [Unito](https://www.unito.io)
test: add browser test for textarea right-click context menu in subgraph (#9891)
## Summary
Add E2E test coverage for the textarea widget right-click context menu
inside subgraphs.
The fix was shipped in #9840 — this PR adds the missing browser test.
## Test
- Loads a subgraph workflow with a CLIPTextEncode (textarea) node
- Navigates into the subgraph
- Right-clicks the textarea DOM element
- Asserts that the ComfyUI "Promote Widget" context menu option appears
## Related
- Fixes the test gap from #9840
- Notion ticket: d7a53160-e1e1-42bb-a5ac-c0c2702c629c
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9891-test-add-browser-test-for-textarea-right-click-context-menu-in-subgraph-3226d73d365081a4be51f89b5d505361)
by [Unito](https://www.unito.io)
feat: expand CDP perf metrics — add DOM nodes, script duration, event listeners (#9887)
## Summary
Expands the performance testing infrastructure to collect 4 additional
CDP metrics that are already returned by `Performance.getMetrics` but
were not being read. This is a zero-cost expansion — no additional CDP
calls, just reading more fields from the existing response.
## New Metrics
| Metric | CDP Source | What It Detects |
|---|---|---|
| `domNodes` | `Nodes` | DOM node count delta — widget DOM leaks during
node create/destroy |
| `jsHeapTotalBytes` | `JSHeapTotalSize` | Total heap delta — combined
with `heapDeltaBytes` shows GC pressure |
| `scriptDurationMs` | `ScriptDuration` | JS execution time vs total
task time — script vs rendering balance |
| `eventListeners` | `JSEventListeners` | Listener count delta — detects
listener accumulation across lifecycle |
## Changes
### `browser_tests/fixtures/helpers/PerformanceHelper.ts`
- Added 4 fields to `PerfSnapshot` interface
- Added 4 fields to `PerfMeasurement` interface
- Wired through `getSnapshot()` and `stopMeasuring()`
### `scripts/perf-report.ts`
- Added 4 fields to `PerfMeasurement` interface
- Expanded `MetricKey` type and `REPORTED_METRICS` array with 3 new
reported metrics (`domNodes`, `scriptDurationMs`, `eventListeners`)
- `jsHeapTotalBytes` is collected but not in `REPORTED_METRICS` — it's
used alongside `heapDeltaBytes` for GC pressure ratio analysis
## Why These 4
From a gap analysis of all ~30 CDP metrics, these were identified as
highest priority for ComfyUI:
- **`Nodes`** (P0): ComfyUI dynamically creates/destroys widget DOM. DOM
bloat from leaked widgets is a key performance risk, especially for Vue
Nodes 2.0.
- **`ScriptDuration`** (P1): Separates JS execution from layout/paint.
Reveals whether perf issues are script-heavy or rendering-heavy.
- **`JSEventListeners`** (P1): Widget lifecycle can leak listeners
across node add/remove cycles.
- **`JSHeapTotalSize`** (P1): With `JSHeapUsedSize`, the ratio shows GC
fragmentation pressure.
## Backward Compatibility
The `PerfMeasurement` interface is extended (not changed). Old baseline
`perf-metrics.json` files without these fields will have `undefined`
values, which the report script handles gracefully (shows `—` for
missing data).
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9887-feat-expand-CDP-perf-metrics-add-DOM-nodes-script-duration-event-listeners-3226d73d3650818abea1d4a441667c38)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
fix: prevent white flash when opening mask editor (#9860)
## Summary
- Remove hardcoded `bg-white` from mask editor canvas background div to
prevent white flash on dialog open
- Add a loading spinner while the mask editor initializes (image
loading, canvas setup, GPU resources)
- Background color is set dynamically by `setCanvasBackground()` after
initialization
Fixes#9852
### AS IS
https://github.com/user-attachments/assets/7da61e32-671b-4056-b5ec-8cb246fc7689
### TO BE
https://github.com/user-attachments/assets/bfdedc69-f690-42c5-8591-619623c04f55
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9860-fix-prevent-white-flash-when-opening-mask-editor-3226d73d365081de9b7ad4622438e6ed)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
fix: prevent live preview dimension flicker between frames (#9937)
## Summary
Fix "Calculating dimensions" text flickering during live sampling
preview in Vue renderer.
## Changes
- **What**: Stop resetting `actualDimensions` to `null` on every
`imageUrl` change. Previous dimensions are retained while the new frame
loads, eliminating the flicker. Error state is still reset correctly.
## Review Focus
The watcher on `props.imageUrl` previously reset both `actualDimensions`
and `imageError`. Now it only resets `imageError`, since
`handleImageLoad` updates dimensions when the new frame actually loads.
This means stale dimensions show briefly between frames, which is
intentionally better than showing "Calculating dimensions" text.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9937-fix-prevent-live-preview-dimension-flicker-between-frames-3246d73d36508154a676e5996112354f)
by [Unito](https://www.unito.io)
feat: make Vue nodes (Nodes 2.0) default for new desktop installs (#9947)
## What
Makes Vue nodes (Nodes 2.0) the default renderer for new desktop app
installs (version ≥1.41.0), matching the behavior already live for cloud
new installs.
## Why
Step 2 of the Nodes 2.0 rollout sequence:
1. ✅ Cloud new installs (≥1.41.0) — DONE
2. 👉 **Desktop app (new installs)** — this PR
3. ⬜ Local installs
4. ⬜ Remove Beta tag
5. ⬜ GTM announcement
No forced migration — only changes the default for new installs.
Existing users keep their setting. Rollback is a settings flip.
## Change
In `coreSettings.ts`, the `defaultsByInstallVersion` for
`Comfy.VueNodes.Enabled` changes from:
```typescript
defaultsByInstallVersion: { '1.41.0': isCloud },
```
to:
```typescript
defaultsByInstallVersion: { '1.41.0': isCloud || isDesktop },
```
## Gated on
- M2 perf target (≥52 FPS on 245-node workflow) — layer merge landed,
likely met
- M-DevRel migration docs (blocks Beta tag removal, not this flip)
Draft PR — ceremonial, to be merged when M2 checkpoint passes.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9947-feat-make-Vue-nodes-Nodes-2-0-default-for-new-desktop-installs-3246d73d365081b280dfff932c7aa016)
by [Unito](https://www.unito.io)
fix: fix perf CI pipeline — z-score baselines, force-push staleness, baseline storage (#9886)
## Summary
Fixes three critical issues with the CI performance reporting pipeline
that made perf reports useless on PRs (demonstrated by PR #9248 — deep
watcher removal merged without useful perf signal).
## Changes
### 1. Fix z-score baseline variance collection (`0/5 runs`)
**Root cause:** PR #9305 added z-score statistical analysis code to
`perf-report.ts`, but the historical data download step was placed in
the wrong workflow file. The report is generated in
`pr-perf-report.yaml` (a `workflow_run`-triggered job), but the
historical download was in `ci-perf-report.yaml` (the test runner) —
different runners, different filesystems.
**Fix:** Implement `perf-data` orphan branch storage:
- On push to main: save `perf-metrics.json` to `perf-data` branch with
timestamped filename
- On PR report: fetch last 5 baselines from `perf-data` branch into
`temp/perf-history/`
- Rolling window of 20 baselines, oldest pruned automatically
- Same pattern used by `github-action-benchmark` (33.7k repos)
### 2. Fix force-push comment staleness
**Root cause:** `cancel-in-progress: true` kills the perf test run
before it uploads artifacts. The downstream report workflow only
triggers on `conclusion == 'success'` — cancelled runs are ignored, so
the comment from the first successful run goes stale.
**Fix:**
- Change `cancel-in-progress: false` — with GitHub's queue depth of 1,
rapid pushes (A,B,C,D) run A and D, skipping B and C
- Add SHA validation in `pr-perf-report.yaml` — before posting, check if
the workflow_run's head SHA still matches the PR's current head. Skip
posting stale results.
### 3. Add permissions for baseline operations
- `contents: write` on CI job (needed for pushing to perf-data branch)
- `actions: read` on both workflows (needed for artifact/baseline
access)
## One-time setup required
After merging, create the `perf-data` orphan branch:
```bash
git checkout --orphan perf-data
git rm -rf .
echo '# Performance Baselines' > README.md
mkdir -p baselines
git add README.md baselines
git commit -m 'Initialize perf-data branch'
git push origin perf-data
```
The first 2 pushes to main after setup will build up variance data, and
z-scores will start appearing in PR reports (threshold is
`historical.length >= 2`).
## Testing
- YAML validated with `yaml.safe_load()`
- `perf-report.ts` `loadHistoricalReports()` already reads from
`temp/perf-history/<index>/perf-metrics.json` — no code changes needed
- All new steps use `continue-on-error: true` for graceful degradation
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9886-fix-fix-perf-CI-pipeline-z-score-baselines-force-push-staleness-baseline-storage-3226d73d365081538424c7945e71f308)
by [Unito](https://www.unito.io)
draft: add red-green-fix skill for verified bug fix workflow (#9954)
## Summary
- Add a Claude Code skill (`/red-green-fix`) that enforces the red-green
commit pattern for bug fixes
- Ensures a failing test is committed first (red CI), then the fix is
committed separately (green CI)
- Gives reviewers proof that the test actually catches the bug
- Includes `reference/testing-anti-patterns.md` with common mistakes
contextualized to this codebase
## Structure
```
.claude/skills/red-green-fix/
├── SKILL.md # Main skill definition
└── reference/
└── testing-anti-patterns.md # Anti-patterns guide
```
## Test Plan
- [ ] Invoke `/red-green-fix <bug description>` in Claude Code and
verify the two-step workflow
- [ ] Confirm PR template includes red-green verification table
- [ ] Review anti-patterns reference for completeness
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9954-draft-add-red-green-fix-skill-for-verified-bug-fix-workflow-3246d73d365081339a83dc09263b0f33)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
test: add large-graph perf test with 245-node workflow (backlog N5) (#9940)
## What
Adds a 245-node workflow asset and two `@perf` tests to establish a
baseline for large-graph performance regressions (Tier 6 in the
performance backlog).
## Why
Backlog item N5: we need CI regression detection for compositor layer
management, GPU texture count, and transform pane cost at 245+ nodes.
This is PR1 of 2 — establishes baseline metrics on main. Future
optimization PRs will show improvement deltas against this baseline.
## Tests Added
- **`large graph idle rendering`** — 120 frames idle with 245 nodes,
measures style recalcs, layouts, task duration, heap delta
- **`large graph pan interaction`** — middle-click pan across 245 nodes,
stresses compositor layer management and transform recalculation
## Workflow Asset
`browser_tests/assets/large-graph-workflow.json` — 245 nodes (49
pipelines of CheckpointLoader → 2× CLIPTextEncode → KSampler +
EmptyLatentImage), 294 links. Minimal structure focused on node count.
## Verification
- [x] `pnpm typecheck:browser` passes
- [x] `pnpm lint` passes (eslint on changed file)
- [x] All link references in JSON validated programmatically
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9940-test-add-large-graph-perf-test-with-245-node-workflow-backlog-N5-3246d73d365081f6b5d8ddb9a85e6ad0)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
feat: add Ingest API codegen with Zod schema generation (#9932)
## Summary
- Add `packages/ingest-types/` package that auto-generates TypeScript
types and Zod schemas from the Ingest service OpenAPI spec
- Uses `@hey-api/openapi-ts` with built-in Zod plugin (Zod v3
compatible)
- Filters out overlapping endpoints shared with the local ComfyUI Python
backend
- Generates **493 TypeScript types** and **256 Zod schemas** covering
cloud-only endpoints
- Configure knip to ignore generated files
## CI automation
The cloud repo pushes generated types to this repo (push model, no
private repo cloning).
See: Comfy-Org/cloud#2858
## How endpoint filtering works
Codegen targets are controlled by the **exclude list** in
`packages/ingest-types/openapi-ts.config.ts`. Everything in the Ingest
`openapi.yaml` is included **except** overlapping endpoints that also
exist in the local ComfyUI Python backend:
**Excluded (overlapping with ComfyUI Python):**
`/prompt`, `/queue`, `/history`, `/object_info`, `/features`,
`/settings`, `/system_stats`, `/interrupt`, `/upload/*`, `/view`,
`/jobs`, `/userdata`, `/webhooks/*`, `/internal/*`
**Included (cloud-only, codegen targets):**
`/workspaces/*`, `/billing/*`, `/secrets/*`, `/assets/*`, `/tasks/*`,
`/auth/*`, `/workflows/*`, `/workspace/*`, `/user`, `/settings/{key}`,
`/tags`, `/feedback`, `/invite_code/*`, `/experiment/models/*`,
`/global_subgraphs/*`
## Follow-up: replace manual types with generated ones
This PR only sets up the codegen infrastructure. A follow-up PR should
replace manually maintained types with imports from
`@comfyorg/ingest-types`:
| File | Lines | Current | Replace with |
|------|-------|---------|-------------|
| `src/platform/workspace/api/workspaceApi.ts` | ~270 | TS interfaces |
`import type { ... } from '@comfyorg/ingest-types'` |
| `src/platform/secrets/types.ts` | ~32 | TS interfaces | `import type {
... } from '@comfyorg/ingest-types'` |
| `src/platform/assets/schemas/assetSchema.ts` | ~125 | Zod schemas |
`import { ... } from '@comfyorg/ingest-types/zod'` |
| `src/platform/assets/schemas/mediaAssetSchema.ts` | ~50 | Zod schemas
| `import { ... } from '@comfyorg/ingest-types/zod'` |
| `src/platform/tasks/services/taskService.ts` | ~70 | Zod schemas |
`import { ... } from '@comfyorg/ingest-types/zod'` |
| `src/platform/workspace/workspaceTypes.ts` | ~6 | TS interface |
`export type { ... } from '@comfyorg/ingest-types'` |
## Test plan
- [x] `pnpm generate` in `packages/ingest-types/` produces
`types.gen.ts` and `zod.gen.ts`
- [x] `pnpm typecheck` passes
- [x] Pre-commit hooks pass (lint, typecheck, format)
- [x] Generated Zod schemas validate correct data and reject invalid
data
- [x] No import conflicts with existing code (generated types are
isolated in separate package)
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
feat: surface missing models in Error Tab for OSS and remove legacy dialog (#9921)
## Summary
- Surface missing models in the Error Tab for OSS environments,
replacing the legacy modal dialog
- Add Download button per model and Download All button in group header
with file size display
- Move download business logic from `components/dialog/content` to
`platform/missingModel`
- Remove legacy missing models dialog components and composable
## Changes
- **Pipeline**: Remove `isCloud` guard from `scanAllModelCandidates` and
`surfaceMissingModels` so OSS detects missing models
- **Grouping**: Group non-asset-supported models by directory in OSS
instead of lumping into UNSUPPORTED
- **UI**: Add Download button (matching Install Node Pack design) and
Download All header button
- **Store**: Add `folderPaths`/`fileSizes` state with setter methods,
race condition guard
- **Cleanup**: Delete `MissingModelsContent`, `MissingModelsHeader`,
`MissingModelsFooter`, `useMissingModelsDialog`, `missingModelsUtils`
- **Tests**: Add OSS/Cloud grouping tests, migrate Playwright E2E to
Error Tab, improve test isolation
- **Snapshots**: Reset Playwright screenshot expectations since OSS
missing model error detection now causes red highlights on affected
nodes
- **Accessibility**: Add `aria-label` with model name, `aria-expanded`
on toggle, warning icon for unknown category
## Test plan
- [x] Unit tests pass (86 tests)
- [x] TypeScript typecheck passes
- [x] knip passes
- [x] Load workflow with missing models in OSS → Error Tab shows missing
models grouped by directory
- [x] Download button triggers browser download with correct URL
- [x] Download All button downloads all downloadable models
- [x] Cloud environment behavior unchanged
- [x] Playwright E2E: `pnpm test:browser:local -- --grep "Missing models
in Error Tab"`
## Screenshots
https://github.com/user-attachments/assets/12f15e09-215a-4c58-87ed-39bbffd1359c
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9921-feat-surface-missing-models-in-Error-Tab-for-OSS-and-remove-legacy-dialog-3236d73d365081f0a9dfc291978f5ecf)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: github-actions <github-actions@github.com>
fix: cloud subscribe redirect hangs waiting for billing init (#9965)
## Summary
Fix /cloud/subscribe route hanging indefinitely because billing context
never initializes during the onboarding flow.
## Changes
- **What**: Replace passive `await until(isInitialized).toBe(true)` with
explicit `await initialize()` in CloudSubscriptionRedirectView. Remove
unused `until` import.

## Review Focus
In the onboarding flow, `useTeamWorkspaceStore().activeWorkspace` is not
set, so `useBillingContext`'s internal watch (which triggers
`initialize()` on workspace change) enters the `!newWorkspaceId` branch
— it resets `isInitialized` to `false` and returns without ever calling
`initialize()`. The old code then awaited `isInitialized` becoming
`true` forever.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9965-fix-cloud-subscribe-redirect-hangs-waiting-for-billing-init-3246d73d3650812d93ebd477c544fa0a)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
feat: add TBT/frameDuration metrics and new perf test scenarios (#9910)
## Summary
Adds Total Blocking Time (TBT) and frame duration metrics to the
performance testing infrastructure, plus three new test scenarios
covering zoom, pan, and many-nodes-idle.
## Changes
### New Metrics
- **`totalBlockingTimeMs`** — Computed from PerformanceObserver
`longtask` entries: `sum(duration - 50ms)` for tasks >50ms. Measures
main thread blocking.
- **`frameDurationMs`** — Average frame duration via rAF timing (16.67ms
= 60fps target). Measures rendering smoothness.
### New Test Scenarios
| Scenario | Description |
|---|---|
| `canvas-zoom-sweep` | 10 zoom-in + 10 zoom-out cycles on default
workflow |
| `canvas-pan-many-nodes` | 10 pan sweeps over 100-node workflow |
| `canvas-many-nodes-idle` | 2-second idle measurement with 100 nodes
rendered |
### Infrastructure
- `PerformanceHelper.ts`: Installs PerformanceObserver for longtask,
collects TBT, measures frame duration via rAF
- `perf-report.ts`: Reports TBT and frame duration in PR comment tables
- `browser_tests/assets/perf/many_nodes_100.json`: 100-node (10×10 grid)
test fixture
## Review Focus
- TBT collection clears entries at `startMeasuring()` and reads at
`stopMeasuring()` — ensure no race with observer buffering
- Frame duration sampling uses 10 frames — enough for signal without
slowing tests
Depends on: #9886, #9887
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9910-feat-add-TBT-frameDuration-metrics-and-new-perf-test-scenarios-3236d73d365081488ae3c594a8bf7cff)
by [Unito](https://www.unito.io)
fix: LGraphGroup paste position (#9962)
## Summary
Fix group paste position: groups now paste at the cursor location
instead of on top of the original.
## Changes
- **What**: Added LGraphGroup offset handling in _deserializeItems()
position adjustment loop, matching existing LGraphNode and Reroute
behavior.
## Screenshots
Before:
https://github.com/user-attachments/assets/e317af10-8009-4092-9d14-de79316cd853
After:
https://github.com/user-attachments/assets/f4ffefd5-519a-4592-812c-c88e3b5940fd
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9962-fix-LGraphGroup-paste-position-3246d73d365081eea5b2e2507da861de)
by [Unito](https://www.unito.io)
fix: tree explorer nodes not filling parent container width (#9964)
## Summary
Fix tree explorer nodes not filling the full width of the sidebar
container, causing text to overflow instead of truncating.
## Changes
- **What**: Add `min-w-0` to `TreeRoot` to allow flex shrinking within
sidebar. Add `w-full` and `min-w-0` to tree node rows so
absolutely-positioned virtualizer items fill the container width and
text truncates correctly.
<img width="365" height="749" alt="image"
src="https://github.com/user-attachments/assets/320910f3-52ad-4634-a935-6bd1a40aea7f"
/>
## Review Focus
The virtualizer renders each item with `position: absolute; left: 0` but
no explicit width, so rows would size to content rather than filling the
container. Adding `w-full` ensures rows stretch to 100% of the
virtualizer container, and `min-w-0` allows proper flex shrinking for
deep indentation levels.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9964-fix-tree-explorer-nodes-not-filling-parent-container-width-3246d73d36508138be38fdcac15ae4ef)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
feat: add Copy URL button to missing model rows for OSS (#9966)
1.42.6 (#9986)
Patch version increment to 1.42.6
**Base branch:** `main`
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9986-1-42-6-3256d73d365081a28bfad82022ce3440)
by [Unito](https://www.unito.io)
---------
Co-authored-by: christian-byrne <72887196+christian-byrne@users.noreply.github.com>
Co-authored-by: github-actions <github-actions@github.com>
fix: block missing e2e regression coverage in CodeRabbit (#9987)
## Summary
Make the CodeRabbit end-to-end regression coverage check actually block
fix-like PRs until it is resolved or explicitly overridden by a
requested reviewer, and harden the prompt so it evaluates only PR-local
metadata.
## Changes
- **What**: Set the `End-to-end regression coverage for fixes` custom
check mode from `warning` to `error`
- **What**: Enable `reviews.request_changes_workflow` so CodeRabbit can
block on failed `error` pre-merge checks
- **What**: Set
`reviews.pre_merge_checks.override_requested_reviewers_only` to `true`
so only requested reviewers can bypass a failed check
- **What**: Tighten the custom check instructions to use only PR
metadata in review context, avoid shell commands, and avoid reverse-diff
or base-branch file evaluation
## Review Focus
Confirm this is the intended CodeRabbit enforcement model for missing
Playwright regression coverage on fix-like PRs and that the prompt
wording is strict enough to avoid false positives from reversed diffs.
fix: add reve and elevenlabs to icon safelist (#9990)
## Summary
Reve and ElevenLabs provider icons were not displaying in the node
library because they were missing from the Tailwind icon safelist.
## Changes
- **What**: Add `reve` and `elevenlabs` to the `@source inline` safelist
in `style.css` so `icon-[comfy--reve]` and `icon-[comfy--elevenlabs]`
classes are generated. Add corresponding `PROVIDER_COLORS` entries in
`categoryUtil.ts`.
<img width="308" height="106" alt="image"
src="https://github.com/user-attachments/assets/d488898a-fbad-4af0-8921-0e8ee7d4705a"
/>
<img width="308" height="78" alt="image"
src="https://github.com/user-attachments/assets/2b3b7172-095b-415e-a49a-d303977e0abc"
/>
## Review Focus
The SVG files already existed in `packages/design-system/src/icons/` but
Tailwind's tree-shaking dropped the classes since they're only used
dynamically via `getProviderIcon()`.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9990-fix-add-reve-and-elevenlabs-to-icon-safelist-3256d73d36508105994fcdd5d0568027)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
fix: mask editor save shows blank image in Load Image node (#9984)
## Summary
Mask editor save was showing a blank image in the Load Image node
(legacy nodes mode, not Nodes 2.0) because
`updateNodeWithServerReferences` called `updateNodeImages`, which
silently no-ops when the node has no pre-existing execution outputs.
Replaced with `setNodeOutputs` which properly creates output entries
regardless of prior state.
**Affects:** Legacy nodes mode only. Nodes 2.0 (Vue Nodes) renders
images via Vue components and is not affected.
- Fixes#9983
- Fixes#9782
- Fixes#9952
## Red-Green Verification
| Commit | SHA | CI Status | Run | Purpose |
|--------|-----|-----------|-----|---------|
| `test: add failing test for mask editor save showing blank image` |
`0ab66e8` | 🔴
[Red](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/23125427860)
| CI: Tests Unit **failure** | Proves the test catches the bug |
| `fix: mask editor save shows blank image in Load Image node` |
`564cc9c` | 🟢
[Green](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/23127289891)
| CI: Tests Unit **success** | Proves the fix resolves the bug |
## manual testing
### as is
https://github.com/user-attachments/assets/8d5c36ce-2c5e-4609-b246-dcf896c4a8e7
### to be
https://github.com/user-attachments/assets/c8ae4f0e-3da0-40f2-a543-d1d5a6bce795
## Test Plan
- [x] CI red on test-only commit
- [x] CI green on fix commit
- [ ] E2E regression test not added: mask editor save requires canvas
pixel manipulation + server upload round-trip which is covered by the
existing unit test mocking the full `save()` flow. The Playwright test
infrastructure does not currently support mask editor interactions (draw
+ save).
- [x] Manual verification (legacy nodes mode): Load Image → upload →
mask editor → draw → save → verify image refreshes
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
fix: allow URL input for free tier users, gate on import button (#10024)
## Summary
- Remove free-tier restriction from the URL input field in
`MissingModelUrlInput.vue` so it is always editable
- Move the subscription check (`canImportModels`) to the Import button
click handler — free-tier users see the upgrade modal only when they
attempt to import
- Extract inline ternary to named `handleImportClick` method for clarity
## Test plan
- [x] Unit tests added (`MissingModelUrlInput.test.ts`) verifying:
- URL input is always editable regardless of subscription tier
- Import button calls `handleImport` for paid users
- Import button calls `showUploadDialog` (upgrade modal) for free-tier
users
- [x] Verify URL input is editable for free-tier users on cloud
- [x] Verify clicking Import as free-tier opens the subscription modal
- [x] Verify paid users can import normally without changes
## E2E test rationale
Playwright E2E regression tests are impractical for this change because
`MissingModelUrlInput` only renders when `isAssetSupported` is true,
which requires `isCloud` — a compile-time constant (`__DISTRIBUTION__`).
The OSS test build always sets `isCloud = false`, so the component never
renders in the E2E environment. Unit tests with mocked feature flags
provide equivalent behavioral coverage.
fix: prevent subscription UI from rendering on non-cloud distributions (#9958)
## Summary
Prevent Plans & Pricing dialog, subscription buttons, and cloud-only
menu items from appearing on desktop/localhost distributions.
## Changes
- **What**: Add `isCloud` guards to
`useSubscriptionDialog.showPricingTable`, `TopbarSubscribeButton`, and
`CurrentUserPopoverLegacy` so subscription UI only renders on cloud
- **Tests**: 24 tests across 3 test files (1 modified, 2 new) covering
cloud/non-cloud behavior
## Review Focus
- Guard placement in `CurrentUserPopoverLegacy.vue` — multiple `v-if`
conditions updated to include `isCloud`
- Early-return in `showPricingTable` as a defense-in-depth measure
Fixes COM-16820
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9958-fix-prevent-subscription-UI-from-rendering-on-non-cloud-distributions-3246d73d365081559a9ee8650409c5b4)
by [Unito](https://www.unito.io)
Co-authored-by: Benjamin Lu <benjaminlu1107@gmail.com>
fix: prevent animated preview duplication on Vue↔Litegraph switch (#9938)
## Problem
SaveAnimatedPNG/WEBP nodes show duplicate output previews when switching
between Vue and Litegraph renderer modes.
## Root Cause
The `ANIM_PREVIEW_WIDGET` (`$$comfy_animation_preview`) DOM widget
lacked `canvasOnly: true`, so `shouldRenderAsVue()` in the widget
registry included it in Vue mode rendering. This caused both:
1. Vue's `ImagePreview.vue` (via `nodeMedia` computed from
`nodeOutputStore`)
2. The legacy `ANIM_PREVIEW_WIDGET` DOM widget (rendered as `WidgetDOM`)
to display simultaneously — duplicating the output preview.
## Fix
Add `canvasOnly: true` to the `ANIM_PREVIEW_WIDGET` options, matching
the pattern used by `IMAGE_PREVIEW` widget in
`useImagePreviewWidget.ts`. This ensures the legacy widget is filtered
out in Vue mode by `shouldRenderAsVue()`, leaving `ImagePreview.vue` as
the single source of truth.
## Testing
- All 539 vueNodes tests pass
- All 22 nodeOutputStore tests pass
- All 140 composables/node tests pass
- Typecheck passes
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9938-fix-prevent-animated-preview-duplication-on-Vue-Litegraph-switch-3246d73d365081019bbfd7e33a9c14fb)
by [Unito](https://www.unito.io)
1.43.0 (#10032)
Minor version increment to 1.43.0
**Base branch:** `main`
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-10032-1-43-0-3256d73d3650818e8408d25fdf28de48)
by [Unito](https://www.unito.io)
Co-authored-by: christian-byrne <72887196+christian-byrne@users.noreply.github.com>
Feat/3d thumbnail inline rendering (#9471)
## Summary
The previous approach generated thumbnails server-side and uploaded them
as `model.glb.png` alongside the model file. This breaks on cloud
deployments where output files are renamed to content hashes, severing
the filename-based association between a model and its thumbnail.
Replace the server-upload approach with client-side Three.js rendering
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9471-Feat-3d-thumbnail-inline-rendering-31b6d73d3650816fbd7dd05b507aa80d)
by [Unito](https://www.unito.io)
test: add FeatureFlagHelper and QueueHelper for E2E test infrastructure (#9554)
## Summary
Add 2 reusable test helpers for Playwright E2E tests, integrated into
the ComfyPage fixture. These provide standardized patterns for mocking
feature flags and queue state across all E2E tests.
## Changes
- **`FeatureFlagHelper.ts`** — manage localStorage `ff:` prefixed
feature flags (`seedFlags` for init-time, `setFlags` for runtime) and
mock `/api/features` route
- **`QueueHelper.ts`** — mock `/api/queue` and `/api/history` routes
with configurable running/pending counts and success/error job entries
- **`ComfyPage.ts`** — integrate both helpers as
`comfyPage.featureFlags` and `comfyPage.queue`
## Review Focus
- Helper API design: are `seedFlags`/`setFlags`/`mockServerFeatures` the
right abstractions for feature flag testing?
- Queue mock fidelity: does the mock history shape match real ComfyUI
API responses closely enough?
- These are test-only infrastructure — no production code changes.
## Stack
This is the base PR for the Playwright E2E coverage stack. Waves 1-4 all
branch from this and can merge independently once this lands:
- **→ This PR**: Test infrastructure helpers
- #9555: Toasts, error overlay, selection toolbox, linear mode,
selection rectangle
- #9556: Node search, bottom panel, focus mode, job history, side panel
- #9557: Errors tab, node headers, queue notifications, settings sidebar
- #9558: Minimap, widget copy, floating menus, node library essentials
---------
Co-authored-by: GitHub Action <action@github.com>
feat: scaffold Astro 5 website app + design-system base.css
- Create apps/website/ with Astro 5, Vue 3, Tailwind v4 integration
- Static output, assetsPrefix /_website/, i18n (en + zh-CN)
- Nx targets: dev, serve, build, preview, typecheck
- Add base.css to design-system: brand tokens + Inter font-face only
- Add catalog entries: astro, @astrojs/vue, @astrojs/check, nanostores, @nanostores/vue
scaffold-01, scaffold-02
fix: add .gitignore and env.d.ts for Astro website app
feat: add layout shell — SEO head, analytics, nav, footer
- BaseLayout: OG/Twitter meta, canonical URL, GA4 GTM-NP9JM6K7,
Vercel Analytics, ClientRouter for SPA navigation
- SiteNav: Comfy logo, Enterprise/Gallery/About/Careers links,
Comfy Cloud + Comfy Hub CTA buttons, mobile hamburger menu
- SiteFooter: Product/Resources/Company/Legal columns,
social icons (GitHub, Discord, X, Reddit, LinkedIn, Instagram)
- Add @vercel/analytics to workspace catalog and website deps
fix: address CodeRabbit review — ARIA wiring, absolute OG URLs, Analytics component
- SiteNav: add aria-controls, aria-expanded, and id for mobile menu
- BaseLayout: use absolute URLs for og:image and twitter:image
- BaseLayout: replace inline inject() with official <Analytics /> component
style: apply oxfmt formatting
fix: remove unused deps from website package.json (knip)
fix: clean up unused catalog entries from pnpm-workspace.yaml
feat: add Wave 3 homepage sections (hero, social proof, pillars, testimonials, CTAs, manifesto, academy, placeholders)
Hide the template selector when a first-time cloud user accepts a shared
workflow from a share link, so the shared workflow opens without the
onboarding template dialog lingering.
- **What**: Added shared-workflow loader behavior to close the global
template selector on accept actions (`copy-and-open` and `open-only`)
while keeping cancel behavior unchanged.
- **What**: Added targeted unit tests covering hide-on-accept and
no-hide-on-cancel behavior in the shared workflow URL loader.
Confirm that share-link accept paths now dismiss the template selector
and that cancel still leaves it available.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9913-fix-hide-template-selector-after-shared-workflow-accept-3236d73d365081099c04e350d499fad2)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
fix: restore native copy/paste events for image paste support (#9914)
- Remove Ctrl+C and Ctrl+V keybindings from the keybinding service
defaults so native browser copy/paste events fire
- This restores image paste into LoadImage nodes, which broke after
PR #9459 moved Ctrl+C/V into the keybinding service, which calls
`event.preventDefault()` on keydown. This prevents the browser `paste`
event from firing, so `usePaste` (which detects images in the clipboard)
never runs. The `PasteFromClipboard` command only reads from
localStorage, completely bypassing image detection.
**Repro:** Copy a node → copy an image externally → try to paste the
image into a LoadImage node → gets old node data from localStorage
instead.
Remove Ctrl+C and Ctrl+V from `CORE_KEYBINDINGS` in `defaults.ts`. The
native browser events now fire as before, and `useCopy`/`usePaste`
handle them correctly. Ctrl+Shift+V, Ctrl+A, Delete, and Backspace
keybindings remain in the keybinding service.
Fixes#9459 (regression)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9914-fix-restore-native-copy-paste-events-for-image-paste-support-3236d73d365081c7ac53f983f316e10f)
by [Unito](https://www.unito.io)
fix: clear stale widget slotMetadata on link disconnect (#9885)
Fixes text field becoming non-editable when a previously linked input is
removed from a custom node.
When a widget's input was promoted to a slot, connected via a link, and
then the input was removed (e.g., by updating the custom node
definition), the widget retained stale `slotMetadata` with `linked:
true`. This prevented the widget from being editable.
In `refreshNodeSlots`, removed the `if (slotInfo)` guard so
`widget.slotMetadata` is always assigned — either to valid metadata or
`undefined`. This ensures stale linked state is cleared when inputs no
longer match widgets.
1. Text field remains editable after promote→connect→disconnect cycle
2. Text field returns to editable state when noodle disconnected
3. No mode switching needed to restore editability
- Added regression test: "clears stale slotMetadata when input no longer
matches widget"
- All existing tests pass (18/18 in affected file)
---
**Note: This PR currently contains only the RED (failing test) commit
for TDD verification. The GREEN (fix) commit will be pushed after CI
confirms the test failure.**
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9885-fix-clear-stale-widget-slotMetadata-on-link-disconnect-3226d73d365081269319c027b42d9f6b)
by [Unito](https://www.unito.io)
fix: stabilize subgraph promoted widget identity and rendering (#9896)
Fix subgraph promoted widget identity/rendering so on-node widgets stay
correct through configure/hydration churn, duplicate names, and
linked+independent coexistence.
- **Subgraph promotion reconciliation**: stabilize linked-entry identity
by subgraph slot id, preserve deterministic linked representative
selection, and prune stale alias/fallback entries without dropping
legitimate independent promotions.
- **Promoted view resolution**: bind slot mapping by promoted view
object identity (`getSlotFromWidget` / `getWidgetFromSlot`) to avoid
same-name collisions.
- **On-node widget rendering**: harden `NodeWidgets` identity and dedup
to avoid visual aliasing, prefer visible duplicates over hidden stale
entries, include type/source execution identity, and avoid collapsing
transient unresolved entries.
- **Mapping correctness**: update `useGraphNodeManager` promoted source
mapping to resolve by input target only when the promoted view is
actually bound to that input.
- **Subgraph input uniqueness**: ensure empty-slot promotion creates
unique input names (`seed`, `seed_1`, etc.) for same-name multi-source
promotions.
- **Safety fix**: guard against undefined canvas in slot-link
interaction.
- **Tests/fixtures**: add focused regressions for fixture path
`subgraph_complex_promotion_1`, linked+independent same-name cases,
duplicate-name identity mapping, dedup behavior, and input-name
uniqueness.
Validate behavior around transient configure/hydration states (`-1` id
to concrete id), duplicate-name promotions, linked representative
recovery, and that dedup never hides legitimate widgets while still
removing true duplicates.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9896-fix-stabilize-subgraph-promoted-widget-identity-and-rendering-3226d73d365081c8a1e8d0a5a22e826d)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
1.42.5 (#9906)
Patch version increment to 1.42.5
**Base branch:** `main`
---------
Co-authored-by: christian-byrne <72887196+christian-byrne@users.noreply.github.com>
Co-authored-by: github-actions <github-actions@github.com>
fix: skip redundant appScalePercentage updates during zoom/pan (#9403)
Add equality check before updating `appScalePercentage` reactive ref.
Firefox profiler shows 586 `setElementText` markers from continuous text
interpolation updates during zoom/pan. The rounded percentage value
often doesn't change between events.
Extract `updateAppScalePercentage()` helper with equality guard —
compares new rounded value to current before assigning to the ref.
Expected: eliminates ~90% of `setElementText` markers during zoom/pan
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9403-fix-skip-redundant-appScalePercentage-updates-during-zoom-pan-31a6d73d3650812db8f2d68ac73c95b0)
by [Unito](https://www.unito.io)
test: add browser test for textarea right-click context menu in subgraph (#9891)
Add E2E test coverage for the textarea widget right-click context menu
inside subgraphs.
The fix was shipped in #9840 — this PR adds the missing browser test.
- Loads a subgraph workflow with a CLIPTextEncode (textarea) node
- Navigates into the subgraph
- Right-clicks the textarea DOM element
- Asserts that the ComfyUI "Promote Widget" context menu option appears
- Fixes the test gap from #9840
- Notion ticket: d7a53160-e1e1-42bb-a5ac-c0c2702c629c
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9891-test-add-browser-test-for-textarea-right-click-context-menu-in-subgraph-3226d73d365081a4be51f89b5d505361)
by [Unito](https://www.unito.io)
feat: expand CDP perf metrics — add DOM nodes, script duration, event listeners (#9887)
Expands the performance testing infrastructure to collect 4 additional
CDP metrics that are already returned by `Performance.getMetrics` but
were not being read. This is a zero-cost expansion — no additional CDP
calls, just reading more fields from the existing response.
| Metric | CDP Source | What It Detects |
|---|---|---|
| `domNodes` | `Nodes` | DOM node count delta — widget DOM leaks during
node create/destroy |
| `jsHeapTotalBytes` | `JSHeapTotalSize` | Total heap delta — combined
with `heapDeltaBytes` shows GC pressure |
| `scriptDurationMs` | `ScriptDuration` | JS execution time vs total
task time — script vs rendering balance |
| `eventListeners` | `JSEventListeners` | Listener count delta — detects
listener accumulation across lifecycle |
- Added 4 fields to `PerfSnapshot` interface
- Added 4 fields to `PerfMeasurement` interface
- Wired through `getSnapshot()` and `stopMeasuring()`
- Added 4 fields to `PerfMeasurement` interface
- Expanded `MetricKey` type and `REPORTED_METRICS` array with 3 new
reported metrics (`domNodes`, `scriptDurationMs`, `eventListeners`)
- `jsHeapTotalBytes` is collected but not in `REPORTED_METRICS` — it's
used alongside `heapDeltaBytes` for GC pressure ratio analysis
From a gap analysis of all ~30 CDP metrics, these were identified as
highest priority for ComfyUI:
- **`Nodes`** (P0): ComfyUI dynamically creates/destroys widget DOM. DOM
bloat from leaked widgets is a key performance risk, especially for Vue
Nodes 2.0.
- **`ScriptDuration`** (P1): Separates JS execution from layout/paint.
Reveals whether perf issues are script-heavy or rendering-heavy.
- **`JSEventListeners`** (P1): Widget lifecycle can leak listeners
across node add/remove cycles.
- **`JSHeapTotalSize`** (P1): With `JSHeapUsedSize`, the ratio shows GC
fragmentation pressure.
The `PerfMeasurement` interface is extended (not changed). Old baseline
`perf-metrics.json` files without these fields will have `undefined`
values, which the report script handles gracefully (shows `—` for
missing data).
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9887-feat-expand-CDP-perf-metrics-add-DOM-nodes-script-duration-event-listeners-3226d73d3650818abea1d4a441667c38)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
fix: prevent white flash when opening mask editor (#9860)
- Remove hardcoded `bg-white` from mask editor canvas background div to
prevent white flash on dialog open
- Add a loading spinner while the mask editor initializes (image
loading, canvas setup, GPU resources)
- Background color is set dynamically by `setCanvasBackground()` after
initialization
Fixes#9852https://github.com/user-attachments/assets/7da61e32-671b-4056-b5ec-8cb246fc7689https://github.com/user-attachments/assets/bfdedc69-f690-42c5-8591-619623c04f55
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9860-fix-prevent-white-flash-when-opening-mask-editor-3226d73d365081de9b7ad4622438e6ed)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
fix: prevent live preview dimension flicker between frames (#9937)
Fix "Calculating dimensions" text flickering during live sampling
preview in Vue renderer.
- **What**: Stop resetting `actualDimensions` to `null` on every
`imageUrl` change. Previous dimensions are retained while the new frame
loads, eliminating the flicker. Error state is still reset correctly.
The watcher on `props.imageUrl` previously reset both `actualDimensions`
and `imageError`. Now it only resets `imageError`, since
`handleImageLoad` updates dimensions when the new frame actually loads.
This means stale dimensions show briefly between frames, which is
intentionally better than showing "Calculating dimensions" text.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9937-fix-prevent-live-preview-dimension-flicker-between-frames-3246d73d36508154a676e5996112354f)
by [Unito](https://www.unito.io)
feat: make Vue nodes (Nodes 2.0) default for new desktop installs (#9947)
Makes Vue nodes (Nodes 2.0) the default renderer for new desktop app
installs (version ≥1.41.0), matching the behavior already live for cloud
new installs.
Step 2 of the Nodes 2.0 rollout sequence:
1. ✅ Cloud new installs (≥1.41.0) — DONE
2. 👉 **Desktop app (new installs)** — this PR
3. ⬜ Local installs
4. ⬜ Remove Beta tag
5. ⬜ GTM announcement
No forced migration — only changes the default for new installs.
Existing users keep their setting. Rollback is a settings flip.
In `coreSettings.ts`, the `defaultsByInstallVersion` for
`Comfy.VueNodes.Enabled` changes from:
```typescript
defaultsByInstallVersion: { '1.41.0': isCloud },
```
to:
```typescript
defaultsByInstallVersion: { '1.41.0': isCloud || isDesktop },
```
- M2 perf target (≥52 FPS on 245-node workflow) — layer merge landed,
likely met
- M-DevRel migration docs (blocks Beta tag removal, not this flip)
Draft PR — ceremonial, to be merged when M2 checkpoint passes.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9947-feat-make-Vue-nodes-Nodes-2-0-default-for-new-desktop-installs-3246d73d365081b280dfff932c7aa016)
by [Unito](https://www.unito.io)
fix: fix perf CI pipeline — z-score baselines, force-push staleness, baseline storage (#9886)
Fixes three critical issues with the CI performance reporting pipeline
that made perf reports useless on PRs (demonstrated by PR #9248 — deep
watcher removal merged without useful perf signal).
**Root cause:** PR #9305 added z-score statistical analysis code to
`perf-report.ts`, but the historical data download step was placed in
the wrong workflow file. The report is generated in
`pr-perf-report.yaml` (a `workflow_run`-triggered job), but the
historical download was in `ci-perf-report.yaml` (the test runner) —
different runners, different filesystems.
**Fix:** Implement `perf-data` orphan branch storage:
- On push to main: save `perf-metrics.json` to `perf-data` branch with
timestamped filename
- On PR report: fetch last 5 baselines from `perf-data` branch into
`temp/perf-history/`
- Rolling window of 20 baselines, oldest pruned automatically
- Same pattern used by `github-action-benchmark` (33.7k repos)
**Root cause:** `cancel-in-progress: true` kills the perf test run
before it uploads artifacts. The downstream report workflow only
triggers on `conclusion == 'success'` — cancelled runs are ignored, so
the comment from the first successful run goes stale.
**Fix:**
- Change `cancel-in-progress: false` — with GitHub's queue depth of 1,
rapid pushes (A,B,C,D) run A and D, skipping B and C
- Add SHA validation in `pr-perf-report.yaml` — before posting, check if
the workflow_run's head SHA still matches the PR's current head. Skip
posting stale results.
- `contents: write` on CI job (needed for pushing to perf-data branch)
- `actions: read` on both workflows (needed for artifact/baseline
access)
After merging, create the `perf-data` orphan branch:
```bash
git checkout --orphan perf-data
git rm -rf .
echo '# Performance Baselines' > README.md
mkdir -p baselines
git add README.md baselines
git commit -m 'Initialize perf-data branch'
git push origin perf-data
```
The first 2 pushes to main after setup will build up variance data, and
z-scores will start appearing in PR reports (threshold is
`historical.length >= 2`).
- YAML validated with `yaml.safe_load()`
- `perf-report.ts` `loadHistoricalReports()` already reads from
`temp/perf-history/<index>/perf-metrics.json` — no code changes needed
- All new steps use `continue-on-error: true` for graceful degradation
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9886-fix-fix-perf-CI-pipeline-z-score-baselines-force-push-staleness-baseline-storage-3226d73d365081538424c7945e71f308)
by [Unito](https://www.unito.io)
draft: add red-green-fix skill for verified bug fix workflow (#9954)
- Add a Claude Code skill (`/red-green-fix`) that enforces the red-green
commit pattern for bug fixes
- Ensures a failing test is committed first (red CI), then the fix is
committed separately (green CI)
- Gives reviewers proof that the test actually catches the bug
- Includes `reference/testing-anti-patterns.md` with common mistakes
contextualized to this codebase
```
.claude/skills/red-green-fix/
├── SKILL.md # Main skill definition
└── reference/
└── testing-anti-patterns.md # Anti-patterns guide
```
- [ ] Invoke `/red-green-fix <bug description>` in Claude Code and
verify the two-step workflow
- [ ] Confirm PR template includes red-green verification table
- [ ] Review anti-patterns reference for completeness
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9954-draft-add-red-green-fix-skill-for-verified-bug-fix-workflow-3246d73d365081339a83dc09263b0f33)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
test: add large-graph perf test with 245-node workflow (backlog N5) (#9940)
Adds a 245-node workflow asset and two `@perf` tests to establish a
baseline for large-graph performance regressions (Tier 6 in the
performance backlog).
Backlog item N5: we need CI regression detection for compositor layer
management, GPU texture count, and transform pane cost at 245+ nodes.
This is PR1 of 2 — establishes baseline metrics on main. Future
optimization PRs will show improvement deltas against this baseline.
- **`large graph idle rendering`** — 120 frames idle with 245 nodes,
measures style recalcs, layouts, task duration, heap delta
- **`large graph pan interaction`** — middle-click pan across 245 nodes,
stresses compositor layer management and transform recalculation
`browser_tests/assets/large-graph-workflow.json` — 245 nodes (49
pipelines of CheckpointLoader → 2× CLIPTextEncode → KSampler +
EmptyLatentImage), 294 links. Minimal structure focused on node count.
- [x] `pnpm typecheck:browser` passes
- [x] `pnpm lint` passes (eslint on changed file)
- [x] All link references in JSON validated programmatically
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9940-test-add-large-graph-perf-test-with-245-node-workflow-backlog-N5-3246d73d365081f6b5d8ddb9a85e6ad0)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
feat: add Ingest API codegen with Zod schema generation (#9932)
- Add `packages/ingest-types/` package that auto-generates TypeScript
types and Zod schemas from the Ingest service OpenAPI spec
- Uses `@hey-api/openapi-ts` with built-in Zod plugin (Zod v3
compatible)
- Filters out overlapping endpoints shared with the local ComfyUI Python
backend
- Generates **493 TypeScript types** and **256 Zod schemas** covering
cloud-only endpoints
- Configure knip to ignore generated files
The cloud repo pushes generated types to this repo (push model, no
private repo cloning).
See: Comfy-Org/cloud#2858
Codegen targets are controlled by the **exclude list** in
`packages/ingest-types/openapi-ts.config.ts`. Everything in the Ingest
`openapi.yaml` is included **except** overlapping endpoints that also
exist in the local ComfyUI Python backend:
**Excluded (overlapping with ComfyUI Python):**
`/prompt`, `/queue`, `/history`, `/object_info`, `/features`,
`/settings`, `/system_stats`, `/interrupt`, `/upload/*`, `/view`,
`/jobs`, `/userdata`, `/webhooks/*`, `/internal/*`
**Included (cloud-only, codegen targets):**
`/workspaces/*`, `/billing/*`, `/secrets/*`, `/assets/*`, `/tasks/*`,
`/auth/*`, `/workflows/*`, `/workspace/*`, `/user`, `/settings/{key}`,
`/tags`, `/feedback`, `/invite_code/*`, `/experiment/models/*`,
`/global_subgraphs/*`
This PR only sets up the codegen infrastructure. A follow-up PR should
replace manually maintained types with imports from
`@comfyorg/ingest-types`:
| File | Lines | Current | Replace with |
|------|-------|---------|-------------|
| `src/platform/workspace/api/workspaceApi.ts` | ~270 | TS interfaces |
`import type { ... } from '@comfyorg/ingest-types'` |
| `src/platform/secrets/types.ts` | ~32 | TS interfaces | `import type {
... } from '@comfyorg/ingest-types'` |
| `src/platform/assets/schemas/assetSchema.ts` | ~125 | Zod schemas |
`import { ... } from '@comfyorg/ingest-types/zod'` |
| `src/platform/assets/schemas/mediaAssetSchema.ts` | ~50 | Zod schemas
| `import { ... } from '@comfyorg/ingest-types/zod'` |
| `src/platform/tasks/services/taskService.ts` | ~70 | Zod schemas |
`import { ... } from '@comfyorg/ingest-types/zod'` |
| `src/platform/workspace/workspaceTypes.ts` | ~6 | TS interface |
`export type { ... } from '@comfyorg/ingest-types'` |
- [x] `pnpm generate` in `packages/ingest-types/` produces
`types.gen.ts` and `zod.gen.ts`
- [x] `pnpm typecheck` passes
- [x] Pre-commit hooks pass (lint, typecheck, format)
- [x] Generated Zod schemas validate correct data and reject invalid
data
- [x] No import conflicts with existing code (generated types are
isolated in separate package)
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
feat: surface missing models in Error Tab for OSS and remove legacy dialog (#9921)
- Surface missing models in the Error Tab for OSS environments,
replacing the legacy modal dialog
- Add Download button per model and Download All button in group header
with file size display
- Move download business logic from `components/dialog/content` to
`platform/missingModel`
- Remove legacy missing models dialog components and composable
- **Pipeline**: Remove `isCloud` guard from `scanAllModelCandidates` and
`surfaceMissingModels` so OSS detects missing models
- **Grouping**: Group non-asset-supported models by directory in OSS
instead of lumping into UNSUPPORTED
- **UI**: Add Download button (matching Install Node Pack design) and
Download All header button
- **Store**: Add `folderPaths`/`fileSizes` state with setter methods,
race condition guard
- **Cleanup**: Delete `MissingModelsContent`, `MissingModelsHeader`,
`MissingModelsFooter`, `useMissingModelsDialog`, `missingModelsUtils`
- **Tests**: Add OSS/Cloud grouping tests, migrate Playwright E2E to
Error Tab, improve test isolation
- **Snapshots**: Reset Playwright screenshot expectations since OSS
missing model error detection now causes red highlights on affected
nodes
- **Accessibility**: Add `aria-label` with model name, `aria-expanded`
on toggle, warning icon for unknown category
- [x] Unit tests pass (86 tests)
- [x] TypeScript typecheck passes
- [x] knip passes
- [x] Load workflow with missing models in OSS → Error Tab shows missing
models grouped by directory
- [x] Download button triggers browser download with correct URL
- [x] Download All button downloads all downloadable models
- [x] Cloud environment behavior unchanged
- [x] Playwright E2E: `pnpm test:browser:local -- --grep "Missing models
in Error Tab"`
https://github.com/user-attachments/assets/12f15e09-215a-4c58-87ed-39bbffd1359c
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9921-feat-surface-missing-models-in-Error-Tab-for-OSS-and-remove-legacy-dialog-3236d73d365081f0a9dfc291978f5ecf)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: github-actions <github-actions@github.com>
fix: cloud subscribe redirect hangs waiting for billing init (#9965)
Fix /cloud/subscribe route hanging indefinitely because billing context
never initializes during the onboarding flow.
- **What**: Replace passive `await until(isInitialized).toBe(true)` with
explicit `await initialize()` in CloudSubscriptionRedirectView. Remove
unused `until` import.

In the onboarding flow, `useTeamWorkspaceStore().activeWorkspace` is not
set, so `useBillingContext`'s internal watch (which triggers
`initialize()` on workspace change) enters the `!newWorkspaceId` branch
— it resets `isInitialized` to `false` and returns without ever calling
`initialize()`. The old code then awaited `isInitialized` becoming
`true` forever.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9965-fix-cloud-subscribe-redirect-hangs-waiting-for-billing-init-3246d73d3650812d93ebd477c544fa0a)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
feat: add TBT/frameDuration metrics and new perf test scenarios (#9910)
Adds Total Blocking Time (TBT) and frame duration metrics to the
performance testing infrastructure, plus three new test scenarios
covering zoom, pan, and many-nodes-idle.
- **`totalBlockingTimeMs`** — Computed from PerformanceObserver
`longtask` entries: `sum(duration - 50ms)` for tasks >50ms. Measures
main thread blocking.
- **`frameDurationMs`** — Average frame duration via rAF timing (16.67ms
= 60fps target). Measures rendering smoothness.
| Scenario | Description |
|---|---|
| `canvas-zoom-sweep` | 10 zoom-in + 10 zoom-out cycles on default
workflow |
| `canvas-pan-many-nodes` | 10 pan sweeps over 100-node workflow |
| `canvas-many-nodes-idle` | 2-second idle measurement with 100 nodes
rendered |
- `PerformanceHelper.ts`: Installs PerformanceObserver for longtask,
collects TBT, measures frame duration via rAF
- `perf-report.ts`: Reports TBT and frame duration in PR comment tables
- `browser_tests/assets/perf/many_nodes_100.json`: 100-node (10×10 grid)
test fixture
- TBT collection clears entries at `startMeasuring()` and reads at
`stopMeasuring()` — ensure no race with observer buffering
- Frame duration sampling uses 10 frames — enough for signal without
slowing tests
Depends on: #9886, #9887
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9910-feat-add-TBT-frameDuration-metrics-and-new-perf-test-scenarios-3236d73d365081488ae3c594a8bf7cff)
by [Unito](https://www.unito.io)
fix: LGraphGroup paste position (#9962)
Fix group paste position: groups now paste at the cursor location
instead of on top of the original.
- **What**: Added LGraphGroup offset handling in _deserializeItems()
position adjustment loop, matching existing LGraphNode and Reroute
behavior.
Before:
https://github.com/user-attachments/assets/e317af10-8009-4092-9d14-de79316cd853
After:
https://github.com/user-attachments/assets/f4ffefd5-519a-4592-812c-c88e3b5940fd
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9962-fix-LGraphGroup-paste-position-3246d73d365081eea5b2e2507da861de)
by [Unito](https://www.unito.io)
fix: tree explorer nodes not filling parent container width (#9964)
Fix tree explorer nodes not filling the full width of the sidebar
container, causing text to overflow instead of truncating.
- **What**: Add `min-w-0` to `TreeRoot` to allow flex shrinking within
sidebar. Add `w-full` and `min-w-0` to tree node rows so
absolutely-positioned virtualizer items fill the container width and
text truncates correctly.
<img width="365" height="749" alt="image"
src="https://github.com/user-attachments/assets/320910f3-52ad-4634-a935-6bd1a40aea7f"
/>
The virtualizer renders each item with `position: absolute; left: 0` but
no explicit width, so rows would size to content rather than filling the
container. Adding `w-full` ensures rows stretch to 100% of the
virtualizer container, and `min-w-0` allows proper flex shrinking for
deep indentation levels.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9964-fix-tree-explorer-nodes-not-filling-parent-container-width-3246d73d36508138be38fdcac15ae4ef)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
feat: add Copy URL button to missing model rows for OSS (#9966)
1.42.6 (#9986)
Patch version increment to 1.42.6
**Base branch:** `main`
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9986-1-42-6-3256d73d365081a28bfad82022ce3440)
by [Unito](https://www.unito.io)
---------
Co-authored-by: christian-byrne <72887196+christian-byrne@users.noreply.github.com>
Co-authored-by: github-actions <github-actions@github.com>
fix: block missing e2e regression coverage in CodeRabbit (#9987)
Make the CodeRabbit end-to-end regression coverage check actually block
fix-like PRs until it is resolved or explicitly overridden by a
requested reviewer, and harden the prompt so it evaluates only PR-local
metadata.
- **What**: Set the `End-to-end regression coverage for fixes` custom
check mode from `warning` to `error`
- **What**: Enable `reviews.request_changes_workflow` so CodeRabbit can
block on failed `error` pre-merge checks
- **What**: Set
`reviews.pre_merge_checks.override_requested_reviewers_only` to `true`
so only requested reviewers can bypass a failed check
- **What**: Tighten the custom check instructions to use only PR
metadata in review context, avoid shell commands, and avoid reverse-diff
or base-branch file evaluation
Confirm this is the intended CodeRabbit enforcement model for missing
Playwright regression coverage on fix-like PRs and that the prompt
wording is strict enough to avoid false positives from reversed diffs.
fix: add reve and elevenlabs to icon safelist (#9990)
Reve and ElevenLabs provider icons were not displaying in the node
library because they were missing from the Tailwind icon safelist.
- **What**: Add `reve` and `elevenlabs` to the `@source inline` safelist
in `style.css` so `icon-[comfy--reve]` and `icon-[comfy--elevenlabs]`
classes are generated. Add corresponding `PROVIDER_COLORS` entries in
`categoryUtil.ts`.
<img width="308" height="106" alt="image"
src="https://github.com/user-attachments/assets/d488898a-fbad-4af0-8921-0e8ee7d4705a"
/>
<img width="308" height="78" alt="image"
src="https://github.com/user-attachments/assets/2b3b7172-095b-415e-a49a-d303977e0abc"
/>
The SVG files already existed in `packages/design-system/src/icons/` but
Tailwind's tree-shaking dropped the classes since they're only used
dynamically via `getProviderIcon()`.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9990-fix-add-reve-and-elevenlabs-to-icon-safelist-3256d73d36508105994fcdd5d0568027)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
fix: mask editor save shows blank image in Load Image node (#9984)
Mask editor save was showing a blank image in the Load Image node
(legacy nodes mode, not Nodes 2.0) because
`updateNodeWithServerReferences` called `updateNodeImages`, which
silently no-ops when the node has no pre-existing execution outputs.
Replaced with `setNodeOutputs` which properly creates output entries
regardless of prior state.
**Affects:** Legacy nodes mode only. Nodes 2.0 (Vue Nodes) renders
images via Vue components and is not affected.
- Fixes#9983
- Fixes#9782
- Fixes#9952
| Commit | SHA | CI Status | Run | Purpose |
|--------|-----|-----------|-----|---------|
| `test: add failing test for mask editor save showing blank image` |
`0ab66e8` | 🔴
[Red](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/23125427860)
| CI: Tests Unit **failure** | Proves the test catches the bug |
| `fix: mask editor save shows blank image in Load Image node` |
`564cc9c` | 🟢
[Green](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/23127289891)
| CI: Tests Unit **success** | Proves the fix resolves the bug |
https://github.com/user-attachments/assets/8d5c36ce-2c5e-4609-b246-dcf896c4a8e7https://github.com/user-attachments/assets/c8ae4f0e-3da0-40f2-a543-d1d5a6bce795
- [x] CI red on test-only commit
- [x] CI green on fix commit
- [ ] E2E regression test not added: mask editor save requires canvas
pixel manipulation + server upload round-trip which is covered by the
existing unit test mocking the full `save()` flow. The Playwright test
infrastructure does not currently support mask editor interactions (draw
+ save).
- [x] Manual verification (legacy nodes mode): Load Image → upload →
mask editor → draw → save → verify image refreshes
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
fix: allow URL input for free tier users, gate on import button (#10024)
- Remove free-tier restriction from the URL input field in
`MissingModelUrlInput.vue` so it is always editable
- Move the subscription check (`canImportModels`) to the Import button
click handler — free-tier users see the upgrade modal only when they
attempt to import
- Extract inline ternary to named `handleImportClick` method for clarity
- [x] Unit tests added (`MissingModelUrlInput.test.ts`) verifying:
- URL input is always editable regardless of subscription tier
- Import button calls `handleImport` for paid users
- Import button calls `showUploadDialog` (upgrade modal) for free-tier
users
- [x] Verify URL input is editable for free-tier users on cloud
- [x] Verify clicking Import as free-tier opens the subscription modal
- [x] Verify paid users can import normally without changes
Playwright E2E regression tests are impractical for this change because
`MissingModelUrlInput` only renders when `isAssetSupported` is true,
which requires `isCloud` — a compile-time constant (`__DISTRIBUTION__`).
The OSS test build always sets `isCloud = false`, so the component never
renders in the E2E environment. Unit tests with mocked feature flags
provide equivalent behavioral coverage.
fix: prevent subscription UI from rendering on non-cloud distributions (#9958)
Prevent Plans & Pricing dialog, subscription buttons, and cloud-only
menu items from appearing on desktop/localhost distributions.
- **What**: Add `isCloud` guards to
`useSubscriptionDialog.showPricingTable`, `TopbarSubscribeButton`, and
`CurrentUserPopoverLegacy` so subscription UI only renders on cloud
- **Tests**: 24 tests across 3 test files (1 modified, 2 new) covering
cloud/non-cloud behavior
- Guard placement in `CurrentUserPopoverLegacy.vue` — multiple `v-if`
conditions updated to include `isCloud`
- Early-return in `showPricingTable` as a defense-in-depth measure
Fixes COM-16820
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9958-fix-prevent-subscription-UI-from-rendering-on-non-cloud-distributions-3246d73d365081559a9ee8650409c5b4)
by [Unito](https://www.unito.io)
Co-authored-by: Benjamin Lu <benjaminlu1107@gmail.com>
fix: prevent animated preview duplication on Vue↔Litegraph switch (#9938)
SaveAnimatedPNG/WEBP nodes show duplicate output previews when switching
between Vue and Litegraph renderer modes.
The `ANIM_PREVIEW_WIDGET` (`$$comfy_animation_preview`) DOM widget
lacked `canvasOnly: true`, so `shouldRenderAsVue()` in the widget
registry included it in Vue mode rendering. This caused both:
1. Vue's `ImagePreview.vue` (via `nodeMedia` computed from
`nodeOutputStore`)
2. The legacy `ANIM_PREVIEW_WIDGET` DOM widget (rendered as `WidgetDOM`)
to display simultaneously — duplicating the output preview.
Add `canvasOnly: true` to the `ANIM_PREVIEW_WIDGET` options, matching
the pattern used by `IMAGE_PREVIEW` widget in
`useImagePreviewWidget.ts`. This ensures the legacy widget is filtered
out in Vue mode by `shouldRenderAsVue()`, leaving `ImagePreview.vue` as
the single source of truth.
- All 539 vueNodes tests pass
- All 22 nodeOutputStore tests pass
- All 140 composables/node tests pass
- Typecheck passes
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9938-fix-prevent-animated-preview-duplication-on-Vue-Litegraph-switch-3246d73d365081019bbfd7e33a9c14fb)
by [Unito](https://www.unito.io)
1.43.0 (#10032)
Minor version increment to 1.43.0
**Base branch:** `main`
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-10032-1-43-0-3256d73d3650818e8408d25fdf28de48)
by [Unito](https://www.unito.io)
Co-authored-by: christian-byrne <72887196+christian-byrne@users.noreply.github.com>
Feat/3d thumbnail inline rendering (#9471)
The previous approach generated thumbnails server-side and uploaded them
as `model.glb.png` alongside the model file. This breaks on cloud
deployments where output files are renamed to content hashes, severing
the filename-based association between a model and its thumbnail.
Replace the server-upload approach with client-side Three.js rendering
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9471-Feat-3d-thumbnail-inline-rendering-31b6d73d3650816fbd7dd05b507aa80d)
by [Unito](https://www.unito.io)
test: add FeatureFlagHelper and QueueHelper for E2E test infrastructure (#9554)
Add 2 reusable test helpers for Playwright E2E tests, integrated into
the ComfyPage fixture. These provide standardized patterns for mocking
feature flags and queue state across all E2E tests.
- **`FeatureFlagHelper.ts`** — manage localStorage `ff:` prefixed
feature flags (`seedFlags` for init-time, `setFlags` for runtime) and
mock `/api/features` route
- **`QueueHelper.ts`** — mock `/api/queue` and `/api/history` routes
with configurable running/pending counts and success/error job entries
- **`ComfyPage.ts`** — integrate both helpers as
`comfyPage.featureFlags` and `comfyPage.queue`
- Helper API design: are `seedFlags`/`setFlags`/`mockServerFeatures` the
right abstractions for feature flag testing?
- Queue mock fidelity: does the mock history shape match real ComfyUI
API responses closely enough?
- These are test-only infrastructure — no production code changes.
This is the base PR for the Playwright E2E coverage stack. Waves 1-4 all
branch from this and can merge independently once this lands:
- **→ This PR**: Test infrastructure helpers
- #9555: Toasts, error overlay, selection toolbox, linear mode,
selection rectangle
- #9556: Node search, bottom panel, focus mode, job history, side panel
- #9557: Errors tab, node headers, queue notifications, settings sidebar
- #9558: Minimap, widget copy, floating menus, node library essentials
---------
Co-authored-by: GitHub Action <action@github.com>
feat: scaffold Astro 5 website app + design-system base.css
- Create apps/website/ with Astro 5, Vue 3, Tailwind v4 integration
- Static output, assetsPrefix /_website/, i18n (en + zh-CN)
- Nx targets: dev, serve, build, preview, typecheck
- Add base.css to design-system: brand tokens + Inter font-face only
- Add catalog entries: astro, @astrojs/vue, @astrojs/check, nanostores, @nanostores/vue
scaffold-01, scaffold-02
fix: add .gitignore and env.d.ts for Astro website app
feat: add layout shell — SEO head, analytics, nav, footer
- BaseLayout: OG/Twitter meta, canonical URL, GA4 GTM-NP9JM6K7,
Vercel Analytics, ClientRouter for SPA navigation
- SiteNav: Comfy logo, Enterprise/Gallery/About/Careers links,
Comfy Cloud + Comfy Hub CTA buttons, mobile hamburger menu
- SiteFooter: Product/Resources/Company/Legal columns,
social icons (GitHub, Discord, X, Reddit, LinkedIn, Instagram)
- Add @vercel/analytics to workspace catalog and website deps
fix: address CodeRabbit review — ARIA wiring, absolute OG URLs, Analytics component
- SiteNav: add aria-controls, aria-expanded, and id for mobile menu
- BaseLayout: use absolute URLs for og:image and twitter:image
- BaseLayout: replace inline inject() with official <Analytics /> component
style: apply oxfmt formatting
fix: remove unused deps from website package.json (knip)
fix: clean up unused catalog entries from pnpm-workspace.yaml
feat: add Wave 3 homepage sections (hero, social proof, pillars, testimonials, CTAs, manifesto, academy, placeholders)
## Summary
- Move `contain-layout contain-style` from the outer node container to
the inner wrapper to prevent CSS containment from interfering with
inter-node compositing
- Add `isolation: isolate` to the outer container for explicit stacking
context isolation
- Fixes#9988
## Root Cause
Originally, PR
[#5169](https://github.com/Comfy-Org/ComfyUI_frontend/pull/5169)
(`57db10f`, 2025-08-23) applied all three containment properties to the
outer node container:
```
contain-layout contain-style contain-paint
```
`contain-paint` ensured each node's painting was clipped to its own
boundaries, guaranteeing render isolation between sibling nodes.
Later, PR
[#6289](https://github.com/Comfy-Org/ComfyUI_frontend/pull/6289)
(`6e4471ad`, 2025-10-28) removed `contain-paint` to fix a node sizing
issue — likely because `contain-paint` clips content to the element's
padding box, which would clip selection overlays and footer tabs that
extend beyond the node boundary via negative insets.
Removing `contain-paint` while keeping `contain-layout contain-style` on
the outer container broke paint isolation between nodes. Without paint
containment, the browser's compositor no longer treated each node as an
isolated painting unit, allowing text from lower z-index nodes to
visually bleed through higher z-index nodes.
## Fix
1. **Move containment inward**: `contain-layout contain-style` is moved
from the outer container to the inner wrapper (`node-inner-wrapper`).
This preserves layout/style containment performance benefits while
keeping them scoped to the node body — away from the stacking context
boundary between sibling nodes.
2. **Explicit isolation**: `isolation: isolate` is added to the outer
container. While `transform` + `z-index` already create a stacking
context, `isolation: isolate` provides an unambiguous signal to the
browser's compositor to treat each node as an isolated compositing unit
— replacing the paint isolation role that `contain-paint` previously
served, without clipping overlays.
## Red-Green Verification
| Commit | Status | Purpose |
|--------|--------|---------|
|
[`feb8555`](https://github.com/Comfy-Org/ComfyUI_frontend/commit/feb855501)
`test:` | 🔴 Red (verified locally) | Test asserts `isolate`
on outer container and `contain-layout contain-style` on inner wrapper
only — fails against current main |
|
[`c47f89e`](https://github.com/Comfy-Org/ComfyUI_frontend/commit/c47f89ed2)
`fix:` | 🟢 Green (CI: Tests Unit passing) | Moves
containment to inner wrapper, adds `isolation: isolate` — test passes |
**Local red verification:**
```
AssertionError: expected [ 'group/node', 'lg-node', …(9) ] to include 'isolate'
❯ src/renderer/extensions/vueNodes/components/LGraphNode.test.ts:264:26
```
## Test Plan
- [x] CI red on test-only commit (verified locally — test fails without
fix)
- [x] CI green on fix commit (CI: Tests Unit passes)
- [ ] Place a String Literal node with long text on the canvas
- [ ] Place a Load Image node (with an image loaded) overlapping the
String Literal
- [ ] Verify text from the String Literal does NOT bleed through the
Load Image node
- [ ] Verify node selection outlines, error borders, and footer tabs
render correctly
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: github-actions <github-actions@github.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
## Summary
- Read `sentry_dsn` from remote config at runtime for cloud builds,
following the same pattern as `firebase.ts` and `comfyApi.ts`
- Upload sourcemaps to both staging (`SENTRY_PROJECT`) and prod
(`SENTRY_PROJECT_PROD`) Sentry projects at build time
- Remove `SENTRY_DSN_PROD` / `USE_PROD_CONFIG`-based DSN selection — DSN
is now delivered per-environment via dynamic config
## Companion PR
- https://github.com/Comfy-Org/cloud/pull/2883 (cloud repo: dynamic
config + workflow)
## Setup required
- Add `SENTRY_PROJECT_PROD` variable (`cloud-frontend-prod`) to GitHub
STAGING environment in cloud repo
## Test plan
- [ ] Verify staging build reads staging DSN from remote config
- [ ] Verify prod build reads prod DSN from remote config
- [ ] Confirm sourcemaps are uploaded to both Sentry projects
- [ ] Validate fallback to `__SENTRY_DSN__` when remote config is
unavailable
## Summary
Fix high-resolution image preview expanding to full screen width, hiding
the close button.
## Changes
- **What**: Override scoped `ComfyImage` styles (`width/height: 100%`,
`object-fit: cover`) in the gallery context with `auto` sizing and
`max-width: 100vw; max-height: 100vh` constraints. The scoped
`[data-v-xxx]` attribute selector gave `comfy-image-main` higher
specificity than the existing `img.galleria-image` rule, so the new
selector `img.comfy-image-main.galleria-image` is used to win
specificity.
## Review Focus
- CSS specificity approach: using `.comfy-image-main.galleria-image` to
override Vue scoped styles without `!important`.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-10129-fix-high-res-image-preview-overflowing-screen-and-hiding-close-button-3266d73d36508191a03bcc75f9912421)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
Co-authored-by: GitHub Action <action@github.com>
## Summary
- Remove the legacy missing nodes modal dialog and migrate all
functionality to the existing Error Overlay / TabErrors system
- Migrate core node version warning from `MissingCoreNodesMessage.vue`
to `MissingNodeCard` in the errors tab
- Remove `Comfy.Workflow.ShowMissingNodesWarning` setting (errors tab
always surfaces missing nodes)
- Delete 6 legacy files: `useMissingNodesDialog.ts`,
`MissingNodesContent.vue`, `MissingNodesFooter.vue`,
`MissingNodesHeader.vue`, `MissingCoreNodesMessage.vue` and its test
- Rename `showMissingNodesDialog`/`showMissingModelsDialog` params to
`showMissingNodes`/`showMissingModels`
- Add `errorOverlay` and `missingNodeCard` to centralized `TestIds`
- Migrate all E2E tests from legacy dialog selectors to error overlay
testIds
- Add new E2E test: MissingNodeCard visible via "See Errors" button flow
- Add new E2E test: subgraph missing node type verified by expanding
pack row
- Add `surfaceMissingNodes` unit tests to `executionErrorStore`
- Guard `semver.compare` against invalid version strings
- Add `role="alert"`, `aria-hidden` for accessibility
- Use reactive props destructuring in `MissingNodeCard` and
`MissingPackGroupRow`
**Net change: -669 lines** (19 files, +323 / -992)
<img width="733" height="579" alt="image"
src="https://github.com/user-attachments/assets/c497809d-b176-43bf-9872-34bd74c6ea0d"
/>
## Test plan
- [x] Unit tests: MissingNodeCard core node warning (7 tests)
- [x] Unit tests: surfaceMissingNodes (4 tests)
- [x] Unit tests: workflowService showPendingWarnings (updated)
- [x] E2E: Error overlay visible on missing nodes workflow
- [x] E2E: Error overlay visible on subgraph missing nodes
- [x] E2E: MissingNodeCard visible via See Errors button
- [x] E2E: Subgraph node type visible after expanding pack row
- [x] E2E: Error overlay does not resurface on undo/redo
- [x] E2E: Error overlay does not reappear on workflow tab switch
- [x] Typecheck, lint, knip all passing
## Related issues
- Closes#9923 (partially — `errorOverlay` and `missingNodeCard` added
to TestIds)
- References #10027 (mock hoisting inconsistency)
- References #10033 (i18n-based test selectors)
- References #10085 (DDD layer violation + focusedErrorNodeId)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-10102-refactor-remove-legacy-missing-nodes-dialog-3256d73d365081c194d2e90bc6401846)
by [Unito](https://www.unito.io)
Resolve all lint warnings (3 oxlint + 1 eslint).
## Changes
- Replace `it.todo` with `it.skip` in subgraph tests (`warn-todo`)
- Move `vi.mock` to top-level in `firebaseAuthStore.test.ts`
(`hoisted-apis-on-top`)
- Rename `DOMPurify` default import to `dompurify`
(`no-named-as-default`)
---
### The Villager Who Ignored the Warnings
Once there lived a villager whose compiler whispered of lint. "They are
only *warnings*," she said, and went about her day. One warning became
three. Three became thirty. The yellow text grew like ivy across the
terminal, until no one could tell the warnings from the errors. One
morning a real error appeared — a misplaced mock, a shadowed import —
but nobody noticed, for the village had long since learned to stop
reading. The build shipped. The users wept. And the warning, faithful to
the last, sat quietly in the log where it had always been.
*Moral: Today's warning is tomorrow's incident report.*
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9972-fix-resolve-all-lint-warnings-3246d73d3650810a89cde5d05e79d948)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
## Summary
Restrict PyPI publishing of `comfyui-frontend-package` to bi-weekly
ComfyUI release cycles only, instead of every nightly version bump.
## Changes
- **What**: Move `publish_pypi` job from `release-draft-create.yaml` to
`release-biweekly-comfyui.yaml`
1. Removed `publish_pypi` job from `release-draft-create.yaml` (no
longer publishes on every merged Release PR)
2. Added `publish-pypi` job to `release-biweekly-comfyui.yaml` with tag
polling, build, publish, and PyPI availability confirmation
3. Gated `create-comfyui-pr` on `publish-pypi` success so the ComfyUI
requirements bump PR is only created after the package is confirmed
available
4. Updated ComfyUI PR body to confirm PyPI availability instead of
warning about a pending release PR
- **Breaking**: None — nightly releases still create GitHub releases and
publish npm types; only PyPI publishing timing changes
- **Dependencies**: None
## Review Focus
- The `publish-pypi` job uses `if: always() &&
needs.resolve-version.result == 'success'` to run even when
`trigger-release-if-needed` is skipped (tag already exists)
- Tag polling (30min timeout) waits for the version bump PR to be merged
before building from the tagged commit
- PyPI propagation polling (15min timeout) confirms the package is
installable before creating the ComfyUI PR
Fixes COM-16778
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9948-ops-restrict-PyPI-publishing-to-bi-weekly-ComfyUI-releases-3246d73d36508198b00fcc247ac5b58c)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
- Adds `gh pr merge --auto --squash` after backport PR creation in the
backport workflow, so backport PRs merge automatically once checks pass
- Uses `|| echo "::warning::..."` fallback to avoid failing the workflow
if auto-merge can't be enabled (e.g. repo setting not configured)
## Test plan
- [ ] Trigger backport workflow on a test PR with `needs-backport` label
- [ ] Verify auto-merge is enabled on the created backport PR
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-10108-chore-enable-auto-merge-on-backport-PRs-3256d73d3650814eb6e5fb2bdf3c5ec7)
by [Unito](https://www.unito.io)
## Summary
Restore the existing job details popover in the expanded queue overlay
after the list row UI moved to `AssetsListItem`.
## Changes
- **What**: Add delayed hover-driven job details popover support back to
`JobAssetsList`, including teleported positioning, row-to-popover hover
handoff, and focused component coverage for the timing behavior.
## Review Focus
Please verify the hover transition between the queue row and the
teleported popover, especially around positioning and whether the
popover remains reachable in the top menu queue overlay.
## Screenshots (if applicable)
N/A
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9549-fix-restore-queue-job-details-popover-31d6d73d3650815eaf45dcc2a14d3dee)
by [Unito](https://www.unito.io)
## Summary
Remove team assignments from CODEOWNERS to reduce notification noise for
internal PRs. Add a workflow that requests team review only when
external contributors open PRs.
## Changes
- **What**: Strip `@Comfy-org/comfy_frontend_devs` and
`@Comfy-Org/comfy_maintainer` from all CODEOWNERS entries (keep
individual user assignments). Add `pr-request-team-review.yaml` workflow
that uses `pull_request_target` to request team review for
non-collaborator PRs.
- **Dependencies**: None
## Review Focus
- The workflow uses `pull_request_target` but does not check out or
execute any untrusted code — it only runs `gh pr edit --add-reviewer`.
- The `author_association` check excludes OWNER, MEMBER, and
COLLABORATOR — internal PRs will not trigger team review requests.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-10104-chore-replace-team-CODEOWNERS-with-external-PR-review-workflow-3256d73d3650813b887ac16b5e97b4c4)
by [Unito](https://www.unito.io)
## Summary
- Adds a **Paste Image** context menu option to nodes that support image
pasting (e.g. Load Image), complementing the existing **Copy Image**
option
- Reads clipboard image data via `navigator.clipboard.read()` and
delegates to `node.pasteFiles()` — same path as `Ctrl+V` paste
- Only shown on nodes that implement `pasteFiles` (e.g. LoadImage)
- Available even when no image is loaded yet (e.g. fresh LoadImage node)
- **Node 2.0 context menu only** — legacy litegraph menu is not
supported
- Fixes#9989
<img width="852" height="685" alt="스크린샷 2026-03-16 오후 5 34 28"
src="https://github.com/user-attachments/assets/219e8162-312a-400b-90ec-961b95b5f472"
/>
## Test plan
- [x] Right-click a Load Image node (with or without image loaded) →
verify "Paste Image" appears
- [x] Copy an image to clipboard → click "Paste Image" → verify image is
loaded into the node
- [x] Verify "Paste Image" does not appear on output-only image nodes
(e.g. Save Image, Preview Image)
- [x] Unit tests pass: `pnpm vitest run
src/composables/graph/useImageMenuOptions.test.ts`
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
# Documentation Audit - File Path Updates
## Summary
Conducted a comprehensive audit of all documentation files across the
repository to ensure 100% accuracy of file paths, code references, and
technical information. Fixed outdated file path references that resulted
from codebase restructuring.
## Changes Made
### File Path Corrections
**docs/SETTINGS.md**
- Updated `src/constants/coreSettings.ts` →
`src/platform/settings/constants/coreSettings.ts` (3 occurrences)
- Updated `src/stores/settingStore.ts` →
`src/platform/settings/settingStore.ts` (2 occurrences)
- Updated `src/services/newUserService.ts` →
`src/services/useNewUserService.ts` (1 occurrence)
**src/locales/CONTRIBUTING.md**
- Updated `src/constants/coreSettings.ts` →
`src/platform/settings/constants/coreSettings.ts` (1 occurrence)
**docs/testing/unit-testing.md**
- Updated `@/domains/workflow/validation/schemas/workflowSchema` →
`@/platform/workflow/validation/schemas/workflowSchema` (1 occurrence)
## Audit Results
### Files Audited (✅ All Pass)
- ✅ Core documentation: README.md, CONTRIBUTING.md, CLAUDE.md, AGENTS.md
- ✅ Docs directory: 27 files in docs/**/*.md
- ✅ Claude commands: 7 files in .claude/commands/*.md
- ✅ Locales documentation: src/locales/CONTRIBUTING.md
- ✅ Package.json scripts: All documented commands verified against
actual scripts
### Accuracy Statistics
- **Total file path references audited:** 50+
- **Broken references found:** 5
- **Broken references fixed:** 5
- **Documentation files scanned:** 31
- **Final accuracy rate:** 100%
### Verified as Accurate
- ✅ All package.json script references in AGENTS.md match actual scripts
- ✅ All configuration file references (.vscode/extensions.json,
vite.config.mts, etc.) point to existing files
- ✅ All code examples reference actual existing files
- ✅ All ADR (Architecture Decision Records) dates and references are
correct
- ✅ Extension API examples match current implementation
- ✅ Feature flag system documentation aligns with actual code
- ✅ Settings system documentation reflects current architecture
- ✅ Testing guides reference correct test file locations
- ✅ Vue component conventions match actual patterns
- ✅ TypeScript conventions align with codebase standards
## Review Notes
This audit was triggered by a codebase restructuring where
settings-related files were moved from `src/constants/` and
`src/stores/` to a new platform-based organization under
`src/platform/settings/`. Additionally, some service files were renamed
to follow the `use*` composable naming convention.
All documentation now accurately reflects the current codebase
structure. No functionality changes were made - only documentation
updates to maintain accuracy.
### Files with Intentionally Deleted References (No Fix Needed)
- `build/plugins/generateImportMapPlugin.ts` - Documented as removed in
ADR-0005, reference is intentionally showing historical context
## Testing
- Verified all updated file paths exist in the codebase
- Confirmed all code examples still compile with updated import paths
- Validated no broken internal documentation links
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-10029-docs-Weekly-Documentation-Update-3256d73d36508177b363d4160085bbe9)
by [Unito](https://www.unito.io)
Co-authored-by: christian-byrne <72887196+christian-byrne@users.noreply.github.com>
## Summary
Add 2 reusable test helpers for Playwright E2E tests, integrated into
the ComfyPage fixture. These provide standardized patterns for mocking
feature flags and queue state across all E2E tests.
## Changes
- **`FeatureFlagHelper.ts`** — manage localStorage `ff:` prefixed
feature flags (`seedFlags` for init-time, `setFlags` for runtime) and
mock `/api/features` route
- **`QueueHelper.ts`** — mock `/api/queue` and `/api/history` routes
with configurable running/pending counts and success/error job entries
- **`ComfyPage.ts`** — integrate both helpers as
`comfyPage.featureFlags` and `comfyPage.queue`
## Review Focus
- Helper API design: are `seedFlags`/`setFlags`/`mockServerFeatures` the
right abstractions for feature flag testing?
- Queue mock fidelity: does the mock history shape match real ComfyUI
API responses closely enough?
- These are test-only infrastructure — no production code changes.
## Stack
This is the base PR for the Playwright E2E coverage stack. Waves 1-4 all
branch from this and can merge independently once this lands:
- **→ This PR**: Test infrastructure helpers
- #9555: Toasts, error overlay, selection toolbox, linear mode,
selection rectangle
- #9556: Node search, bottom panel, focus mode, job history, side panel
- #9557: Errors tab, node headers, queue notifications, settings sidebar
- #9558: Minimap, widget copy, floating menus, node library essentials
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
The previous approach generated thumbnails server-side and uploaded them
as `model.glb.png` alongside the model file. This breaks on cloud
deployments where output files are renamed to content hashes, severing
the filename-based association between a model and its thumbnail.
Replace the server-upload approach with client-side Three.js rendering
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9471-Feat-3d-thumbnail-inline-rendering-31b6d73d3650816fbd7dd05b507aa80d)
by [Unito](https://www.unito.io)
## Problem
SaveAnimatedPNG/WEBP nodes show duplicate output previews when switching
between Vue and Litegraph renderer modes.
## Root Cause
The `ANIM_PREVIEW_WIDGET` (`$$comfy_animation_preview`) DOM widget
lacked `canvasOnly: true`, so `shouldRenderAsVue()` in the widget
registry included it in Vue mode rendering. This caused both:
1. Vue's `ImagePreview.vue` (via `nodeMedia` computed from
`nodeOutputStore`)
2. The legacy `ANIM_PREVIEW_WIDGET` DOM widget (rendered as `WidgetDOM`)
to display simultaneously — duplicating the output preview.
## Fix
Add `canvasOnly: true` to the `ANIM_PREVIEW_WIDGET` options, matching
the pattern used by `IMAGE_PREVIEW` widget in
`useImagePreviewWidget.ts`. This ensures the legacy widget is filtered
out in Vue mode by `shouldRenderAsVue()`, leaving `ImagePreview.vue` as
the single source of truth.
## Testing
- All 539 vueNodes tests pass
- All 22 nodeOutputStore tests pass
- All 140 composables/node tests pass
- Typecheck passes
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9938-fix-prevent-animated-preview-duplication-on-Vue-Litegraph-switch-3246d73d365081019bbfd7e33a9c14fb)
by [Unito](https://www.unito.io)
## Summary
Prevent Plans & Pricing dialog, subscription buttons, and cloud-only
menu items from appearing on desktop/localhost distributions.
## Changes
- **What**: Add `isCloud` guards to
`useSubscriptionDialog.showPricingTable`, `TopbarSubscribeButton`, and
`CurrentUserPopoverLegacy` so subscription UI only renders on cloud
- **Tests**: 24 tests across 3 test files (1 modified, 2 new) covering
cloud/non-cloud behavior
## Review Focus
- Guard placement in `CurrentUserPopoverLegacy.vue` — multiple `v-if`
conditions updated to include `isCloud`
- Early-return in `showPricingTable` as a defense-in-depth measure
Fixes COM-16820
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9958-fix-prevent-subscription-UI-from-rendering-on-non-cloud-distributions-3246d73d365081559a9ee8650409c5b4)
by [Unito](https://www.unito.io)
Co-authored-by: Benjamin Lu <benjaminlu1107@gmail.com>
## Summary
- Remove free-tier restriction from the URL input field in
`MissingModelUrlInput.vue` so it is always editable
- Move the subscription check (`canImportModels`) to the Import button
click handler — free-tier users see the upgrade modal only when they
attempt to import
- Extract inline ternary to named `handleImportClick` method for clarity
## Test plan
- [x] Unit tests added (`MissingModelUrlInput.test.ts`) verifying:
- URL input is always editable regardless of subscription tier
- Import button calls `handleImport` for paid users
- Import button calls `showUploadDialog` (upgrade modal) for free-tier
users
- [x] Verify URL input is editable for free-tier users on cloud
- [x] Verify clicking Import as free-tier opens the subscription modal
- [x] Verify paid users can import normally without changes
## E2E test rationale
Playwright E2E regression tests are impractical for this change because
`MissingModelUrlInput` only renders when `isAssetSupported` is true,
which requires `isCloud` — a compile-time constant (`__DISTRIBUTION__`).
The OSS test build always sets `isCloud = false`, so the component never
renders in the E2E environment. Unit tests with mocked feature flags
provide equivalent behavioral coverage.
## Summary
Mask editor save was showing a blank image in the Load Image node
(legacy nodes mode, not Nodes 2.0) because
`updateNodeWithServerReferences` called `updateNodeImages`, which
silently no-ops when the node has no pre-existing execution outputs.
Replaced with `setNodeOutputs` which properly creates output entries
regardless of prior state.
**Affects:** Legacy nodes mode only. Nodes 2.0 (Vue Nodes) renders
images via Vue components and is not affected.
- Fixes#9983
- Fixes#9782
- Fixes#9952
## Red-Green Verification
| Commit | SHA | CI Status | Run | Purpose |
|--------|-----|-----------|-----|---------|
| `test: add failing test for mask editor save showing blank image` |
`0ab66e8` | 🔴
[Red](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/23125427860)
| CI: Tests Unit **failure** | Proves the test catches the bug |
| `fix: mask editor save shows blank image in Load Image node` |
`564cc9c` | 🟢
[Green](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/23127289891)
| CI: Tests Unit **success** | Proves the fix resolves the bug |
## manual testing
### as is
https://github.com/user-attachments/assets/8d5c36ce-2c5e-4609-b246-dcf896c4a8e7
### to be
https://github.com/user-attachments/assets/c8ae4f0e-3da0-40f2-a543-d1d5a6bce795
## Test Plan
- [x] CI red on test-only commit
- [x] CI green on fix commit
- [ ] E2E regression test not added: mask editor save requires canvas
pixel manipulation + server upload round-trip which is covered by the
existing unit test mocking the full `save()` flow. The Playwright test
infrastructure does not currently support mask editor interactions (draw
+ save).
- [x] Manual verification (legacy nodes mode): Load Image → upload →
mask editor → draw → save → verify image refreshes
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
## Summary
Make the CodeRabbit end-to-end regression coverage check actually block
fix-like PRs until it is resolved or explicitly overridden by a
requested reviewer, and harden the prompt so it evaluates only PR-local
metadata.
## Changes
- **What**: Set the `End-to-end regression coverage for fixes` custom
check mode from `warning` to `error`
- **What**: Enable `reviews.request_changes_workflow` so CodeRabbit can
block on failed `error` pre-merge checks
- **What**: Set
`reviews.pre_merge_checks.override_requested_reviewers_only` to `true`
so only requested reviewers can bypass a failed check
- **What**: Tighten the custom check instructions to use only PR
metadata in review context, avoid shell commands, and avoid reverse-diff
or base-branch file evaluation
## Review Focus
Confirm this is the intended CodeRabbit enforcement model for missing
Playwright regression coverage on fix-like PRs and that the prompt
wording is strict enough to avoid false positives from reversed diffs.
## Summary
Fix tree explorer nodes not filling the full width of the sidebar
container, causing text to overflow instead of truncating.
## Changes
- **What**: Add `min-w-0` to `TreeRoot` to allow flex shrinking within
sidebar. Add `w-full` and `min-w-0` to tree node rows so
absolutely-positioned virtualizer items fill the container width and
text truncates correctly.
<img width="365" height="749" alt="image"
src="https://github.com/user-attachments/assets/320910f3-52ad-4634-a935-6bd1a40aea7f"
/>
## Review Focus
The virtualizer renders each item with `position: absolute; left: 0` but
no explicit width, so rows would size to content rather than filling the
container. Adding `w-full` ensures rows stretch to 100% of the
virtualizer container, and `min-w-0` allows proper flex shrinking for
deep indentation levels.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9964-fix-tree-explorer-nodes-not-filling-parent-container-width-3246d73d36508138be38fdcac15ae4ef)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
## Summary
Adds Total Blocking Time (TBT) and frame duration metrics to the
performance testing infrastructure, plus three new test scenarios
covering zoom, pan, and many-nodes-idle.
## Changes
### New Metrics
- **`totalBlockingTimeMs`** — Computed from PerformanceObserver
`longtask` entries: `sum(duration - 50ms)` for tasks >50ms. Measures
main thread blocking.
- **`frameDurationMs`** — Average frame duration via rAF timing (16.67ms
= 60fps target). Measures rendering smoothness.
### New Test Scenarios
| Scenario | Description |
|---|---|
| `canvas-zoom-sweep` | 10 zoom-in + 10 zoom-out cycles on default
workflow |
| `canvas-pan-many-nodes` | 10 pan sweeps over 100-node workflow |
| `canvas-many-nodes-idle` | 2-second idle measurement with 100 nodes
rendered |
### Infrastructure
- `PerformanceHelper.ts`: Installs PerformanceObserver for longtask,
collects TBT, measures frame duration via rAF
- `perf-report.ts`: Reports TBT and frame duration in PR comment tables
- `browser_tests/assets/perf/many_nodes_100.json`: 100-node (10×10 grid)
test fixture
## Review Focus
- TBT collection clears entries at `startMeasuring()` and reads at
`stopMeasuring()` — ensure no race with observer buffering
- Frame duration sampling uses 10 frames — enough for signal without
slowing tests
Depends on: #9886, #9887
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9910-feat-add-TBT-frameDuration-metrics-and-new-perf-test-scenarios-3236d73d365081488ae3c594a8bf7cff)
by [Unito](https://www.unito.io)
## Summary
Fix /cloud/subscribe route hanging indefinitely because billing context
never initializes during the onboarding flow.
## Changes
- **What**: Replace passive `await until(isInitialized).toBe(true)` with
explicit `await initialize()` in CloudSubscriptionRedirectView. Remove
unused `until` import.

## Review Focus
In the onboarding flow, `useTeamWorkspaceStore().activeWorkspace` is not
set, so `useBillingContext`'s internal watch (which triggers
`initialize()` on workspace change) enters the `!newWorkspaceId` branch
— it resets `isInitialized` to `false` and returns without ever calling
`initialize()`. The old code then awaited `isInitialized` becoming
`true` forever.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9965-fix-cloud-subscribe-redirect-hangs-waiting-for-billing-init-3246d73d3650812d93ebd477c544fa0a)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
## What
Adds a 245-node workflow asset and two `@perf` tests to establish a
baseline for large-graph performance regressions (Tier 6 in the
performance backlog).
## Why
Backlog item N5: we need CI regression detection for compositor layer
management, GPU texture count, and transform pane cost at 245+ nodes.
This is PR1 of 2 — establishes baseline metrics on main. Future
optimization PRs will show improvement deltas against this baseline.
## Tests Added
- **`large graph idle rendering`** — 120 frames idle with 245 nodes,
measures style recalcs, layouts, task duration, heap delta
- **`large graph pan interaction`** — middle-click pan across 245 nodes,
stresses compositor layer management and transform recalculation
## Workflow Asset
`browser_tests/assets/large-graph-workflow.json` — 245 nodes (49
pipelines of CheckpointLoader → 2× CLIPTextEncode → KSampler +
EmptyLatentImage), 294 links. Minimal structure focused on node count.
## Verification
- [x] `pnpm typecheck:browser` passes
- [x] `pnpm lint` passes (eslint on changed file)
- [x] All link references in JSON validated programmatically
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9940-test-add-large-graph-perf-test-with-245-node-workflow-backlog-N5-3246d73d365081f6b5d8ddb9a85e6ad0)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
- Add a Claude Code skill (`/red-green-fix`) that enforces the red-green
commit pattern for bug fixes
- Ensures a failing test is committed first (red CI), then the fix is
committed separately (green CI)
- Gives reviewers proof that the test actually catches the bug
- Includes `reference/testing-anti-patterns.md` with common mistakes
contextualized to this codebase
## Structure
```
.claude/skills/red-green-fix/
├── SKILL.md # Main skill definition
└── reference/
└── testing-anti-patterns.md # Anti-patterns guide
```
## Test Plan
- [ ] Invoke `/red-green-fix <bug description>` in Claude Code and
verify the two-step workflow
- [ ] Confirm PR template includes red-green verification table
- [ ] Review anti-patterns reference for completeness
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9954-draft-add-red-green-fix-skill-for-verified-bug-fix-workflow-3246d73d365081339a83dc09263b0f33)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
## Summary
Fixes three critical issues with the CI performance reporting pipeline
that made perf reports useless on PRs (demonstrated by PR #9248 — deep
watcher removal merged without useful perf signal).
## Changes
### 1. Fix z-score baseline variance collection (`0/5 runs`)
**Root cause:** PR #9305 added z-score statistical analysis code to
`perf-report.ts`, but the historical data download step was placed in
the wrong workflow file. The report is generated in
`pr-perf-report.yaml` (a `workflow_run`-triggered job), but the
historical download was in `ci-perf-report.yaml` (the test runner) —
different runners, different filesystems.
**Fix:** Implement `perf-data` orphan branch storage:
- On push to main: save `perf-metrics.json` to `perf-data` branch with
timestamped filename
- On PR report: fetch last 5 baselines from `perf-data` branch into
`temp/perf-history/`
- Rolling window of 20 baselines, oldest pruned automatically
- Same pattern used by `github-action-benchmark` (33.7k repos)
### 2. Fix force-push comment staleness
**Root cause:** `cancel-in-progress: true` kills the perf test run
before it uploads artifacts. The downstream report workflow only
triggers on `conclusion == 'success'` — cancelled runs are ignored, so
the comment from the first successful run goes stale.
**Fix:**
- Change `cancel-in-progress: false` — with GitHub's queue depth of 1,
rapid pushes (A,B,C,D) run A and D, skipping B and C
- Add SHA validation in `pr-perf-report.yaml` — before posting, check if
the workflow_run's head SHA still matches the PR's current head. Skip
posting stale results.
### 3. Add permissions for baseline operations
- `contents: write` on CI job (needed for pushing to perf-data branch)
- `actions: read` on both workflows (needed for artifact/baseline
access)
## One-time setup required
After merging, create the `perf-data` orphan branch:
```bash
git checkout --orphan perf-data
git rm -rf .
echo '# Performance Baselines' > README.md
mkdir -p baselines
git add README.md baselines
git commit -m 'Initialize perf-data branch'
git push origin perf-data
```
The first 2 pushes to main after setup will build up variance data, and
z-scores will start appearing in PR reports (threshold is
`historical.length >= 2`).
## Testing
- YAML validated with `yaml.safe_load()`
- `perf-report.ts` `loadHistoricalReports()` already reads from
`temp/perf-history/<index>/perf-metrics.json` — no code changes needed
- All new steps use `continue-on-error: true` for graceful degradation
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9886-fix-fix-perf-CI-pipeline-z-score-baselines-force-push-staleness-baseline-storage-3226d73d365081538424c7945e71f308)
by [Unito](https://www.unito.io)
## What
Makes Vue nodes (Nodes 2.0) the default renderer for new desktop app
installs (version ≥1.41.0), matching the behavior already live for cloud
new installs.
## Why
Step 2 of the Nodes 2.0 rollout sequence:
1. ✅ Cloud new installs (≥1.41.0) — DONE
2. 👉 **Desktop app (new installs)** — this PR
3. ⬜ Local installs
4. ⬜ Remove Beta tag
5. ⬜ GTM announcement
No forced migration — only changes the default for new installs.
Existing users keep their setting. Rollback is a settings flip.
## Change
In `coreSettings.ts`, the `defaultsByInstallVersion` for
`Comfy.VueNodes.Enabled` changes from:
```typescript
defaultsByInstallVersion: { '1.41.0': isCloud },
```
to:
```typescript
defaultsByInstallVersion: { '1.41.0': isCloud || isDesktop },
```
## Gated on
- M2 perf target (≥52 FPS on 245-node workflow) — layer merge landed,
likely met
- M-DevRel migration docs (blocks Beta tag removal, not this flip)
Draft PR — ceremonial, to be merged when M2 checkpoint passes.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9947-feat-make-Vue-nodes-Nodes-2-0-default-for-new-desktop-installs-3246d73d365081b280dfff932c7aa016)
by [Unito](https://www.unito.io)
## Summary
Fix "Calculating dimensions" text flickering during live sampling
preview in Vue renderer.
## Changes
- **What**: Stop resetting `actualDimensions` to `null` on every
`imageUrl` change. Previous dimensions are retained while the new frame
loads, eliminating the flicker. Error state is still reset correctly.
## Review Focus
The watcher on `props.imageUrl` previously reset both `actualDimensions`
and `imageError`. Now it only resets `imageError`, since
`handleImageLoad` updates dimensions when the new frame actually loads.
This means stale dimensions show briefly between frames, which is
intentionally better than showing "Calculating dimensions" text.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9937-fix-prevent-live-preview-dimension-flicker-between-frames-3246d73d36508154a676e5996112354f)
by [Unito](https://www.unito.io)
## Summary
Expands the performance testing infrastructure to collect 4 additional
CDP metrics that are already returned by `Performance.getMetrics` but
were not being read. This is a zero-cost expansion — no additional CDP
calls, just reading more fields from the existing response.
## New Metrics
| Metric | CDP Source | What It Detects |
|---|---|---|
| `domNodes` | `Nodes` | DOM node count delta — widget DOM leaks during
node create/destroy |
| `jsHeapTotalBytes` | `JSHeapTotalSize` | Total heap delta — combined
with `heapDeltaBytes` shows GC pressure |
| `scriptDurationMs` | `ScriptDuration` | JS execution time vs total
task time — script vs rendering balance |
| `eventListeners` | `JSEventListeners` | Listener count delta — detects
listener accumulation across lifecycle |
## Changes
### `browser_tests/fixtures/helpers/PerformanceHelper.ts`
- Added 4 fields to `PerfSnapshot` interface
- Added 4 fields to `PerfMeasurement` interface
- Wired through `getSnapshot()` and `stopMeasuring()`
### `scripts/perf-report.ts`
- Added 4 fields to `PerfMeasurement` interface
- Expanded `MetricKey` type and `REPORTED_METRICS` array with 3 new
reported metrics (`domNodes`, `scriptDurationMs`, `eventListeners`)
- `jsHeapTotalBytes` is collected but not in `REPORTED_METRICS` — it's
used alongside `heapDeltaBytes` for GC pressure ratio analysis
## Why These 4
From a gap analysis of all ~30 CDP metrics, these were identified as
highest priority for ComfyUI:
- **`Nodes`** (P0): ComfyUI dynamically creates/destroys widget DOM. DOM
bloat from leaked widgets is a key performance risk, especially for Vue
Nodes 2.0.
- **`ScriptDuration`** (P1): Separates JS execution from layout/paint.
Reveals whether perf issues are script-heavy or rendering-heavy.
- **`JSEventListeners`** (P1): Widget lifecycle can leak listeners
across node add/remove cycles.
- **`JSHeapTotalSize`** (P1): With `JSHeapUsedSize`, the ratio shows GC
fragmentation pressure.
## Backward Compatibility
The `PerfMeasurement` interface is extended (not changed). Old baseline
`perf-metrics.json` files without these fields will have `undefined`
values, which the report script handles gracefully (shows `—` for
missing data).
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9887-feat-expand-CDP-perf-metrics-add-DOM-nodes-script-duration-event-listeners-3226d73d3650818abea1d4a441667c38)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
## Summary
Add E2E test coverage for the textarea widget right-click context menu
inside subgraphs.
The fix was shipped in #9840 — this PR adds the missing browser test.
## Test
- Loads a subgraph workflow with a CLIPTextEncode (textarea) node
- Navigates into the subgraph
- Right-clicks the textarea DOM element
- Asserts that the ComfyUI "Promote Widget" context menu option appears
## Related
- Fixes the test gap from #9840
- Notion ticket: d7a53160-e1e1-42bb-a5ac-c0c2702c629c
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9891-test-add-browser-test-for-textarea-right-click-context-menu-in-subgraph-3226d73d365081a4be51f89b5d505361)
by [Unito](https://www.unito.io)
## What
Add equality check before updating `appScalePercentage` reactive ref.
## Why
Firefox profiler shows 586 `setElementText` markers from continuous text
interpolation updates during zoom/pan. The rounded percentage value
often doesn't change between events.
## How
Extract `updateAppScalePercentage()` helper with equality guard —
compares new rounded value to current before assigning to the ref.
## Perf Impact
Expected: eliminates ~90% of `setElementText` markers during zoom/pan
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9403-fix-skip-redundant-appScalePercentage-updates-during-zoom-pan-31a6d73d3650812db8f2d68ac73c95b0)
by [Unito](https://www.unito.io)
## Summary
Fixes text field becoming non-editable when a previously linked input is
removed from a custom node.
## Problem
When a widget's input was promoted to a slot, connected via a link, and
then the input was removed (e.g., by updating the custom node
definition), the widget retained stale `slotMetadata` with `linked:
true`. This prevented the widget from being editable.
## Solution
In `refreshNodeSlots`, removed the `if (slotInfo)` guard so
`widget.slotMetadata` is always assigned — either to valid metadata or
`undefined`. This ensures stale linked state is cleared when inputs no
longer match widgets.
## Acceptance Criteria
1. Text field remains editable after promote→connect→disconnect cycle
2. Text field returns to editable state when noodle disconnected
3. No mode switching needed to restore editability
## Testing
- Added regression test: "clears stale slotMetadata when input no longer
matches widget"
- All existing tests pass (18/18 in affected file)
---
**Note: This PR currently contains only the RED (failing test) commit
for TDD verification. The GREEN (fix) commit will be pushed after CI
confirms the test failure.**
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9885-fix-clear-stale-widget-slotMetadata-on-link-disconnect-3226d73d365081269319c027b42d9f6b)
by [Unito](https://www.unito.io)
## Summary
- Remove Ctrl+C and Ctrl+V keybindings from the keybinding service
defaults so native browser copy/paste events fire
- This restores image paste into LoadImage nodes, which broke after
#9459
## Problem
PR #9459 moved Ctrl+C/V into the keybinding service, which calls
`event.preventDefault()` on keydown. This prevents the browser `paste`
event from firing, so `usePaste` (which detects images in the clipboard)
never runs. The `PasteFromClipboard` command only reads from
localStorage, completely bypassing image detection.
**Repro:** Copy a node → copy an image externally → try to paste the
image into a LoadImage node → gets old node data from localStorage
instead.
## Fix
Remove Ctrl+C and Ctrl+V from `CORE_KEYBINDINGS` in `defaults.ts`. The
native browser events now fire as before, and `useCopy`/`usePaste`
handle them correctly. Ctrl+Shift+V, Ctrl+A, Delete, and Backspace
keybindings remain in the keybinding service.
Fixes#9459 (regression)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9914-fix-restore-native-copy-paste-events-for-image-paste-support-3236d73d365081c7ac53f983f316e10f)
by [Unito](https://www.unito.io)
## Summary
Hide the template selector when a first-time cloud user accepts a shared
workflow from a share link, so the shared workflow opens without the
onboarding template dialog lingering.
## Changes
- **What**: Added shared-workflow loader behavior to close the global
template selector on accept actions (`copy-and-open` and `open-only`)
while keeping cancel behavior unchanged.
- **What**: Added targeted unit tests covering hide-on-accept and
no-hide-on-cancel behavior in the shared workflow URL loader.
## Review Focus
Confirm that share-link accept paths now dismiss the template selector
and that cancel still leaves it available.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9913-fix-hide-template-selector-after-shared-workflow-accept-3236d73d365081099c04e350d499fad2)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
## Summary
Reorder condition checks in `forwardPanEvent` so the cheap
`isMiddlePointerInput()` check runs first, avoiding expensive
`hasTextSelection()` → `window.getSelection().toString().trim()` on
every pointermove event.
## Changes
- **What**: `forwardPanEvent` now early-returns on
`!isMiddlePointerInput(e)` before calling `shouldIgnoreCopyPaste`, which
internally calls `hasTextSelection()`. Since most pointermove events are
not middle-button, this skips the expensive `toString()` call entirely.
## Review Focus
Semantic equivalence: the original condition was `(A && B) || C →
return`. Rewritten as two guards: `if (C) return; if (A && B) return;`.
The logic is equivalent — if `C` (`!isMiddlePointerInput`) is true, we
return regardless of `A && B`. If `C` is false (middle button), we check
`A && B` (`shouldIgnoreCopyPaste && activeElement`).
## Evidence
Backlog item #19. rizumu's Chrome DevTools profiling (Mar 2026) of a
245-node workflow showed `toString()` in `hasTextSelection` called on
every pointermove. Source: Slack `#C095BJSFV24` thread
`p1772823346206479`.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9892-fix-reorder-forwardPanEvent-conditions-to-skip-expensive-hasTextSelection-on-non-middle--3226d73d365081d38b89c8bb1dde3693)
by [Unito](https://www.unito.io)
## Summary
- Add `quickRegister()` entries for 3 model directories missing UI
backlinks:
- `CogVideo` → `DownloadAndLoadCogVideoModel` (covers CogVideo,
CogVideo/ControlNet/*, CogVideo/VAE)
- `inpaint` → `INPAINT_LoadInpaintModel`
- `layer_model` → `LayeredDiffusionApply`
These mappings ensure the "Use" button in the model browser correctly
creates
the appropriate loader node when users click on models from these node
packs.
## Test plan
- [ ] Verify "Use" button works for CogVideo models in model browser
- [ ] Verify "Use" button works for inpaint models (fooocus, lama, MAT)
- [ ] Verify "Use" button works for LayerDiffuse models
🤖 Generated with [Claude Code](https://claude.com/claude-code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9890-feat-add-model-to-node-mappings-for-CogVideo-inpaint-and-LayerDiffuse-3226d73d3650816ea547fbcdf3a20c35)
by [Unito](https://www.unito.io)
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Replacement for https://github.com/Comfy-Org/ComfyUI_frontend/pull/9364
## Summary
Add a generic mechanism for widgets to reactively display upstream
linked values when disabled, with concrete implementations for curve and
imagecrop widgets.
## Changes
- What: When a widget input is linked to an upstream node, the widget
enters a disabled state and displays the upstream node's current value
reactively. This is built as a two-layer system:
- Infrastructure layer: Resolve link origin info (originNodeId,
originOutputName) in buildSlotMetadata, pass it through
SimplifiedWidget.linkedUpstream, and provide a generic useUpstreamValue
composable that reads upstream values from widgetValueStore.
- Widget layer: Each widget type provides its own ValueExtractor to
interpret upstream data. singleValueExtractor handles simple
type-matched values (e.g. CurvePoint[]); boundsExtractor composes a
Bounds object from either a single upstream widget or four individual
x/y/width/height number widgets.
- Curve widget: shows upstream curve points in read-only mode
- ImageCrop widget: shows upstream bounding box with disabled crop
handles and number inputs
- CurveEditor and WidgetBoundingBox: gain disabled prop support
## Adapting future widgets
The system is designed so that any widget needing upstream value display
only needs to:
1. Accept widget: SimplifiedWidget as a prop (provides linkedUpstream
automatically)
2. Call useUpstreamValue(() => widget.linkedUpstream, extractor) with a
suitable extractor
3. Use singleValueExtractor(typeGuard) for single-value types, or write
a custom ValueExtractor for composite cases like boundsExtractor
4. Compute an effectiveValue that switches between upstream and local
based on disabled state
No infrastructure changes are needed — linkedUpstream is already
populated for all widget types that have a corresponding input slot.
## Review Focus
- The buildSlotMetadata helper is shared between extractVueNodeData and
refreshNodeSlots — verify the graph ref is reliably available in both
paths
- boundsExtractor composing from 4 individual number widgets
(x/y/width/height) — this handles the BBox→ImageCrop case where upstream
exposes separate widgets rather than a single Bounds object
## Screenshots
https://github.com/user-attachments/assets/dbc57a44-c5df-44f0-acce-d347797ee8fb
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9851-feat-reactive-upstream-value-display-for-disabled-curve-and-imagecrop-widgets-3226d73d36508134b386ddc9b9f1266b)
by [Unito](https://www.unito.io)
## Problem
When navigating back from a subgraph to the root graph, the SubgraphNode
can retain a stale progress bar. This happens because the progress
watcher in `GraphCanvas.vue` watches `[nodeLocationProgressStates,
canvasStore.canvas]`, but neither value changes reference during
subgraph navigation:
- `nodeLocationProgressStates` is already `{}` (execution completed
while viewing the subgraph)
- `canvasStore.canvas` is a `shallowRef` set once at startup — only
`canvas.graph` changes (via `setGraph()`)
**Reproduction** (from PR #4382 comment thread by @guill):
1. Create a subgraph with a KSampler
2. Execute the workflow
3. While progress bar is halfway, enter the subgraph
4. Wait for execution to complete
5. Navigate back to root graph
6. Progress bar is stuck at 50%
## Root Cause
`canvasStore.canvas` is a `shallowRef` — subgraph navigation mutates
`canvas.graph` (a nested property) via `LGraphCanvas.setGraph()`, which
doesn't trigger a shallow watch. The watcher never re-fires to clear
stale `node.progress` values.
## Fix
Add `canvasStore.currentGraph` to the watcher's dependency array. This
is already a `shallowRef` in `canvasStore` that's updated on every
`litegraph:set-graph` event. Zero overhead, precise targeting.
## Context
- Original discussion:
https://github.com/Comfy-Org/ComfyUI_frontend/pull/4382/files/BASE..868e047272f6c5d710db7e607b8997d4c243490f#r2202024855
- PR #9248 correctly removed `deep: true` from this watcher but missed
the subgraph edge case
- `deep: true` was the wrong fix — `canvasStore.currentGraph` is the
precise solution
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9865-fix-clear-stale-progress-bar-on-SubgraphNode-after-navigation-3226d73d3650811c8f9de612d81ef98a)
by [Unito](https://www.unito.io)
## Summary
Previously when I switch from nodes 1.0 to 2.0, positions and sizes of
nodes do not follow 'always snap to grid'. You can guess what a mess it
is for people relying on snap to grid to retain sanity. This PR fixes
it.
## Changes
In `ensureCorrectLayoutScale`, we call `snapPoint` after the position
and the size are updated.
We also need to ensure that the snapped size is larger than the minimal
size required by the content, so I've added 'ceil' mode to `snapPoint`,
and the patch is larger than I thought first.
I'd happily try out nodes 2.0 once this is addressed :)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9332-fix-respect-always-snap-to-grid-when-auto-scale-layout-from-nodes-1-0-to-2-0-3176d73d365081f5b6bcc035a8ffa648)
by [Unito](https://www.unito.io)
## Summary
Adds `view_mode` and `is_app_mode` properties to the
`app:run_button_click` telemetry event so analytics can segment run
button clicks by the user's current view context.
## Changes
- **`types.ts`**: Added `view_mode?: string` and `is_app_mode?: boolean`
to `RunButtonProperties`
- **`PostHogTelemetryProvider.ts`**: Computes `view_mode` and
`is_app_mode` from `useAppMode()` in `trackRunButton()`
- **`MixpanelTelemetryProvider.ts`**: Same as PostHog (providers are
mirrors)
## New Properties
| Property | Type | Description | Example Values |
|----------|------|-------------|----------------|
| `view_mode` | `string` | Granular AppMode value | `'graph'`, `'app'`,
`'builder:inputs'`, `'builder:outputs'`, `'builder:arrange'` |
| `is_app_mode` | `boolean` | Simplified flag for app mode vs non-app
mode | `true` when `mode === 'app' \|\| mode === 'builder:arrange'` |
## Design Decisions
- **Both granular and simplified**: `view_mode` gives exact mode for
detailed analysis; `is_app_mode` gives a quick boolean for simple
segmentation
- **Computed in providers**: View mode is read from `useAppMode()` at
tracking time, same pattern as `getExecutionContext()` — no changes
needed at call sites
- **`trigger_source` unchanged**: `keybindingService.ts` already reports
`trigger_source: 'keybinding'` regardless of view mode, satisfying the
requirement that keybinding invocations are correctly identified even in
app mode
## Testing
- Typecheck passes (no new errors)
- Format and lint pass (no new errors)
- Manual verification: all pre-existing errors are in unrelated files
(`draftCacheV2.property.test.ts`, `workflowDraftStoreV2.fsm.test.ts`)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9881-feat-telemetry-add-view_mode-and-is_app_mode-to-run_button_click-event-3226d73d36508101b3a8c7ff27706f81)
by [Unito](https://www.unito.io)
## Summary
Add structured preload error logging with Sentry context enrichment and
a user-facing toast notification when chunk loading fails (e.g. after a
deploy with new hashed filenames).
## Changes
- **`parsePreloadError` utility** (`src/utils/preloadErrorUtil.ts`):
Extracts structured info from `vite:preloadError` events — URL, file
type (JS/CSS/unknown), chunk name, and whether it looks like a hash
mismatch.
- **Sentry enrichment** (`src/App.vue`): Sets Sentry context and tags on
preload errors so they are searchable/filterable in the Sentry
dashboard.
- **User-facing toast**: Shows an actionable "please refresh" message
when a preload error occurs, across all distributions (cloud, desktop,
localhost).
- **Capture-phase resource error listener** (`src/App.vue`): Catches
CSS/script load failures that bypass `vite:preloadError` and reports
them to Sentry with the same structured context.
- **Unit tests** (`src/utils/preloadErrorUtil.test.ts`): 9 tests
covering URL parsing, chunk name extraction, hash mismatch detection,
and edge cases.
## Files Changed
| File | What |
|------|------|
| `src/App.vue` | Preload error handler + resource error listener |
| `src/locales/en/main.json` | Toast message string |
| `src/utils/preloadErrorUtil.ts` | `parsePreloadError()` utility |
| `src/utils/preloadErrorUtil.test.ts` | Unit tests |
## Review Focus
- Toast fires for all distributions (cloud/desktop/localhost) —
intentional so all users see stale chunk errors
- `parsePreloadError` is defensive — returns `unknown` for any field it
cannot parse
- Capture-phase listener filters to only `<script>` and `<link
rel="stylesheet">` elements
## References
- [Vite preload error
handling](https://vite.dev/guide/build#load-error-handling)
---------
Co-authored-by: bymyself <cbyrne@comfy.org>
## Summary
Custom names set on subgraph output nodes are ignored in the v2 renderer
— it always shows the data type name (e.g. "texts") instead of the
user-defined label. Works correctly in v1.
## Changes
- **What**: Made `outputs` in `extractVueNodeData` reactive via
`shallowReactive` + `defineProperty` (matching the existing `inputs`
pattern). Added a `node:slot-label:changed` graph trigger that
`SubgraphNode` fires when input/output labels are renamed, so the Vue
layer picks up the change.
## Review Focus
- The `outputs` reactivity mirrors `inputs` exactly — same
`shallowReactive` + setter pattern. The new trigger event forces
`shallowReactive` to detect the deep property change by re-assigning the
array.
- Also handles input label renames for consistency, even though the
current bug report is output-specific.
## Screenshots
**v1 — output correctly shows custom label "output_text":**
<img width="1076" height="628" alt="Screenshot 2026-02-26 at 4 43 00 PM"
src="https://github.com/user-attachments/assets/b4d6ae4c-9970-4d99-a872-4ce1b28522f2"
/>
**v2 before fix — output shows type name "texts" instead of custom
label:**
<img width="808" height="298" alt="Screenshot 2026-02-26 at 4 43 30 PM"
src="https://github.com/user-attachments/assets/cf06aa6c-6d4d-4be9-9bcd-dcc072ed1907"
/>
**v2 after fix — output correctly shows "output_text":**
<img width="1013" height="292" alt="Screenshot 2026-02-26 at 5 14 44 PM"
src="https://github.com/user-attachments/assets/3c43fa9b-0615-4758-bee6-be3481168675"
/>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9266-fix-subgraph-output-slot-labels-not-updating-in-v2-renderer-3146d73d365081979327fd775a6ef62b)
by [Unito](https://www.unito.io)
## Summary
Fix node layout drift from repeated `ensureCorrectLayoutScale` scaling,
simplify it to a pure one-time normalizer, and fix links not following
Vue nodes during drag.
## Changes
- **What**:
- `ensureCorrectLayoutScale` simplified to a one-time normalizer:
unprojects legacy Vue-scaled coordinates back to canonical LiteGraph
coordinates, marks the graph as corrected, and does nothing else. No
longer touches the layout store, syncs reroutes, or changes canvas
scale.
- Removed no-op calls from `useVueNodeLifecycle.ts` (a renderer version
string was passed where an `LGraph` was expected).
- `layoutStore.finalizeOperation` now calls `notifyChange` synchronously
instead of via `setTimeout`. This ensures `useLayoutSync`'s `onChange`
callback pushes positions to LiteGraph `node.pos` and calls
`canvas.setDirty()` within the same RAF frame as a drag update, fixing
links not following Vue nodes during drag.
- **Tests**: Added tests for `ensureCorrectLayoutScale` (idempotency,
round-trip, unknown-renderer no-op) and `graphRenderTransform`
(project/unproject round-trips, anchor caching).
## Review Focus
- The `setTimeout(() => this.notifyChange(change), 0)` →
`this.notifyChange(change)` change in `layoutStore.ts` is the key fix
for the drag-link-sync bug. The listener (`useLayoutSync`) only writes
to LiteGraph, not back to the layout store, so synchronous notification
is safe.
- `ensureCorrectLayoutScale` no longer has any side effects beyond
normalizing coordinates and setting `workflowRendererVersion` metadata.
---------
Co-authored-by: Amp <amp@ampcode.com>
Co-authored-by: Christian Byrne <cbyrne@comfy.org>
Co-authored-by: jaeone94 <89377375+jaeone94@users.noreply.github.com>
Co-authored-by: AustinMroz <austin@comfy.org>
Co-authored-by: Hunter <huntcsg@users.noreply.github.com>
Co-authored-by: GitHub Action <action@github.com>
## Summary
Fix duplicate LLink objects created during subgraph unpacking, where
output.links contains multiple link IDs for the same connection but
input.link only references one, leaving orphaned links.
## Changes
- **What**: Three layers of defense against duplicate links:
1. **Serialization fix** (`slotUtils.ts`): Clone `output.links` array in
`outputAsSerialisable` to prevent shared-reference mutation during
serialization round-trips
2. **Self-healing** (`LGraph.ts`): `_removeDuplicateLinks()` sanitizes
corrupted data during `configure()`, keeping the link referenced by
`input.link` and removing orphaned duplicates from `output.links` and
`_links`
3. **Unpack dedup** (`LGraph.ts`): Subgraph unpacking filters `newLinks`
via a `seenLinks` Set before creating connections
Runtime diagnostic logging via `graph.events` (no Sentry import in
litegraph):
- `_dupLinkIndex` Map for O(1) duplicate detection, only allocated when
enabled
- `_checkDuplicateLink()` called at the 3 link-creation sites
(`connectSlots`, `SubgraphInput.connect`, `SubgraphOutput.connect`)
- App layer listens for `diagnostic:duplicate-link` events and forwards
to Sentry with rate-limiting (1 per key per 60s)
## Review Focus
- The `_removeDuplicateLinks` strategy of keeping the link referenced by
`input.link` and removing others from `output.links` + `_links`
- The diagnostic index lifecycle: built on enable, updated on link
create/remove, cleared on disable
- Sentry integration in `app.ts` using the existing `graph.events`
system to avoid coupling litegraph to Sentry
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9120-fix-detect-and-remove-duplicate-links-in-subgraph-unpacking-3106d73d3650815b995ddf8f41da67ae)
by [Unito](https://www.unito.io)
## What
Replace `canvas.offsetHeight` with `canvas.height / devicePixelRatio` in
`renderInfo` to avoid forced synchronous layout.
## Why
`renderInfo` is called ~2,631 times in a typical session. Each call
reads `this.canvas.offsetHeight`, which forces the browser to flush
pending style/layout changes synchronously. With PrimeVue injecting
styles dynamically and Vue patching the DOM, there are almost always
pending mutations — converting every canvas-only `renderInfo` call into
a forced layout.
## How
`canvas.height` is the DPR-scaled internal resolution (set in
`resizeCanvas` as `cssHeight * devicePixelRatio`). Dividing by
`devicePixelRatio` yields the same CSS pixel value as `offsetHeight`
without triggering layout.
## Verification
- [x] Unit test: verifies `offsetHeight` is not accessed when y is
provided
- [x] Unit test: verifies fallback uses `canvas.height /
devicePixelRatio`
- [x] `pnpm typecheck` passes
- [x] `pnpm lint` passes
- [x] All litegraph tests pass (538 passed)
## Perf Impact
Eliminates ~2,631 forced synchronous layouts per session from the canvas
info panel.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9304-fix-avoid-forced-layout-in-renderInfo-by-using-canvas-height-3156d73d36508171973dda289b30d5ee)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## What
Add a per-frame text measurement cache for all hot-path
ctx.measureText() calls.
## Why
drawTruncatingText() in BaseWidget calls ctx.measureText() per widget
per frame with zero caching. For a 50-node workflow at 60fps:
~78,000-243,000 measureText calls/sec. Text labels rarely change between
frames.
## How
Global Map<string, number> cache keyed by font+text, cleared once per
frame at the start of drawFrontCanvas(). Replaces direct
ctx.measureText() calls in BaseWidget.drawTruncatingText, draw.ts
truncateTextToWidth/drawTextInArea, LGraphBadge.getWidth,
LGraphButton.getWidth, and textUtils.truncateText.
## Perf Impact
Expected: ~95% reduction in measureText calls (only cache misses on
first frame and value changes). Firefox has slower measureText than
Chrome, so this disproportionately benefits Firefox.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9404-fix-cache-ctx-measureText-results-to-avoid-redundant-calls-in-draw-loop-31a6d73d3650814e9cdac16949c55cb7)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
Add frontend support for `prompt_id` in `progress_text` binary WS
messages, enabling parallel workflow execution to route progress text to
the correct active prompt.
Backend PR: https://github.com/Comfy-Org/ComfyUI/pull/12540
## Changes
- Advertise `supports_progress_text_metadata` in client feature flags
- Decode `prompt_id` from new binary format when flag is active
- Add optional `prompt_id` to `zProgressTextWsMessage` schema
- Filter `progress_text` events by `activePromptId` — skip messages for
non-active prompts
## Deployment
Can be deployed independently in any order — feature flag negotiation
ensures graceful degradation.
Part of COM-12671
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9002-Add-prompt_id-support-to-progress_text-WS-messages-30d6d73d365081acabbcdac939e0c751)
by [Unito](https://www.unito.io)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Progress text updates can include optional metadata so richer context
is available.
* **Bug Fixes / Improvements**
* Progress updates are now filtered to show only the currently active
prompt, reducing cross-talk from concurrent operations and improving
update accuracy.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
Automatically clears transient validation errors
(`value_bigger_than_max`, `value_smaller_than_min`, `value_not_in_list`,
`required_input_missing`) when the user modifies a widget value or
connects an input slot, so resolved errors don't linger in the error
panel. Also clears missing model state when the user changes a combo
widget value.
## Changes
- **`useNodeErrorAutoResolve` composable**: watches widget changes and
slot connections, clears matching errors via `executionErrorStore`
- **`executionErrorStore`**: adds `clearSimpleNodeErrors` and
`clearSimpleWidgetErrorIfValid` with granular per-slot error removal
- **`executionErrorUtil`**: adds `isValueStillOutOfRange` to prevent
premature clearing when a new value still violates the constraint
- **`graphTraversalUtil`**: adds `getExecutionIdFromNodeData` for
subgraph-aware execution ID resolution
- **`GraphCanvas.vue`**: fixes subgraph error key lookup by using
`getExecutionIdByNode` instead of raw `node.id`
- **`NodeWidgets.vue`**: wires up the new composable to the widget layer
- **`missingModelStore`**: adds `removeMissingModelByWidget` to clear
missing model state on widget value change
- **`useGraphNodeManager`**: registers composable per node
- **Tests**: 126 new unit tests covering error clearing, range
validation, and graph traversal edge cases
## Screenshots
https://github.com/user-attachments/assets/515ea811-ff84-482a-a866-a17e5c779c39https://github.com/user-attachments/assets/a2b30f02-4929-4537-952c-a0febe20f02e
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9464-feat-auto-resolve-simple-validation-errors-on-widget-change-and-slot-connection-31b6d73d3650816b8afdc34f4b40295a)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Use the last previewable output as the batch cover/thumbnail instead
of the first, so the most recently generated image (e.g., `_00010`) is
shown as the representative
- Reverse output order in batch folder view so newest images appear at
the top
- Gitignore `.claude/worktrees` to fix knip scanning untracked worktree
copies
## Linked Issues
- Fixes#9354
- Related to #9080
## Test plan
- [ ] Generate a batch of images (e.g., 10 images) and verify the
sidebar shows the last generated image as the cover
- [ ] Expand the batch folder view and verify images are in
reverse-chronological order (newest first)
- [ ] Verify existing unit tests pass (`pnpm test:unit`)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9467-fix-show-most-recent-image-first-in-asset-sidebar-batch-view-31b6d73d365081cbaf30f81009c7fcfa)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
- Fixes#9319
- Add [fast-check](https://github.com/dubzzz/fast-check) property-based
testing with FSM (Finite State Machine) traversal to automatically
explore state combinations in the workflow persistence system
- Fix a real bug in `saveDraft()` discovered by the FSM test: orphan
cleanup in `loadIndex()` could delete a just-written payload when the
in-memory cache was empty
## Why this is needed
#9317 exposed a class of bug where two independently correct changes
interact to cause workflow loss. Conventional unit tests verify
specific, hand-picked scenarios and cannot catch these cross-PR
interaction bugs.
### AS IS (before)
| Aspect | Status |
|---|---|
| Testing approach | Example-based: developer picks specific inputs and
expected outputs |
| State coverage | Only explicitly written scenarios are tested |
| Cross-interaction bugs | Not detectable — each test runs one isolated
path |
| Bug in `saveDraft` | Undetected — `loadIndex()` orphan cleanup could
delete a just-written payload after `reset()` |
### TO BE (after)
| Aspect | Status |
|---|---|
| Testing approach | Property-based: fast-check generates **200 random
command sequences** per run |
| State coverage | Random exploration of `SaveDraft → GetDraft →
RemoveDraft → MoveDraft → GetMostRecentPath → Reset` combinations |
| Cross-interaction bugs | Detected automatically — fast-check shrinks
failing sequences to minimal reproductions |
| Bug in `saveDraft` | Found and fixed — `loadIndex()` now runs
**before** `writePayload()` to prevent orphan cleanup race |
## What fast-check does
fast-check is a property-based testing library. Instead of testing "does
this specific input produce this specific output?", it tests "does this
**property** hold for **all possible inputs**?"
For FSM testing specifically, fast-check:
1. Takes a set of **commands** (SaveDraft, GetDraft, RemoveDraft,
MoveDraft, GetMostRecentPath, Reset)
2. Generates **random sequences** of these commands
3. Runs each sequence against both a **model** (simplified oracle) and
the **real system** (store + localStorage)
4. Verifies **invariants** after every mutating command (index/payload
consistency, no orphans, LRU correctness, model agreement)
5. When a failure is found, **shrinks** the sequence to the minimal
reproduction
Example: the bug this PR fixes was shrunk to just 4 commands:
```
SaveDraft(d.json) → RemoveDraft(d.json) → Reset() → SaveDraft(a.json) ✗
```
## Changes
| File | Change |
|---|---|
| `package.json` / `pnpm-workspace.yaml` | Add `fast-check`
devDependency |
| `draftCacheV2.property.test.ts` | 7 property tests for pure index
functions |
| `workflowDraftStoreV2.fsm.test.ts` | FSM test: 6 commands, invariant
checking, 200 runs |
| `workflowDraftStoreV2.ts` | Fix: move `loadIndex()` before
`writePayload()` in `saveDraft()` |
## Test plan
- [x] `pnpm test:unit` — all 117 persistence tests pass (including 7
property + 1 FSM)
- [x] `pnpm typecheck` — clean
- [x] `pnpm lint` — clean
- [x] Pre-commit hooks pass (format, lint, typecheck)
- [x] Pre-push hook passes (knip)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9370-test-add-property-based-FSM-tests-for-workflow-persistence-3196d73d3650813daa98cdd8bef7e975)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
## Summary
Right-clicking a textarea widget (e.g. text node) shows the browser's
native context menu instead of ComfyUI's context menu, preventing access
to promote/un-promote options in subgraphs.
## Changes
- **What**: Replace `@contextmenu.capture.stop` on
`WidgetTextarea.vue`'s `<Textarea>` with a handler implementing
double-right-click toggling: first right-click shows ComfyUI's context
menu, second right-click (while menu is open) allows browser native
menu. Exposes `isNodeOptionsOpen()` from `useMoreOptionsMenu.ts` to
check menu state.
## Review Focus
The capture-phase handler in `WidgetTextarea.vue` only changes
`contextmenu` handling — pointer event modifiers
(`pointerdown/move/up.capture.stop`) that prevent canvas panning are
untouched. The double-right-click pattern matches Notion/YouTube
behavior for editable text fields.
<!-- Pipeline-Ticket: d7a53160-e1e1-42bb-a5ac-c0c2702c629c -->
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9840-fix-show-ComfyUI-context-menu-on-textarea-widget-right-click-3216d73d36508102b4c9c13a5915bc48)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Alexander Brown <drjkl@comfy.org>
## Summary
Adds VS Code-style multi-keybinding support to the Keybinding settings
panel. Commands can now have multiple keybindings displayed, expanded,
and individually managed.
- Fixes#1088
## Changes
### Store (`keybindingStore.ts`)
- `removeAllKeybindingsForCommand(commandId)` — unsets all bindings for
a command
- `updateSpecificKeybinding(old, new)` — replaces a single binding
without affecting others
- `resetKeybindingForCommand` — updated to restore **all** default
bindings, not just the first
- `isCommandKeybindingModified` — updated to compare full sorted sets of
bindings
### UI (`KeybindingPanel.vue`)
- **Data model**: `keybinding: KeybindingImpl | null` → `keybindings:
KeybindingImpl[]`
- **Multi-binding display**: shows up to 2 combos inline with `, `
separator, then `+ N more` badge
- **Expand/collapse**: click any row with 2+ bindings to expand
individual binding rows; chevron-right icon rotates on expand
- **Per-binding actions**: edit (pencil), reset, trash on each expanded
sub-row
- **Parent row actions**: `+`/trash for 2+ bindings, pencil/reset/trash
for 1, `+`/disabled for 0
- **Edit modes**: `edit` (replace specific binding via
`updateSpecificKeybinding`) and `add` (append via `addUserKeybinding`)
- **Right-click context menu**: Change keybinding, Add new, Reset to
default, Remove keybinding — with proper disabled states and lucide
icons
- **Remove all dialog**: confirmation via `showSmallLayoutDialog` with
`RemoveAllKeybindingsHeader`/`Content` components
- **Reset all dialog**: confirmation via `showConfirmDialog` before
resetting all keybindings to defaults
- **Double-click**: 0 bindings → add, 1 → edit, 2+ → no-op (single click
toggles expand)
- **Consistent alignment**: commands without chevron get `pl-5` padding
to align with those that have it
### Tests (`keybindingStore.test.ts`)
- 7 new tests covering `removeAllKeybindingsForCommand`,
`updateSpecificKeybinding`, multi-binding `isCommandKeybindingModified`,
and multi-binding `resetKeybindingForCommand`
### i18n (`main.json`)
- 11 new keys: removeAllKeybindingsTitle/Message, removeAll,
changeKeybinding, addNewKeybinding, resetToDefault, removeKeybinding,
nMoreKeybindings, resetAllKeybindingsTitle/Message, allKeybindingsReset
### New components
- `RemoveAllKeybindingsHeader.vue` — dialog header
- `RemoveAllKeybindingsContent.vue` — dialog body with Close/Remove all
buttons
## Test plan
- [x] `pnpm typecheck` passes
- [x] `pnpm lint` passes (no new errors)
- [x] `pnpm vitest run src/platform/keybindings/` — 45 tests pass
- [x] CodeRabbit review — 0 findings
- [ ] Manual: open Settings → Keybindings, verify multi-binding commands
(e.g. Delete Selected Items, Zoom In) show multiple combos
- [ ] Manual: click row to expand, verify per-binding actions work
- [ ] Manual: right-click row, verify context menu actions
- [ ] Manual: click trash on 2+ binding command, verify "Remove all"
confirmation dialog
- [ ] Manual: click "Reset All" button, verify confirmation dialog
appears
- [ ] Manual: add/edit/remove individual bindings, verify persistence
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9738-feat-multi-keybinding-support-in-settings-panel-3206d73d365081e9b08bd3cfe21495f1)
by [Unito](https://www.unito.io)
## Summary
Unify the search bar + action buttons layout across all left sidebar
panels (Node Library, Workflows, Model Library, Media Assets) using a
shared `SidebarTopArea` presentation component.
## Changes
- **What**:
- Add `SidebarTopArea.vue` — layout component with `flex-1` default slot
(search) and `#actions` slot (buttons), plus optional `bottomDivider`
prop
- Replace raw `<button>` elements in Node Library with `<Button
variant="secondary" size="icon">`
- Replace reka-ui `TabsTrigger` with shared `Tab/TabList` component in
Node Library
- Move Media Assets tab list from hover-only `#tool-buttons` to
always-visible header below search area
- Unify spacing (`gap-2`, `p-2 2xl:px-4`) and divider styles across all
sidebar panels
- Remove unused `assetType` prop and header from
`AssetsSidebarGridView`/`AssetsSidebarListView`
## Review Focus
- `SidebarTopArea` API simplicity — just slots + one optional prop
- Node Library still requires `TabsRoot` in the body for reka-ui
`TabsContent` in child panels
- Media Assets tabs are now always visible instead of hover-only
[screen-capture
(1).webm](https://github.com/user-attachments/assets/fe1d8f7b-5674-4bb3-9842-569e4c3af6c9)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9740-feat-unify-sidebar-panel-header-layout-with-SidebarTopArea-component-3206d73d365081ea8ba7fd6ac54e0169)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
## Summary
- Fixes a bug where widgets marked as `advanced` were always visible,
ignoring the "Always show advanced widgets on all nodes" setting
- Root cause: `extractWidgetDisplayOptions` in `useGraphNodeManager.ts`
read `widget.advanced` (always `undefined` on BaseWidget) instead of
`widget.options?.advanced` (where `litegraphService` actually sets the
flag)
- Consistent with how `hidden` is already read from
`widget.options.hidden` on the adjacent line
## Test plan
- [ ] Load a node with advanced inputs (e.g. `LTXVScheduler`)
- [ ] Verify `max_shift`, `base_shift`, `stretch`, `terminal` are hidden
when "Always show advanced widgets on all nodes" is disabled
- [ ] Verify they become visible when the setting is enabled or the
per-node toggle is clicked
- [ ] Verify the advanced toggle button appears on nodes with advanced
widgets
🤖 Generated with [Claude Code](https://claude.com/claude-code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9857-fix-advanced-widgets-always-visible-regardless-of-setting-3226d73d36508132b338d098fbf19433)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
## Problem
The lint/format CI workflow was broken for fork PRs in two ways:
### 1. Node version mismatch in setup-frontend action
The `setup-frontend` shared action (created in #8377) was missed when
Node version was standardized to `.nvmrc` in #9521. It still used
`node-version: 'lts/*'` instead of `node-version-file: '.nvmrc'`.
### 2. Fork PRs with lint issues silently passed CI
Fork PRs with auto-fixable lint/format issues got a **green checkmark**
despite having unfixed issues:
1. Auto-fix steps (`lint:fix`, `format`) fix issues in the workspace
2. `Commit changes` is correctly skipped for forks (can't push to fork
branches)
3. `Final validation` passes because it runs on the already-fixed
workspace
4. The `Comment on PR about manual fix needed` step tries to post a
comment via `actions/github-script`, but fork PRs have a read-only
`GITHUB_TOKEN` — the comment silently fails (`continue-on-error: true`)
5. **Result**: workflow reports success, contributor thinks their code
is clean
## Fix
- **setup-frontend**: Use `node-version-file: '.nvmrc'` instead of
`node-version: 'lts/*'`
- **ci-lint-format**: Replace the broken fork comment step with an
explicit `exit 1` that fails CI and prints clear fix instructions in the
log. This follows the principle from `.github/AGENTS.md`: fork PRs can't
post comments, so don't try.
## Testing
- [ ] Verify fork PRs with clean code still pass
- [ ] Verify fork PRs with lint issues now properly fail (instead of
silently passing)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9846-fix-restore-fork-PR-lint-format-CI-workflow-3226d73d3650811cb5bfe9f1f989cc0c)
by [Unito](https://www.unito.io)
## Summary
Add a setting to select all children (nodes, reroutes, nested groups)
when clicking a group on the canvas.
## Changes
- **What**: New `LiteGraph.Group.SelectChildrenOnClick` boolean setting
(default: `false`). When enabled, selecting a group cascades `select()`
to all its `_children`, and deselecting cascades `deselect()`. Recursion
handles nested groups naturally. No double-move risk — the drag handler
already uses `skipChildren=true`. The setting is wired via `onChange` to
`canvas.groupSelectChildren`, keeping litegraph free of platform
imports.
## Review Focus
- The select/deselect cascading in `LGraphCanvas.select()` /
`deselect()` — verify no infinite recursion risk with deeply nested
groups.
- The `groupSelectChildren` property is set via the setting's `onChange`
callback on `LGraphCanvas.active_canvas` — confirm this covers canvas
re-creation scenarios.
## Screenshots (if applicable)
N/A — behavioral change behind a setting toggle.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9149-feat-select-group-children-on-click-3116d73d365081a1a7b8c82dea95b242)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
All three watchers used { deep: true } unnecessarily because their
watched sources already produce new object/array references on change,
making deep traversal redundant:
- GraphView.vue: queueStore.tasks is a computed that spreads three
shallowRef arrays into a new array each time. TaskItemImpl instances
are immutable (readonly fields, replaced not mutated).
- DomWidget.vue: Replaced opaque widgetState deep watcher with explicit
property deps (pos, size, zIndex, readonly, positionOverride). During
60 FPS pan/zoom, Vue no longer walks the entire DomWidgetState object
graph including the markRaw widget — only 5 leaf properties are
checked. pos and size are new array literals each frame; the rest are
primitives.
- GraphCanvas.vue: nodeLocationProgressStates is a computed returning a
new Record on every execution progress event (nodeProgressStates is
replaced wholesale via WebSocket handler).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9248-perf-remove-deep-true-from-3-hot-watchers-to-reduce-reactivity-overhead-3136d73d365081278f18da5a2eef6971)
by [Unito](https://www.unito.io)
Co-authored-by: bymyself <cbyrne@comfy.org>
## Summary
- Replace PrimeVue `ColorPicker` with a custom component built on Reka
UI Popover
- New `ColorPicker` supports HSV saturation-value picking, hue/alpha
sliders, hex/rgba display toggle
- Simplify `WidgetColorPicker` by removing PrimeVue-specific
normalization logic
- Add Storybook stories for both `ColorPicker` and `WidgetColorPicker`
## Test plan
- [x] Unit tests pass (9 widget tests, 47 colorUtil tests)
- [x] Typecheck passes
- [x] Lint passes
- [ ] Verify color picker visually in Storybook
- [ ] Test color picking in node widgets with hex/rgb/hsb formats
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9647-feat-replace-PrimeVue-ColorPicker-with-custom-component-31e6d73d36508114bc54d958ff8d0448)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
## Summary
Inline splash screen CSS into `index.html` to fix broken loading
animation on cloud/ephemeral environments.
## Changes
- **What**: On cloud/ephemeral environments (e.g.
`fe-pr-*.testenvs.comfy.org`), SPA fallback serves `index.html` for
unknown paths. The `<link href="splash.css">` request resolves to
`/cloud/splash.css`, which the server does not find as a static file —
so it returns `index.html` with `200 OK`. The browser receives HTML
instead of CSS, the CSS parser silently ignores it, and the splash
screen renders without any styles or animations.
- Inlined `splash.css` directly into `index.html` `<style>` block —
eliminates the external request entirely
- Moved `splash.css` to `src/assets/` for content-hashed Vite processing
as source of truth
- Removed `public/splash.css`
## Review Focus
- The inline CSS is byte-for-byte identical to the original
`public/splash.css`
- `src/assets/splash.css` preserved as canonical source for future
changes
[screen-capture
(1).webm](https://github.com/user-attachments/assets/06729641-d1fd-47aa-9dd4-4acd28c2cfcf)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9849-fix-inline-splash-CSS-to-prevent-SPA-fallback-breakage-on-cloud-environments-3226d73d365081418741eb0944a74977)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
## Summary
Fix download icon not appearing when file size is successfully fetched
in the missing models dialog.
## Changes
- **What**: Restructured the `v-if/v-else-if` chain in
`MissingModelsContent.vue` so that file size and download icon render
together instead of being mutually exclusive. Previously, a successful
file size fetch would prevent the download button from rendering.
## Review Focus
The file size span and download/gated-link are now inside a shared
`<template v-else-if="model.isDownloadable">` block. File size uses
`v-if` (independent), while gated link and download button remain
`v-if/v-else` (mutually exclusive with each other).
[screen-capture.webm](https://github.com/user-attachments/assets/f2f04d52-265b-4d05-992e-0ffe9bf64026)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9850-fix-show-download-icon-alongside-file-size-in-missing-models-dialog-3226d73d365081fd943bcfdedda87c73)
by [Unito](https://www.unito.io)
## Summary
Bake the frontend git commit hash into the build so it no longer needs
to be fetched from the server via `/api/system_stats`.
## Changes
- **What**: Add `__COMFYUI_FRONTEND_COMMIT__` build-time constant (via
Vite `define`) sourced from `git rev-parse HEAD` at build time. Falls
back to `"unknown"` if git is unavailable. `SystemStatsPanel` uses this
baked-in value for the "Frontend Version" row in cloud mode instead of
the server-provided `comfyui_frontend_version` field.
## Testing
Confirmed to display the actual commit.
<img width="1448" height="908" alt="Screenshot 2026-03-12 at 7 09 52 PM"
src="https://github.com/user-attachments/assets/2b42348a-5c3e-4509-aa84-1a259bba5f3f"
/>
## Review Focus
- The `getDisplayValue` override for `comfyui_frontend_version` —
cleanest way to swap the data source without restructuring the column
system.
- No cloud-side changes needed: the `sync-frontend-build` workflow
already checks out the frontend repo at the exact commit ref, so `git
rev-parse HEAD` returns the correct hash.
## Summary
Fix asset browser sidebar missing categories because `typeCategories`
assumed a fixed tag order from the API.
## Changes
- **What**: Replace `tags[0] === 'models'` / `tags[1]` with
`tags.includes(MODELS_TAG)` and `flatMap`+`filter`, matching the pattern
used by `getAssetModelFolders` and `filterByCategory`.
## Review Focus
The API returns tags in arbitrary order (e.g. `['checkpoints',
'models']` instead of `['models', 'checkpoints']`). The old code
filtered out most assets, resulting in an empty sidebar. New test
validates arbitrary tag ordering.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9843-fix-use-order-independent-tag-matching-in-asset-browser-categories-3216d73d365081b886f3d5ab1790e19d)
by [Unito](https://www.unito.io)
Co-authored-by: Amp <amp@ampcode.com>
This has semi-significant performance impact if you use the same
workflow for more than a day. I left Comfy running with a mouse jiggler
and this reduced my instance to a crawl after about an hour.
`nodeProgressStatesByJob` accumulated an entry for every job that ever
executed during a session. In long-running sessions this grew without
bound, since entries were only removed for the active job on
`resetExecutionState`.
Add `MAX_PROGRESS_JOBS` (1000) and `evictOldProgressJobs()`, called
after each `handleProgressState` update. When the map exceeds the limit,
the oldest entries (by ES2015+ insertion order) are pruned — keeping
only the most recent 1000. This mirrors the pattern used by
assetsStore's `MAX_HISTORY_ITEMS`.
Also adds tests for:
- nodeLocationProgressStates computed reactivity (recomputes on
wholesale replacement, produces new references)
- Eviction behavior (retains below limit, evicts oldest above limit,
preserves most recent, no-op when updating existing job)
- API event handler wiring via captured apiEventHandlers map
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9249-fix-cap-nodeProgressStatesByJob-to-prevent-unbounded-memory-growth-3136d73d365081e49b36d8ade0d4dd6e)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
Muted (NEVER mode) subgraph nodes throw "No inner node DTO found" during
prompt serialization because `resolveOutput()` falls through to subgraph
resolution for nodes whose inner DTOs were never registered.
## Changes
- **What**: Add early return in `ExecutableNodeDTO.resolveOutput()` for
`NEVER` mode nodes, matching the existing `BYPASS` mode guard. Add 5
tests covering muted, bypassed, and normal mode resolution.
## Review Focus
The fix is a single-line early return. The key insight is that
`graphToPrompt` in `executionUtil.ts` correctly skips `getInnerNodes()`
for muted/bypassed nodes, so their inner DTOs are never in the map — but
`resolveOutput()` was missing the corresponding guard for `NEVER` mode.
Fixes#8986
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9302-fix-return-undefined-for-muted-node-output-resolution-3156d73d3650811e9697c7281f11cf96)
by [Unito](https://www.unito.io)
## Summary
Adds `auxclick` event listener to prevent the browser's default
middle-click paste behavior on Linux systems.
**Problem:** On Linux, middle-clicking anywhere triggers a paste from
the PRIMARY clipboard. When middle-dragging to pan the canvas, this
causes the entire workflow to be duplicated as new nodes on mouse
release.
**Solution:** Add `auxclick` event listener with `preventDefault()` to
the graph canvas, blocking the paste while preserving pan functionality.
## Changes
- Add `auxclick` event listener in `bindEvents()`
- Add corresponding `removeEventListener` in `unbindEvents()`
## Test Plan
- [ ] On Linux: Middle-drag to pan canvas - should pan without
duplicating nodes
- [ ] On Linux: Verify left/right click behaviors unchanged
- [ ] On Windows/macOS: Verify no regression (auxclick should have no
effect)
Fixes#4464
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-8259-fix-prevent-middle-click-paste-duplicating-workflow-on-Linux-2f16d73d3650812b98f9cada699f5508)
by [Unito](https://www.unito.io)
---------
Co-authored-by: bymyself <cbyrne@comfy.org>
## Summary
Update workspace creation modal copy to clarify that creating a
workspace establishes a **new** credit pool, rather than sharing from
the owner's existing credits.
## Changes
- **What**: Changed i18n message from "Workspaces let members share a
single credits pool" to "Workspaces create a new credit pool that can be
shared among members"
## Review Focus
Copy change only — single i18n string update in
`src/locales/en/main.json`.
Fixes COM-16521
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9811-fix-update-workspace-creation-modal-phrasing-for-credit-pool-clarity-3216d73d36508186850ac5a8ad97461d)
by [Unito](https://www.unito.io)
## Summary
- When bulk exporting, `job_asset_name_filters` was always sent for
every job, restricting each job to only the assets the user clicked on.
For multi-output jobs, this meant the ZIP only contained 1 asset per job
instead of all outputs.
- Now compares selected asset count per job against `outputCount`
metadata and omits the filter for fully-selected jobs, so the backend
returns all assets.
## Test plan
- [x] Unit tests: all outputs selected → no filter sent
- [x] Unit tests: subset selected → filter sent
- [x] Unit tests: mixed selection → filter only for partial jobs
- [x] Unit tests: multiple fully-selected jobs → no filters
- [x] Typecheck, lint pass
- [ ] Manual: bulk export multi-output jobs in cloud env, verify ZIP
contains all outputs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9684-fix-omit-job_asset_name_filters-when-all-job-outputs-selected-31f6d73d3650814482f8c59d05027d79)
by [Unito](https://www.unito.io)
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
## Summary
nodeLocationProgressStates runs executionIdToNodeLocatorId for every
ancestor prefix of every node's display_node_id on every WebSocket
progress update. Each call traverses the subgraph hierarchy via
getNodeById. For nested subgraphs with many executing nodes, this
results in O(N × D²) graph lookups per progress tick, where N is
the number of executing nodes and D is the nesting depth.
Add a plain Map cache inside the execution store that memoizes
executionIdToNodeLocatorId results by execution ID string. The
cache persists across computed re-evaluations within a single
execution run, reducing subsequent progress updates to O(1)
lookups per execution ID.
Cache invalidation:
- Cleared at execution start (handleExecutionStart) to ensure
fresh graph state for each new run
- Cleared at execution end (resetExecutionState) to prevent
stale entries leaking across runs
The cache stores strings only (no graph node references), with
typical size of ~50-200 entries per run, so memory impact is
negligible.
- **What**: Caches a mapping of ids to ids, preventing an exponential
re-scan of subgraphs
- **Breaking**: Nothing
- **Dependencies**: None
You will need to set the feature flag
`ff:expose_executionId_to_node_locator_id_cache_counters` from the
underlying counter PR if you want to measure the impact.
```js
localStorage.setItem(
'ff:expose_executionId_to_node_locator_id_cache_counters',
'true'
)
```
## Review Focus
Before pulling this PR, pull
https://github.com/Comfy-Org/ComfyUI_frontend/pull/9243 , set the
feature flag, and run a workflow with nested subgraphs, then look in
console to see a cache measurement.
Next, pull this PR and load the same workflow. Note the massive
reduction in visits.
## Screenshots
Login problems due to cross-opener policy are preventing me from taking
screenshots from a local dev build at this time
## Thread
There isn't one. I don't have access to AmpCode or Unito.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9244-Cache-execution-id-to-node-locator-id-mappings-3136d73d365081e680eae3c891480ee7)
by [Unito](https://www.unito.io)
Co-authored-by: bymyself <cbyrne@comfy.org>
## Summary
Add Sentry breadcrumbs to subgraph proxy widget operations for better
observability of widget state changes.
## Changes
- **What**: Add `Sentry.addBreadcrumb()` calls with category
`'subgraph'` to `promoteWidget`, `demoteWidget`, and `pruneDisconnected`
in `proxyWidgetUtils.ts`
## Review Focus
Breadcrumbs are info-level and don't affect control flow. They log
widget name/node ID for promote/demote and removed count for prune.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-8996-chore-add-Sentry-breadcrumbs-to-subgraph-proxy-widget-operations-30d6d73d365081a5abbccabd39fc7129)
by [Unito](https://www.unito.io)
## Summary
Adds a comprehensive `TROUBLESHOOTING.md` guide to help developers
resolve common development issues.
## Motivation
a developer reported issues where `pnpm dev` would get stuck on 'nx
serve'. This highlighted the need for centralized troubleshooting
documentation to help developers quickly resolve common issues without
having to wait for help.
## Changes
- Created `TROUBLESHOOTING.md` with FAQ-style documentation
- Added Mermaid flowchart for quick issue diagnosis
- Documented solutions for common problems:
- Development server issues (nx serve hanging)
- Build and TypeScript errors
- Dependency and package management problems
- Testing issues
- Git and branch conflicts
## Structure
The guide includes:
- Quick diagnostic flowchart (Mermaid)
- Frequently Asked Questions with:
- Clear symptoms
- Step-by-step solutions
- Explanations of why issues occur
- Links to community support resources
- Contribution guidelines
## Test Plan
- [x] File created and committed
- [x] Mermaid flowchart renders correctly
- [x] All commands are accurate and tested
- [x] Links to Discord and GitHub are valid
🤖 Generated with [Claude Code](https://claude.com/claude-code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7738-docs-Add-TROUBLESHOOTING-md-guide-for-common-development-issues-2d26d73d365081eda291e619b3067bc6)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: snomiao <snomiao@gmail.com>
Co-authored-by: GitHub Action <action@github.com>
## Summary
When many nodes are rendered in the transform container, both zoom and
pan can cause FPS drops because the browser re-rasterizes all visible
content at the new transform. `will-change: transform` tells the browser
to keep the layer as a GPU texture and skip re-rasterization during
active interaction, restoring visual quality only after settling.
- Add pointer drag detection so `will-change: transform` covers pan in
addition to zoom. Without this, dragging with 256+ nodes causes jank as
the browser re-rasterizes the entire layer on every frame of the pan.
- Fix settleDelay from 16ms to 256ms. At 16ms the debounce fires between
consecutive wheel events (~50ms apart on a physical mouse), causing
`will-change` to toggle on/off rapidly. Each toggle forces the browser
to promote/demote the compositor layer, which is more expensive than not
having the optimization at all.
- Replace scoped CSS with Tailwind `will-change-transform`.
- Remove per-node `will-change: transform` on `.lg-node`. Promoting each
node to its own compositor layer (256 nodes = 256 GPU textures)
increases memory pressure and compositing overhead, making performance
worse than a single promoted container.
- Previously, the virtual DOM of Nodes was updated during zooming and
dragging, but now this update is avoided through some techniques.
- Using the 3D versions of scale and translate can provide a smoother
experience when dealing with a large number of nodes.
## Test plan
- [x] Unit tests updated and passing
- [x] Manual: verify during both zoom and pan
- [x] Manual: compare pan FPS with 256 nodes before/after
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9649-perf-detect-pointer-drag-in-useTransformSettling-for-pan-optimization-31e6d73d3650818bb2c3ccd01a465140)
by [Unito](https://www.unito.io)
---------
Co-authored-by: github-actions <github-actions@github.com>
Previously, MatchType and Autogrow inputs would not be considered would
filtering searchbox entires. For example, "Batch Images" would not show
as a suggestion would dragging a noodle from a "Load Image" node.
This is resolved by adding a step during nodeDef registration to
precalculate a list of all input types. This may have performance
implications.
- Search filtering should be more performant
- Initial node registration will be slower
- There's additional memory cost to store this information on every
node.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9388-Support-search-filtering-to-dynamic-input-types-3196d73d365081d9939eff5e167a7e83)
by [Unito](https://www.unito.io)
## Summary
Centralize the inline `display_name || name` pattern into
`getAssetDisplayName`, adding `display_name` to the existing metadata
fallback chain.
## Changes
- **What**: Update `getAssetDisplayName` fallback chain to
`user_metadata.name → metadata.name → display_name → name`. Replace all
6 inline `asset.display_name || asset.name` call sites with the shared
utility. Remove duplicate local function in `AssetsSidebarListView.vue`.
## Review Focus
The fallback order preserves user_metadata overrides while incorporating
the `display_name` field added in #9626. All callers now go through a
single code path.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9641-refactor-centralize-display_name-name-into-getAssetDisplayName-31e6d73d365081e09e5de85486583443)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
Co-authored-by: GitHub Action <action@github.com>
## Summary
Wires the nightly survey system into the app by adding a controller
component and a convenience composable for feature-site usage tracking.
## Changes
- **What**: NightlySurveyController iterates enabled surveys from the
registry and renders a NightlySurveyPopover for each.
useSurveyFeatureTracking wraps useFeatureUsageTracker with a
config-enabled guard for use at feature call sites.
- **Tree-shaking**: Controller is loaded via defineAsyncComponent behind
a compile-time isNightly/isCloud/isDesktop guard in SideToolbar.vue, so
the entire survey module subtree is eliminated from cloud/desktop/stable
builds.
## Review Focus
- DCE pattern: controller imported conditionally via
defineAsyncComponent + distribution guard (same pattern as
ComfyRunButton/index.ts)
- useSurveyFeatureTracking short-circuits early when config is
absent/disabled (avoids initializing tracker storage)
- No user-facing behavior change: FEATURE_SURVEYS registry is still
empty
## Part of Nightly Survey System
This is part 5 of a stacked PR chain:
1. feat/feature-usage-tracker - useFeatureUsageTracker (merged in #8189)
2. feat/survey-eligibility - useSurveyEligibility (#8189, merged)
3. feat/survey-config - surveyRegistry.ts (#8355, merged)
4. feat/survey-popover - NightlySurveyPopover.vue (#9083, merged)
5. **feat/survey-integration** - NightlySurveyController.vue (this PR)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
Fix input asset previews (images/videos) disappearing from
LoadImage/LoadVideo nodes after execution and a browser tab switch.
## Changes
- **What**: Guard `setOutputsByLocatorId` in `imagePreviewStore` to
preserve existing input-type preview images (`type: 'input'`) when the
incoming execution output has no images. Execution outputs with actual
images still overwrite as expected.
## Review Focus
- The guard only applies when existing output is an input preview (`type
=== 'input'` for all images) AND incoming output has no images — this is
the exact scenario where execution clobbers upload widget previews.
- Root cause: execution results from the backend overwrite the upload
widget's synthetic preview for LoadImage/LoadVideo nodes (which produce
no output images). Combined with the deferred resize-observer
re-observation from PR #8805, returning to a hidden tab reads the
now-empty store entry.
## Summary
- When a media node (LoadImage/LoadAudio/LoadVideo) is selected and the
clipboard contains stale node metadata from a prior Ctrl+C, pasting
skips the node-metadata deserialization so that the paste falls through
to litegraph's default handler instead of incorrectly pasting the old
copied node.
- FixesComfy-Org/ComfyUI#12896
## Root Cause
The paste handler in `usePaste.ts` checks clipboard `text/html` for
`data-metadata` (serialized node data) **before** falling through to
litegraph's default paste. When a user copies a node, then copies a web
image, the browser clipboard may retain the old `data-metadata` in
`text/html` while the image data is not available as a
`DataTransferItem` file. This causes the stale node to be pasted instead
of the image.
## Fix
Skip `pasteClipboardItems()` when a media node is selected, allowing the
paste to fall through to litegraph's default handler which can handle
the clipboard content appropriately.
## Test plan
- [x] Added unit test verifying node metadata paste is skipped when
media node is selected
- [x] Manual: Copy a node → copy a web image → select LoadImage node →
Ctrl+V → verify image is pasted, not the node
## AS IS
https://github.com/user-attachments/assets/210d77d3-5c49-4e38-91b7-b9d9ea0e7ca0
## TO BE
https://github.com/user-attachments/assets/b68e4582-0c57-48b8-9ed9-0b3243bb1554🤖 Generated with [Claude Code](https://claude.com/claude-code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9773-fix-skip-node-metadata-paste-when-media-node-is-selected-3216d73d3650814d92dadcd0c0ec79c7)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
Adds a reusable agent skill for managing cherry-pick backports across
stable release branches.
## What
Agent skill at `.claude/skills/backport-management/` with routing-table
SKILL.md + 4 reference files (discovery, analysis, execution, logging).
## Why
Codifies lessons from backporting 57 PRs across cloud/1.41, core/1.41,
and core/1.40. Makes future backport sessions faster and less
error-prone.
## Key learnings baked in
- Cloud-only PRs must not be backported to `core/*` branches (wasted
effort)
- Wave verification (`pnpm typecheck`) between batches to catch breakage
early
- Human review required for non-trivial conflict resolutions before
admin-merge
- MUST vs SHOULD decision guide with clear criteria
- Continuous backporting preference over bulk sessions
- Mermaid diagram as final session deliverable
- Conflict triage table (never skip based on file count alone)
- `gh api` for labels instead of `gh pr edit` (Projects Classic
deprecation)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9619-chore-add-backport-management-agent-skill-31d6d73d3650815b9808c3916b8e3343)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
Asset widget dropdown search only matched against `item.name`
(filename), but users see `item.label` (display name). Now searches both
fields so filtering matches what is visually displayed.
## Changes
- **What**: `defaultSearcher` in `FormDropdown` now matches against both
`name` and `label` fields
- Added 3 unit tests covering label-based search scenarios
## Review Focus
- The change only affects cloud asset mode where `name` (filename) and
`label` (display name) differ. In local mode, `label` is either
`undefined` or identical to `name`, so behavior is unchanged.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9774-bugfix-Asset-widget-search-matches-display-label-3216d73d365081ca8befdf7260c66a26)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
- Fixes incorrect workflow loading after page refresh when the active
workflow is saved and unmodified
- Adds saved-workflow fallback in `loadPreviousWorkflowFromStorage()`
before falling back to latest draft
- Calls `openWorkflow()` in `restoreWorkflowTabsState()` to activate the
correct tab after restoration
- Fixes#9317
## Test plan
- [x] Unit tests for saved-workflow fallback (draft missing, saved
workflow exists)
- [x] Unit tests for correct tab activation after restoration
- [x] Unit tests for existing behavior preservation (draft preferred
over saved, no-session-path fallback)
- [x] `pnpm typecheck` passes
- [x] `pnpm lint` passes
- [x] All 117 persistence tests pass
🤖 Generated with [Claude Code](https://claude.com/claude-code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9318-fix-restore-correct-workflow-on-page-reload-3166d73d365081ba9139f7c23c917aa4)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
Co-authored-by: GitHub Action <action@github.com>
## Summary
When a workflow is loaded with missing models, users currently have no
way to identify or resolve them from within the UI. This PR adds a full
missing-model detection and resolution pipeline that surfaces missing
models in the Errors tab, allowing users to install or import them
without leaving the editor.
## Changes
### Missing Model Detection
- Scan all COMBO widgets across root graph and subgraphs for model-like
filenames during workflow load
- Enrich candidates with embedded workflow metadata (url, hash,
directory) when available
- Verify asset-supported candidates against the asset store
asynchronously to confirm installation status
- Propagate missing model state to `executionErrorStore` alongside
existing node/prompt errors
### Errors Tab UI — Model Resolution
- Group missing models by directory (e.g. `checkpoints`, `loras`, `vae`)
with collapsible category cards
- Each model row displays:
- Model name with copy-to-clipboard button
- Expandable list of referencing nodes with locate-on-canvas button
- **Library selector**: Pick an alternative from the user's existing
models to substitute the missing model with one click
- **URL import**: Paste a Civitai or HuggingFace URL to import a model
directly; debounced metadata fetch shows filename and file size before
confirming; type-mismatch warnings (e.g. importing a LoRA into
checkpoints directory) are surfaced with an "Import Anyway" option
- **Upgrade prompt**: In cloud environment, free-tier subscribers are
shown an upgrade modal when attempting URL import
- Separate "Import Not Supported" section for custom-node models that
cannot be auto-resolved
- Status card with live download progress, completion, failure, and
category-mismatch states
### Canvas Integration
- Highlight nodes and widgets that reference missing models with error
indicators
- Propagate missing-model badges through subgraph containers so issues
are visible at every graph level
### Code Cleanup
- Simplify `surfacePendingWarnings` in workflowService, remove stale
widget-detected model merging logic
- Add `flattenWorkflowNodes` utility to workflowSchema for traversing
nested subgraph structures
- Extract `MissingModelUrlInput`, `MissingModelLibrarySelect`,
`MissingModelStatusCard` as focused single-responsibility components
## Testing
- Unit tests for scan pipeline (`missingModelScan.test.ts`): enrichment,
skip-installed, subgraph flattening
- Unit tests for store (`missingModelStore.test.ts`): state management,
removal helpers
- Unit tests for interactions (`useMissingModelInteractions.test.ts`):
combo select, URL input, import flow, library confirm
- Component tests for `MissingModelCard` and error grouping
(`useErrorGroups.test.ts`)
- Updated `workflowService.test.ts` and `workflowSchema.test.ts` for new
logic
## Review Focus
- Missing model scan + enrichment pipeline in `missingModelScan.ts`
- Interaction composable `useMissingModelInteractions.ts` — URL metadata
fetch, library install, upload fallback
- Store integration and canvas-level error propagation
## Screenshots
https://github.com/user-attachments/assets/339a6d5b-93a3-43cd-98dd-0fb00681b66f
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9743-feat-Surface-missing-models-in-Errors-tab-Cloud-3206d73d365081678326d3a16c2165d8)
by [Unito](https://www.unito.io)
## Summary
Fix all app mode widgets (including seed) disappearing after hard
refresh due to a race condition in `pruneLinearData` and a missing
reactivity dependency in `mappedSelections`.
## Changes
- **What**: Guard `pruneLinearData` with `!ChangeTracker.isLoadingGraph`
so inputs are preserved while `rootGraph.configure()` hasn't populated
nodes yet. Add `graphNodes` dependency to `mappedSelections` computed in
`LinearControls.vue` so it re-evaluates when the graph finishes
configuring.
## Review Focus
The core fix is a one-line guard change: `app.rootGraph &&
!ChangeTracker.isLoadingGraph` instead of just `app.rootGraph`. The
previous guard failed because `rootGraph` exists as an empty graph
during loading — `resolveNode()` returns `undefined` for all nodes and
everything gets filtered out.
Fixes COM-16193
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9621-fix-app-mode-widgets-disappear-after-hard-refresh-31d6d73d3650811193f5e1bc8f3c15c8)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
- align the linear-mode `DropZone` upload indicator with the Figma file
upload states
- add a co-located Storybook story for the default and hover variants
- add a `forceHovered` preview prop so Storybook can render the hover
state deterministically
## Validation
- `pnpm typecheck` (run in the original workspace with dependencies
installed)
- `pnpm lint` (passes with one pre-existing warning in
`src/lib/litegraph/src/ContextMenu.ts`)
- Storybook smoke check is currently blocked by an existing workspace
issue: `vite-plugin-inspect` fails with `Can not found environment
context for client`
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9690-feat-add-DropZone-Storybook-coverage-for-file-upload-states-31f6d73d365081ae9eabdde6b5915f26)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
## Summary
Rename the `Comfy.Queue.QPOV2` settings label to `Docked job
history/queue panel` to improve searchability/discoverability in the
settings UI.
## Changes
- **What**: Updated the visible setting name in the core settings
definition and the English locale string.
## Review Focus
The change is intentionally limited to the display label. The persisted
setting key remains `Comfy.Queue.QPOV2`, so existing user configuration
is preserved.
## Screenshots (if applicable)
Not applicable.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9620-fix-rename-docked-queue-panel-setting-31d6d73d36508189a2d1d3a621739a22)
by [Unito](https://www.unito.io)
## Summary
This makes the focus ring only appear on keyboard navigation (Tab), not
on mouse click for widgets like toggle switches, while text inputs still
show the ring on click since browsers apply ` :focus-visible` to them.
## Mozilla Standard
Legacy `focus-within` triggers a highlight ring on every mouse click,
creating unnecessary visual noise during canvas navigation. Following
[MDN
standards](https://developer.mozilla.org/en-US/docs/Web/CSS/:focus-visible),
`:focus-visible` only triggers the highlight when the browser determines
a visual cue is needed (e.g., keyboard navigation).
Using the `:has()` relational selector allows the container to react to
the state of its children natively in CSS. Removes the need for Vue
event listeners or complex state bubbling to highlight the field border.
This reduces JavaScript overhead and simplifies component logic. FYI
[MDN
:has()](https://developer.mozilla.org/en-US/docs/Web/CSS/Reference/Selectors/:has).
Reordered Tailwind classes to move `transition-all` to the end,
following official best practices. Groups layout/shape first, followed
by interaction states, and finally animations. This improves code
readability and maintainability.
## Screenshots
before
<img width="558" height="252" alt="12efd5721fb792a7e2dab7e022c2bed6"
src="https://github.com/user-attachments/assets/f881fe13-9f4f-40fd-a8cc-f438b1ba4bde"
/>
after
<img width="538" height="235" alt="e5ffec0a34d3b237c4fca9818ec598dd"
src="https://github.com/user-attachments/assets/5ada4112-64bd-48a4-9e9c-b59de6984370"
/>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9456-fix-update-WidgetLayoutField-border-styling-31b6d73d36508193a31ed02bfdef414f)
by [Unito](https://www.unito.io)
Co-authored-by: Terry Jia <terryjia88@gmail.com>
## Summary
Show tab-specific empty state messages instead of a misleading registry
connection error on tabs that don't depend on the Manager API.
## Changes
- **What**: Scoped `comfyManagerStore.error` to only affect tabs that
actually use the Manager API (AllInstalled, UpdateAvailable,
Conflicting). Missing and Workflow tabs use the registry directly, so
the Manager API error is irrelevant to them. Previously, any prior
Manager API failure would cause `"Error connecting to the Comfy Node
Registry"` to appear on the Missing tab even when there are simply no
missing nodes.
## Review Focus
The `isManagerErrorRelevant` computed only shows the error for Manager
API-dependent tabs. Verify that AllInstalled/UpdateAvailable/Conflicting
still correctly show the error when Manager API fails.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9640-fix-show-correct-empty-state-on-Missing-tab-instead-of-misleading-registry-error-31e6d73d365081868da1d226a9bb5fdc)
by [Unito](https://www.unito.io)
## Summary
Image input changes (dropdown selection and file upload) in app/linear
mode did not create their own undo entries, causing undo to skip or
bundle image changes with subsequent actions.
## Changes
- **What**: Add explicit `checkState()` calls in
`WidgetSelectDropdown.vue` after `modelValue` is set in
`updateSelectedItems()` (dropdown selection) and `handleFilesUpdate()`
(file upload), ensuring each image change gets its own undo entry.
## Review Focus
The fix is intentionally scoped to `WidgetSelectDropdown` rather than
the generic `updateHandler` in `NodeWidgets.vue`, which would create
excessive undo entries for text inputs. The pattern follows existing
usage in `useSelectedNodeActions.ts` and other composables.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9623-fix-call-checkState-after-image-input-changes-for-proper-undo-tracking-31d6d73d3650814781dbca5db459ab6d)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
Restores `widget.inputEl` assignment on STRING multiline widgets that
was removed in commit a7c211516 (PR #8594) when it was renamed to
`widget.element`. Custom nodes (e.g. comfyui-custom-scripts) rely on
`widget.inputEl` to call `addEventListener` or set `readOnly`.
- FixesComfy-Org/ComfyUI#12893
## Test plan
- Verify custom nodes that access `widget.inputEl` on STRING widgets
work correctly
- Verify `widget.element` still works as before
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
Fix#9732
To clarify how preventing the 60 FPS object assignment solves the
`vue-i18n` (intlify) issue, here is the complete chain reaction leading
to the performance loop:
1. The Root Cause: In `useMinimapViewport.ts`, `useRafFn` acts as a
timer bound to the browser's **refresh rate** (60 FPS). In the original
code, it unconditionally executed the `viewportTransform.value = { ... }
`assignment 60 times a second.
2. Vue's Reactivity Interception: Because `viewportTransform` is a
reactive variable (`ref`), updating it causes its corresponding
**computed** property (`viewportStyles`) to register a data dependency
update.
3. Forced Re-rendering: The `<template> ` in `MiniMap.vue` is bound to
`:style="viewportStyles"`. Since the dependent value changed, Vue's
Virtual DOM decides: "I must re-render the entire `MiniMap.vue`
interface **60 times** per second to ensure the element positions are
up-to-date!"
4. The Victim Emerges: Inside the template of `MiniMap.vue`, there are
several internationalization translation functions: `<button
:aria-label="$t('g.settings')" ...> <button :aria-label="$t('g.close')"
...> `In Vue, whenever a component re-renders, all functions within its
template (including `$t()`) must be re-evaluated. Because the component
was being forced to re-render **60 times** per second, and there are
approximately **6 calls** to `$t()` within this UI, it multiplied into
60 × 6 = **360** intlify compilation and evaluate events per second.
## Solution
Only assemble objects and hand them over to Vue for rendering when the
mouse is actually dragging the canvas.
By extracting the math into **stack-allocated** primitive variables `(x,
y, w, h) `and strictly comparing them, it completely halts the CPU burn
at the source with minimal runtime overhead.
## Screenshot
before
<img width="1820" height="908" alt="image"
src="https://github.com/user-attachments/assets/b48d1e76-6498-47c0-af41-e0594d4e7e2f"
/>
after
<img width="1566" height="486" alt="image"
src="https://github.com/user-attachments/assets/5848b7b7-c57c-494f-a99e-4f7c92889ed0"
/>
Tooltips are normally resolved through the node definition. Since
DynamicCombo added widgets are nested in the spec definition, this
lookup fails to find them. This PR makes it so that when a widget is
dynamically added using `litegraphService:addNodeInput`, any 'tooltiptip
defined in the provided inputSpec is applied on the widget.
The tooltip system does not current support tooltips for dynamiclly
added inputs. That can be considered for a followup PR.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9717-Support-tooltips-on-DynamicCombos-31f6d73d365081dc93f9eadd98572b3c)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Alexander Brown <drjkl@comfy.org>
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
## Summary
Fix Vue warning about extraneous non-props attributes (`data-index`,
`style`) in `TreeExplorerV2Node`, which caused the Node Library sidebar
to freeze.
## Changes
- **What**: Added `defineOptions({ inheritAttrs: false })` and
`v-bind="$attrs"` on both node/folder `<div>` elements so the
virtualizer's positioning attributes are properly applied to the
rendered DOM.
## Review Focus
`TreeExplorerV2Node` has a fragment root (`<TreeItem as-child>` +
`<Teleport>`), so Vue cannot auto-inherit attrs. The virtualizer's
`style` (positioning) merges cleanly with `rowStyle` (`paddingLeft`
only).
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9735-fix-resolve-extraneous-attrs-warning-in-TreeExplorerV2Node-3206d73d3650817c8619e2145e98813d)
by [Unito](https://www.unito.io)
## Summary
The canvas mode selector popover (Select/Hand mode) uses plain `div`
elements for its menu items, making them completely inaccessible to
keyboard-only users and screen readers. This PR replaces them with
proper semantic HTML and ARIA attributes.
## Problem (AS-IS)
As reported in #9519, the canvas mode selector popover has the following
accessibility issues:
1. **Menu items are `div` elements** — they cannot receive keyboard
focus, so users cannot Tab into the popover or activate items with
Enter/Space. Keyboard-only users are completely locked out of switching
between Select and Hand (pan) mode via the popover.
2. **No ARIA roles** — screen readers announce the popover content as
generic text rather than an interactive menu. Users relying on assistive
technology have no way to understand that these are selectable options.
3. **No keyboard navigation within the popover** — even if a user
somehow focuses an item, there are no ArrowUp/ArrowDown handlers to move
between options, which is the standard interaction pattern for
`role="menu"` widgets (WAI-ARIA Menu Pattern).
4. **Decorative icons are not hidden from assistive technology** — icon
elements (`<i>` tags) are exposed to screen readers, adding noise to the
announcement.
5. **The bottom toolbar (`GraphCanvasMenu`) lacks semantic grouping** —
the `ButtonGroup` container has no `role="toolbar"` or `aria-label`, so
screen readers cannot convey that these buttons form a related control
group.
> Note: Pan mode itself already has keyboard shortcuts (`H` for
Hand/Lock, `V` for Select/Unlock), but the popover UI that surfaces
these options is not keyboard-accessible.
## Solution (TO-BE)
1. **Replace `div` → `button`** for menu items in `CanvasModeSelector` —
buttons are natively focusable and activatable via Enter/Space without
extra work.
2. **Add `role="menu"` on the container and `role="menuitem"` on each
option** — screen readers now announce "Canvas Mode menu" with two menu
items.
3. **Add ArrowUp/ArrowDown keyboard navigation** with wrap-around —
pressing ArrowDown on "Select" moves focus to "Hand", and vice versa.
This follows the WAI-ARIA Menu Pattern.
4. **Add `aria-label` to each menu item and `aria-hidden="true"` to
decorative icons** — screen readers announce "Select" / "Hand" clearly
without icon noise.
5. **Add `role="toolbar"` with `aria-label="Canvas Toolbar"` to the
`ButtonGroup`** — screen readers identify the bottom control bar as a
coherent toolbar.
## Changes
- **What**: Accessibility improvements to `CanvasModeSelector` popover
and `GraphCanvasMenu` toolbar
- **Dependencies**: None — only HTML/ARIA attribute changes and two new
i18n keys (`canvasMode`, `canvasToolbar`)
## Review Focus
- Verify the `button` elements render visually identical to the previous
`div` elements (same padding, hover styles, cursor)
- Confirm ArrowUp/ArrowDown navigation works correctly in the popover
- Check that screen readers announce the menu and toolbar correctly
Fixes#9519
> Note: The issue also requests Space-bar hold-to-pan, Tab through node
ports, and link creation mode keyboard shortcuts. These are significant
new features beyond the scope of this accessibility fix and should be
tracked separately.
## Test plan
- [x] Unit tests for ARIA roles, button elements, aria-labels,
aria-hidden, and arrow key navigation (7 tests)
- [ ] Manual: open canvas mode selector popover → Tab focuses first item
→ ArrowDown/ArrowUp navigates → Enter/Space selects
- [ ] Screen reader: verify "Canvas Mode menu" with "Select" and "Hand"
menu items are announced
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9526-fix-improve-canvas-menu-keyboard-navigation-and-ARIA-accessibility-31c6d73d3650814c9487ecf748cf6a99)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
Fixes Zod validation crash when the jobs batch contains text-only
preview outputs (e.g. from LLM nodes), which caused the entire Assets
sidebar to show nothing.
## Changes
- **What**: Made `filename`, `subfolder`, and `type` optional in
`zPreviewOutput` and added `.passthrough()` for extra fields like
`content`. Text-only jobs are safely filtered out downstream by
`supportsPreview`.
- Added tests in `fetchJobs.test.ts` verifying a mixed batch (image +
text-only + no-preview) parses successfully.
- Added test in `assetsStore.test.ts` verifying text-only jobs are
skipped without breaking sibling image jobs. Improved `TaskItemImpl`
mock to realistically handle media types.
## Review Focus
- The `zPreviewOutput` schema now uses `.passthrough()` to allow extra
fields from new preview types (like `content` on text previews). This is
consistent with `zRawJobListItem` and `zExecutionError` which also use
`.passthrough()`.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9724-fix-make-zPreviewOutput-accept-text-only-job-outputs-31f6d73d36508119a7aef99f9b765ecd)
by [Unito](https://www.unito.io)
## Summary
Fix cloud login page showing only the splash loader (wave SVG) instead
of the login form when the user is not authenticated.
## Changes
- **What**: Remove splash loader on `CloudLayoutView` mount. Cloud
onboarding pages do not use workspace initialization, so the
`workspaceStore.spinner` transition (`true→false`) that normally removes
the splash never occurs — leaving it visible indefinitely.
Co-authored-by: Amp <amp@ampcode.com>
<img width="842" height="488" alt="스크린샷 2026-03-07 오후 9 39 20"
src="https://github.com/user-attachments/assets/9ac8bfcd-c882-4661-851f-b08838d4fed1"
/>
## Summary
- Add Storybook stories for WidgetInputText, WidgetTextarea, and
ScrubableNumberInput
- Reorganize story titles under `Components/Input/` to align with Figma
design system
- Fix PrimeIcons not rendering in Storybook (caused by
`[&_*]:!font-inter` override)
- Fix knip unused export warnings (dead code removal + workspace config)
## Test plan
- [ ] Run `pnpm storybook` and verify Components/Input/InputText stories
render
- [ ] Verify Components/Input/TextArea stories render with label and
copy button
- [ ] Verify Components/Input/Number stories render with -/+ icons
- [ ] Toggle Storybook theme between Light/Dark and confirm Number story
adapts
- [ ] Verify existing Button stories still render correctly
🤖 Generated with [Claude Code](https://claude.com/claude-code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9527-feat-add-text-widget-stories-and-Number-input-stories-31c6d73d3650817ba351cdef26a356c8)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
Replace legacy PrimeVue `AutoCompletePlus` with a new
`SearchAutocomplete` component built on Reka UI, matching the
`SearchInput` design system.
## Changes
- **What**: Add `SearchAutocomplete` component extending `SearchInput`
with dropdown suggestions, IME composition handling, and generic typed
`optionLabel` support. Replace `AutoCompletePlus` usage in
`ManagerDialog`.
- **Dependencies**: None (uses existing Reka UI Combobox primitives)
## Review Focus
- `SearchAutocomplete` feature parity with the replaced
`AutoCompletePlus` (suggestions, option selection, IME handling)
- Dropdown styling and positioning via Reka UI `ComboboxContent`
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9645-feat-replace-PrimeVue-AutoComplete-with-SearchAutocomplete-in-ManagerDialog-31e6d73d36508117ba0bef3d30dd0863)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
## Summary
Add support for deploying full ephemeral preview environments from
frontend PRs. This is the frontend-side half — it sends `pr_number` and
`variant` (cpu/gpu) in the dispatch payload, and adds a cleanup dispatch
on PR close/unlabel.
### Changes
- **`cloud-dispatch-build.yaml`** — Add `pr_number` and `variant` to the
`frontend-asset-build` dispatch payload. Variant is derived from which
preview label triggered the event (`preview-cpu` → cpu, else gpu).
- **`cloud-dispatch-cleanup.yaml`** (new) — Fire-and-forget dispatch of
`frontend-preview-cleanup` to the cloud repo when a frontend PR is
closed or has its preview label removed. Enables synchronized teardown.
### Companion PR
Cloud-side: Comfy-Org/cloud (creates the `deploy-frontend-preview` job,
extends the reconciler)
### How it works
1. Label a frontend PR with `preview`, `preview-cpu`, or `preview-gpu`
2. Assets build and upload to GCS (existing flow)
3. Cloud deploys a full ephemeral env at `fe-pr-{N}.testenvs.comfy.org`
using all `:main` service tags
4. Subsequent pushes update the frontend SHA via AppSet upsert
5. On close/unlabel, cleanup dispatch triggers immediate teardown
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9715-feat-dispatch-frontend-PR-preview-environments-to-cloud-31f6d73d3650819da1b5ca5ce419e06e)
by [Unito](https://www.unito.io)
- Buttons are marked as `touch-manipulation` so double-tapping on them
doesn't initiate a zoom.
- Move scrubable inputs to usePointerSwipe
- Strangely, swipe direction was inverted on mobile. This solves the
issue and simplifies code
- Moves event handlers into the scrubbable input component
- Make the slightly bigger buttons only apply when on mobile.
- Updates the workflows dropdown to have a check by the activeWorkflow
and truncate workflow names
- Displays dropzones (for the image preview) on mobile, but disables the
prompt to drag and drop an image if none is selected.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9686-Mobile-input-tweaks-31f6d73d3650811d9025d0cd1ac58534)
by [Unito](https://www.unito.io)
## Summary
Fix HoneyToast footer content (text + buttons) being squished when
collapsed due to `sm:w-min` on the outer container.
## Changes
- **What**: Replace `sm:w-min sm:min-w-0` with state-aware sizing
(`sm:w-fit` when collapsed, `sm:w-[max(400px,40vw)]` when expanded) on
the HoneyToast container. Add `gap-4` between footer text and buttons in
ManagerProgressToast, `min-w-0` on text area, and `shrink-0` on button
area to prevent overlap.
## Review Focus
- HoneyToast is used in 3 places: ManagerProgressToast,
ModelImportProgressDialog, AssetExportProgressDialog. The change moves
width control from the inner content div to the outer container, which
should have no negative impact on the other two consumers.
## As-is
<img width="481" height="146" alt="스크린샷 2026-03-10 오전 10 40 52"
src="https://github.com/user-attachments/assets/b5e12e20-23ea-4f11-9778-ad4e6c10a425"
/>
## To-be
<img width="506" height="62" alt="스크린샷 2026-03-10 오후 1 46 18"
src="https://github.com/user-attachments/assets/f2b7963d-eedb-4885-bc57-f8b377962e92"
/>
<img width="630" height="83" alt="스크린샷 2026-03-10 오후 1 46 10"
src="https://github.com/user-attachments/assets/e9c35f1c-3441-4fb2-8fa4-f66b7d53b3e5"
/>
<img width="683" height="103" alt="스크린샷 2026-03-10 오후 1 46 02"
src="https://github.com/user-attachments/assets/afb94a16-cfba-4da9-8676-35f4d3133b57"
/>
## Summary
`loadApiJson` early-returns when missing node types are detected,
preventing the entire API-format workflow from loading onto the canvas.
## Changes
- **What**: Remove early `return` in `loadApiJson` so missing nodes are
skipped while the rest of the workflow loads normally, consistent with
how `loadGraphData` handles missing nodes in standard workflow format.
## Review Focus
The existing code already handles missing nodes gracefully:
- `LiteGraph.createNode()` returns `null` for unregistered types
- `if (!node) continue` skips missing nodes during graph construction
- `if (!fromNode) continue` skips connections to missing nodes
- `if (!node) return` skips input processing for missing nodes
The early `return` was unnecessarily preventing the entire load. The
warning modal is still shown via `showMissingNodesError`.
## Test workflow & screen recording
[04wan2.2smoothmix图生视频
(3).json](https://github.com/user-attachments/files/25858354/04wan2.2smoothmix.3.json)
[screen-capture.webm](https://github.com/user-attachments/assets/9c396f80-fff1-4d17-882c-35ada86542c1)
<img width="373" height="535" alt="스크린샷 2026-03-09 오후 2 48 10"
src="https://github.com/user-attachments/assets/7fea3fd4-0d90-4022-ad78-c53e3d5be887"
/>
## Summary
- Reorganize Select-related stories under `Components/Select/` hierarchy
(SingleSelect, MultiSelect, Select)
- Add `size` prop (`lg`/`md`) to SingleSelect, MultiSelect,
SelectTrigger for Figma Large (40px) / Medium (32px) variants
- Add `invalid` prop (red border) to SingleSelect and SelectTrigger
- Add `loading` prop (spinner) to SingleSelect
- Add `hover:bg-secondary-background-hover` to all select triggers
- Align disabled opacity to 30% per Figma spec
- Add new stories: Disabled, Invalid, Loading, MediumSize, AllStates
## Test plan
- [ ] Verify Storybook renders all stories under `Components/Select/`
- [ ] Check hover state visually on all select triggers
- [ ] Verify Medium size (32px) renders correctly
- [ ] Verify Invalid state shows red border
- [ ] Verify Loading state shows spinner
- [ ] Verify Disabled state has 30% opacity and no hover effect
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9639-refactor-reorganize-Select-stories-and-add-size-state-variants-31e6d73d36508142b835f04ab6bdaefe)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
When a logged-in user navigates to `/cloud/signup`, they are now
redirected to `cloud-user-check` (which handles survey or main page
routing).
This mirrors the existing `beforeEnter` guard on the `cloud-login`
route. The `switchAccount` query param bypass is preserved for
consistency.
## Changes
- Added `beforeEnter` guard to the `cloud-signup` route in
`onboardingCloudRoutes.ts`
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9691-Redirect-authenticated-users-from-signup-page-to-cloud-31f6d73d365081e08cb5c3360a862a37)
by [Unito](https://www.unito.io)
## Summary
Replace legacy `SearchBox` (PrimeVue) and `SearchBoxV2` with the unified
`SearchInput` (reka-ui) component across all consumers.
## Changes
- **What**: Remove `SearchBox.vue`, `SearchBoxV2.vue`, their tests and
stories. Migrate all 14 consumers to `SearchInput`. Move layout classes
to `ComboboxRoot` for proper flex sizing. Extract filter button/chips in
`NodeLibrarySidebarTab`. Standardize modal search width to `flex-1
max-w-lg`.
- **Dependencies**: None new — `SearchInput` already existed using
reka-ui
## Review Focus
- `NodeLibrarySidebarTab.vue`: filter button and `SearchFilterChip`
rendering moved outside the search component
- `SearchInput.vue`: `className` now applied to `ComboboxRoot` instead
of `ComboboxAnchor` for correct flex layout
- Modal dialogs (`WorkflowTemplateSelectorDialog`, `AssetBrowserModal`,
`SampleModelSelector`) unified to `flex-1 max-w-lg`
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9644-feat-unify-search-components-by-replacing-SearchBox-SearchBoxV2-with-SearchInput-31e6d73d365081ebac55cb265f33b631)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
Co-authored-by: github-actions <github-actions@github.com>
## Summary
Fix virtual scroll pagination not triggering in media asset panel list
view.
## Changes
**What**: `VirtualGrid` in `AssetsSidebarListView` was missing
`maxColumns=1` and had an incorrect default item height (200px vs actual
~48px). Without `maxColumns`, `cols` was calculated as
`floor(containerWidth / 200)` (e.g. 2), causing the row count to be
halved and `isNearEnd` to never fire correctly. Added `:max-columns="1"`
and `:default-item-height="48"` to fix pagination. Added regression
tests to `VirtualGrid.test.ts`.
## Review Focus
The root cause: `VirtualGrid.cols` computed as `floor(width/200)`
instead of `1` for single-column list layout, breaking spacer heights
and `approach-end` detection.
Test covers both the fix (approach-end fires with maxColumns=1) and the
bug reproduction (does not fire without it).
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9646-fix-virtual-scroll-pagination-not-working-in-media-asset-list-view-31e6d73d3650813d973ad19638ad6933)
by [Unito](https://www.unito.io)
Thumbnail downscaling is currently being used in more places than it
should be.
- Nodes which display images will display incorrect resolution
indicators
<img width="255" height="372" alt="image"
src="https://github.com/user-attachments/assets/674790b6-04c8-4db0-84c2-2fa2dbaf123d"
/> <img width="255" height="372" alt="image"
src="https://github.com/user-attachments/assets/1dbe751b-7462-4408-9236-9446b005f5fc"
/>
This is particularly confusing with output nodes, which claim the output
is not of the intended resolution
- The "Download Image" and "Open Image" context menu actions will
incorrectly download the downscaled thumbnail.
- The assets panel will incorrectly display the thumbnail resolution as
the resolution of the output
- The lightbox (zoom) of an image will incorrectly display a downscaled
thumbnail.
This PR is a quick workaround to staunch the major problems
- Nodes always display full previews.
- Resolution downscaling is applied on the assert card, not on the
assetItem itself
- Due to implementation, this means that asset cards will still
incorrectly show the resolution of the thumbnail instead of the size of
the full image.
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
## Summary
Cloud build dispatch was only triggering on the `labeled` event, not on
subsequent pushes to PRs that already had a preview label.
## Changes
- **What**: Add `synchronize` to `pull_request` event types and update
the `if` condition to support all three preview labels (`preview`,
`preview-cpu`, `preview-gpu`). For `labeled` events, check the added
label name; for `synchronize` events, check existing PR labels.
## Review Focus
The `if` condition now branches on `github.event.action` to use the
correct label-checking mechanism for each event type.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9636-fix-dispatch-cloud-build-on-synchronize-for-preview-labeled-PRs-31e6d73d3650814e9069e37d6199ffc9)
by [Unito](https://www.unito.io)
As a temporary fix for widgets being incorrectly hidden, #9669 allowed
all disabled widgets to be displayed.
This PR provides a more robust implementation to derive whether the
widget, as would be displayed from the root graph, is disabled.
Potential regression:
- Drag drop handlers are applied on node, not widgets. A subgraph
containing a "Load Image" node, does not allow dragging and dropping an
image onto the subgraphNode in order to load it. Because app mode
widgets would display from the original owning node prior to this PR,
these drag/drop handlers would apply. Placing "Load Image" nodes. I
believe this change makes behavior more consistent, but it warrants
consideration.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9671-Restore-hiding-of-linked-inputs-in-app-mode-31e6d73d365081688e37fbb931f3af68)
by [Unito](https://www.unito.io)
## Summary
Adds `isGraphReady` getter to `ComfyApp` and uses it in
`executionErrorStore` guards to prevent false 'ComfyApp graph accessed
before initialization' error logs during early store evaluation.
## Changes
- **What**: Added `isGraphReady` boolean getter to `ComfyApp` that
safely checks graph initialization without triggering the `rootGraph`
getter's error log. Updated 5 guard sites in `executionErrorStore` to
use `app.isGraphReady` instead of `app.rootGraph`.
- **Why**: The `rootGraph` getter logs an error when accessed before
initialization. Computed properties and watch callbacks in
`executionErrorStore` are evaluated early (before graph init), causing
false error noise in the console.
## Review Focus
- `isGraphReady` is intentionally minimal — just
`!!this.rootGraphInternal` — to avoid duplicating the error-logging
behavior of `rootGraph`
- The `watch(lastNodeErrors, ...)` callback now checks `isGraphReady` at
the top and early-returns, consistent with the computed property pattern
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9672-fix-add-isGraphReady-guard-to-prevent-premature-graph-access-error-logs-31e6d73d365081be8e1fc77114ce9382)
by [Unito](https://www.unito.io)
Co-authored-by: Alexander Brown <drjkl@comfy.org>
Adds a concise guide to `docs/release-process.md` explaining how the
release workflows interact, with focus on the version semantics that
differ between minor and patch bumps.
Key sections:
- How minor bumps freeze the previous minor into `core/` and `cloud/`
branches
- How patch bumps on `main` vs `core/X.Y` differ (published vs draft
releases)
- Why unreleased commits are dual-homed when a minor bump happens
- Summary table, backporting, publishing, and bi-weekly automation
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9548-docs-add-release-process-guide-31d6d73d365081f2bdaace48a7cb81ae)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Alexander Brown <drjkl@comfy.org>
App mode stores the state of selected widgets as a tuple of `[NodeId,
WidgetName]`. With recent subgraph changes, for a given node,
`widget.name` will no longer uniquely resolve to a single widget.
- From both Vue and Litegraph, selecting an input for display in App
mode will now resolve the NodeId of the node which owns the widget
instead of the selected node.
- When displaying selections in litegraph, if the NodeId does not exist
in the current graph, instead of resolving the actual node the rootGraph
is searched for any subgraphNode which contains a view matching the
`[NodeId, WidgetName]` pair.
- When displaying widgets in App mode, the widget is always set as being
a view of the real widget (This means that they will not display a
purple promotion border.
Known Issue:
- These same subgraph changes made it so that a widget can be linked
without being disabled. This PR makes it so widgets which have been
linked instead display normally under the assumption that they are
incorrectly marked as disabled. As disabled widgets can not be selected
as inputs, this should handle normal usage fine, but a better solution
is being investigated
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9669-Always-use-interior-nodeId-for-app-mode-31e6d73d365081f8a918d0e43cb659ee)
by [Unito](https://www.unito.io)
## Summary
When the actionbar is floating and has no docked buttons, the container
is now hidden (zero-width, transparent border) to avoid showing an empty
rounded box. Additionally, the error/destructive border is now applied
to the floating actionbar panel itself (via `ComfyActionbar`) instead of
the container, so it appears in the correct location when floating.
## Changes
- **TopMenuSection**: Added `hasDockedButtons` and
`isActionbarContainerEmpty` computed properties to detect when the
docked container has no visible buttons; `actionbarContainerClass`
computed hides the container by collapsing it when empty and floating,
while preserving the legacy drop zone via `:has(.border-dashed)` CSS
selector
- **TopMenuSection**: Error border
(`border-destructive-background-hover`) is now only applied to the
docked container when the actionbar is **not** floating
- **ComfyActionbar**: Accepts new `hasAnyError` prop and applies the
error border to the floating panel's `panelClass` when floating
## Review Focus
- The `has-[.border-dashed]` CSS selector restores the container visuals
when a legacy drag-target element is present inside it — verify this
works as expected
- Error border placement: docked mode shows border on container,
floating mode shows border on the fixed panel
## Screenshots
https://github.com/user-attachments/assets/75caabac-e391-4bfd-b4dc-62d564e55d37
## Problem
Output assets in the assets panel show content hashes (e.g.,
`a1b2c3d4.png`) instead of display names (e.g., `ComfyUI_00001_.png`).
## Root Cause
Cloud inference replaces `filename` with the content hash in the output
transform pipeline. The hashed filename gets stored in the jobs table's
`preview_output` JSONB. The frontend uses this hash as the display name.
## Solution
- Add `display_name` field to `AssetItem` schema and `ResultItemImpl`
- Backend (cloud PR) joins job→assets table to resolve the original name
and injects `display_name` into job responses
- Frontend prefers `display_name` over `name` **only for display text
and download filenames**
- `asset.name` remains unchanged (the hash) for URLs, drag-to-canvas,
export filters, and output key dedup
## Backwards Compatible
- OSS: `display_name` is undefined, falls back to `asset.name` (which is
already the real filename in OSS)
- Cloud pre-deploy: `display_name` absent from API, falls back
gracefully
- Old jobs with no assets: `display_name` not injected, no change
## Cloud PR
https://github.com/Comfy-Org/cloud/pull/2747https://github.com/user-attachments/assets/8a4c9cac-4ade-4ea2-9a70-9af240a56602
## Summary
After a user runs the workflow once in graph mode, switching to app mode
with no app built, incorrectly showed the app mode outputs view instead
of the intro screen
## Changes
- **What**: don't try and select outputs if no outputs & filter out all
outputs when nothing chosen
## Summary
Fix regression from PR #9535 where the multi-output count badge stopped
appearing in the asset sidebar.
## Root Cause
PR #9535 changed `outputCount` in `mapHistoryToAssets` from
`job.outputs_count` (backend-provided total) to
`task.previewableOutputs.length`. However, `TaskItemImpl` constructed
from a job listing only has the single `preview_output`, so
`previewableOutputs.length` is always **1** — the multi-output badge
never appears.
## Fix
Use the backend-provided `outputs_count` (via `task.outputsCount`) with
fallback to `task.previewableOutputs.length` when unavailable. This
restores the correct count while preserving the fallback for jobs that
don't have `outputs_count` from the server.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9627-fix-restore-backend-outputs_count-for-asset-sidebar-multi-output-badge-31d6d73d36508160b93fd03af4a01aa3)
by [Unito](https://www.unito.io)
## Summary
fix#8097.
This PR shifts the Mask Editor Brush Size slider from a linear scale to
a logarithmic (exponential) scale. Previously, the linear 1-250 range
heavily clumped the usable, small "fine-detail" brush sizes (e.g., 1px
to 20px) into the very first 10% of the slider, making it extremely
difficult to select precise sizes with the mouse.
This update borrows UX paradigms from other standard image editors like
Photoshop and GIMP, which map their scale entry widgets on an
exponential curve.
## GIMP Source
By inspecting the official **GIMP** source code under
`libgimpwidgets/gimpscaleentry.c`, we can see this exact mathematical
relationship being utilized when the logarithmic property is marked TRUE
on a brush radius adjustment widget:
```
// Mapping visual slider to internal value
value = gtk_adjustment_get_lower(...) + exp(t);
// Mapping internal value to visual slider
t = log (value - gtk_adjustment_get_lower(...) + 0.1);
```
https://github.com/user-attachments/assets/6d59ff12-f623-42cc-a52b-84147e9bb90b
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9534-fix-maskeditor-make-brush-size-slider-logarithmic-8097-31c6d73d365081118508e8363e0c5312)
by [Unito](https://www.unito.io)
## Summary
Fix asset sidebar badge showing inflated output count by using
previewable outputs length instead of raw server count.
## Changes
- **What**: Changed `outputCount` in `mapHistoryToAssets` from
`job.outputs_count` (includes all output types: text, JSON, custom data)
to `task.previewableOutputs.length` (only image, video, audio, 3D). The
badge now matches what users actually see in the expanded view.
## Review Focus
One-line change. The `task.previewableOutputs` array is already computed
on the line immediately below and used for `allOutputs`, so this
introduces no new computation.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9535-fix-use-previewable-output-count-for-asset-sidebar-badge-31c6d73d365081c49161caec64cf3921)
by [Unito](https://www.unito.io)
## Summary
Fix non-widget inputs on nested subgraphs appearing twice — once as
slots and once as unresolved button widgets.
## Changes
- **What**: Add `getTargetWidget()` guard in the `isSubgraphNode()`
branch of `resolveSubgraphInputTarget`, matching the existing check for
non-subgraph nodes. Non-widget inputs (e.g. AUDIO, IMAGE) now return
`undefined` instead of a bogus promotion entry.
## Review Focus
`resolveSubgraphInputTarget` had an asymmetry: the non-subgraph branch
checked `getTargetWidget()` before returning, but the `isSubgraphNode()`
branch returned unconditionally for every input. For nested subgraphs
where non-widget slots are linked through to inner SubgraphNode inputs,
this created `PromotedWidgetView` entries that failed `resolveDeepest()`
(falling back to `type: 'button'`), while the inputs also rendered as
normal slot circles since `input.widget` was never set by
`_resolveInputWidget` (which correctly skipped them).
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9542-fix-prevent-non-widget-inputs-on-nested-subgraphs-from-appearing-as-button-widgets-31c6d73d3650816387c3f97f0385e762)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
## Summary
Remove the workspace switching confirmation dialog since switching
workspaces no longer discards unsaved changes.
## Changes
- **What**: Remove `hasUnsavedChanges` check, `dialogService.confirm`
call, and unused imports (`useI18n`, `useWorkflowStore`,
`useDialogService`) from `useWorkspaceSwitch`. Rename
`switchWithConfirmation` to `switchWorkspace`. Update callers
(`WorkspaceSwitcherPopover.vue`, `InviteAcceptedToast.vue`). Remove
`workspace.unsavedChanges` i18n entries from all 12 locale files.
Simplify tests to cover core switching behavior only.
## Review Focus
The confirmation dialog was showing inaccurate information (warning
about discarding unsaved changes when that no longer happens). This is a
pure removal with no new behavior.
<!-- Pipeline-Ticket: COM-15441 -->
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9250-fix-remove-workspace-switching-confirmation-dialog-3136d73d365081d3b959da22e8f151d1)
by [Unito](https://www.unito.io)
## Summary
During workflow loading, the workflow data & active workflow object can
be out of sync, meaning any checkState calls will overwrite data into
the wrong workflow.
Recreation steps:
* Open 2-3 workflows
* Enter builder mode > select step
* Select some different inputs on each
* Quickly tap the shift key (this triggers checkState) while switching
tabs
* After a while, you'll see the wrong inputs on the workflows
Alternatively, register an extension that guarantees to call checkState
during the bad phase, run this in browser devtools and switch tabs:
```
window.app.registerExtension({
name: 'bad',
async afterConfigureGraph() {
window.app.extensionManager.workflow.activeWorkflow.changeTracker.checkState()
}
})
```
## Changes
- **What**:
- Add loading graph flag
- Prevent checkState calls while loading
- Prevent app mode data sync while loading
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9531-fix-Prevent-corruption-of-workflow-data-due-to-checkState-during-graph-loading-31c6d73d365081e2ab91d9145bf1d025)
by [Unito](https://www.unito.io)
## Summary
Standardize the repo's Node contract on 24 while centralizing workflow
resolution through `.nvmrc` so local setup, CI, and package metadata
stay aligned from one version file.
## Changes
- **What**: Add `package.json` `engines.node = 24.x`, switch every
`actions/setup-node` workflow in the repo to `node-version-file:
'.nvmrc'`, and update contributor and Playwright docs to point to
`.nvmrc` as the Node source of truth.
## Review Focus
The workflow behavior should be unchanged apart from sourcing the Node
version from `.nvmrc` instead of repeating literals like `20`, `22`,
`24.x`, or `lts/*`. GitHub's formatter also moved the new `engines`
block to the package metadata section near the end of `package.json`.
---------
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: Alexander Brown <drjkl@comfy.org>
## Summary
Dispatch a `frontend-asset-build` event to the cloud repo when the
`preview` label is added to a PR, so cloud can build preview assets.
## Changes
- **What**: Extended `cloud-dispatch-build.yaml` to trigger on
`pull_request` `labeled` events filtered to the `preview` label. The
payload sends the PR head SHA and branch.
## Review Focus
- The `pull_request` trigger gives a read-only `GITHUB_TOKEN`, but the
dispatch step uses `CLOUD_DISPATCH_TOKEN` so this is fine.
- Fork PRs are blocked by the existing `github.repository` guard.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9518-feat-dispatch-cloud-build-when-preview-label-is-added-to-PR-31c6d73d365081a8aab6f585960977f6)
by [Unito](https://www.unito.io)
## Summary
Additional fixes and updates based on testing
## Changes
- **What**:
- add warning to welcome screen & when sharing an app that has had all
outputs removed
- fix target workflow when changing mode via tab right click menu
- change build app text to be conditional "edit" vs "build" depending on
if an app is already defined
- update empty apps sidebar tab button text to make it clearer
- remove templates button from app mode (we will reintroduce this once
we have app templates)
- add "exit to graph" after applying default mode of node graph
- update cancel button to remove item from queue if it hasn't started
yet
- improve scoping of jobs/outputs to the current workflow [not perfect
but should be much improved]
- close sidebar tabs on entering app mode
- change tooltip to be under the workflow menu rather than covering the
button
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9511-feat-fix-App-mode-QA-feedback-2-31b6d73d365081d59bbbc13111100d46)
by [Unito](https://www.unito.io)
## Summary
Refactors the error and subgraph node footer UI by extracting a
dedicated `NodeFooter` component and replacing the CSS `outline`
approach with a layered border overlay for selection/executing state
indicators.
## Changes
- **What**: Extracted `NodeFooter.vue` from `LGraphNode.vue` to
encapsulate the footer tab logic (subgraph enter, error, advanced
inputs). Replaced CSS `outline` with an absolutely-positioned border
overlay div for selection and executing state. Added a separate root
border overlay div for the node body border. Removed unused
`isTransparent` function from `colorUtil.ts`.
- **Dependencies**: None
## Review Focus
- The layered overlay approach (`absolute -inset-[3px] border-3`) for
selection/executing outlines vs the previous `outline-3` approach —
ensures the outline renders outside the node bounds correctly including
the footer area
- `NodeFooter` handles 4 cases: subgraph+error (dual tabs), error only,
subgraph only, advanced inputs — verify edge cases render correctly
- Resize handle bottom offset adjustments for nodes with footers
(`hasFooter`)
## Screenshots
<img width="1142" height="603" alt="image"
src="https://github.com/user-attachments/assets/e0d401f0-8516-4f5f-ab77-48a79530f4bd"
/>
<img width="1175" height="577" alt="image"
src="https://github.com/user-attachments/assets/bcf08fff-728a-491c-add9-5b96d2f3bfce"
/>
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9360-style-Update-error-subgraph-node-footer-design-with-layered-overlay-approach-3186d73d365081b2ac31f166f4d1944a)
by [Unito](https://www.unito.io)
---------
Co-authored-by: github-actions <github-actions@github.com>
Co-authored-by: GitHub Action <action@github.com>
Disable missing model warnings in browser tests by default.
Browser tests run without model files on disk, so workflows that embed
model metadata can render differently in CI than the test actually
intends to cover. The viewport screenshot golden had started depending
on the missing-model popup even though the test is only about restoring
an offscreen viewport.
Set `Comfy.Workflow.ShowMissingModelsWarning` to `false` in the shared
Playwright fixture, keep the missing-model dialog coverage by explicitly
enabling the setting in the dialog tests, and update the viewport
screenshot expectation to the no-popup rendering.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9513-test-disable-missing-model-warnings-in-browser-tests-31b6d73d365081d1908bfe11ec0c3bc2)
by [Unito](https://www.unito.io)
## Summary
Extract the inner link-resolution loop from
`_internalConfigureAfterSlots` into a private `_resolveInputWidget`
method to reduce cognitive complexity below the sonarjs threshold of 15.
## Changes
- **What**: Extract nested loop body (lines 654-689) into
`_resolveInputWidget` private method in `SubgraphNode.ts`
- Pure refactoring with no behavioral changes
## Review Focus
Straightforward extract-method refactoring. The new method contains the
exact same logic that was previously inline.
Fixes#9297
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9383-refactor-extract-input-widget-resolution-from-SubgraphNode-configure-3196d73d365081ba9124cfd0d312fcb0)
by [Unito](https://www.unito.io)
## Summary
Extract and harden subgraph node ID deduplication to prevent widget
store key collisions when multiple subgraph copies share identical node
IDs.
## Changes
- **What**: Extract `deduplicateSubgraphNodeIds` from `LGraph.ts` into
`utils/subgraphDeduplication.ts`, decomposed into focused helpers
(`remapNodeIds`, `findNextAvailableId`, `patchSerialisedLinks`,
`patchPromotedWidgets`, `patchProxyWidgets`). Clone inputs internally so
caller data is never mutated. Add safety limit on ID search to prevent
unbounded loops. Add `console.warn` on remapped IDs matching existing
`ensureGlobalIdUniqueness` behavior. Add test fixture and 5 behavioral
tests covering ID remapping, link patching, promoted widget patching,
proxyWidget patching, and no-op when IDs are unique.
## Review Focus
- The cloning strategy in `deduplicateSubgraphNodeIds` — it
`structuredClone`s subgraphs and rootNodes, returning the clones. The
caller uses `effectiveNodesData` to thread the patched root nodes
through to node creation.
- The `MAX_NODE_ID` safety limit (100M) — is this a reasonable ceiling?
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9510-fix-extract-and-harden-subgraph-node-ID-deduplication-31b6d73d365081f48c7de75e2bfc48b3)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Amp <amp@ampcode.com>
- Increased the z-index on app mode outputs so that they display above a
zoomed image
- The "view job" button on the job queued toast in mobile app mode will
take you to outputs instead of assets
- Image previews now have a minimum zoom of ~20% and a maximum zoom of
~50x
- The enter panel in linear mode now has a minimum size of ~1/5th screen
size
- In arrange mode, dragging to rearrange inputs will no longer cause a
horizontal scrollbar to appear.
- Videos will now display the first frame instead of a generic video
icon
- Muted/Bypassed nodes can no longer be selected as inputs/outputs, or
be displayed when in app mode.
- Linked input can no longer be selected or displayed
- Adds a share workflow button in app mode and wires up the existing
context menu
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9432-More-app-fixes-31a6d73d365081509cd0ea74bfdc9b95)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
Fixes flakey test to only assert that the upload request is made with
the correct URL
## Changes
- **What**
- Replace waitForResponse with waitForRequest for the no_workflow.webp
upload test to only assert the request is initiated with the correct URL
- Move request listener setup before the drag-drop action to avoid race
conditions
- Remove screenshot assertion for the upload case since the upload may
not complete before the screenshot is taken
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9488-fix-update-loadWorkflowInMedia-test-to-only-assert-upload-request-URL-31b6d73d365081f69a9aeb1095da7d60)
by [Unito](https://www.unito.io)
## Summary
Migrate hardcoded litegraph canvas keybindings (Ctrl+A/C/V, Delete,
Backspace) into the customizable keybinding system so users can remap
them via Settings > Keybindings.
## Changes
- **What**: Register Ctrl+A (SelectAll), Ctrl+C (CopySelected), Ctrl+V
(PasteFromClipboard), Ctrl+Shift+V (PasteFromClipboardWithConnect),
Delete/Backspace (DeleteSelectedItems) as core keybindings in
`defaults.ts`. Add new `PasteFromClipboardWithConnect` command. Remove
hardcoded handling from litegraph `processKey()`, the `app.ts` Ctrl+C/V
monkey-patch, and the `keybindingService` canvas forwarding logic.
Fixes#1082Fixes#2015
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-9459-feat-expose-litegraph-internal-keybindings-31b6d73d3650819a8499fd96c8a6678f)
by [Unito](https://www.unito.io)
2026-03-06 18:30:35 +01:00
831 changed files with 56884 additions and 8386 deletions
description: Manages cherry-pick backports across stable release branches. Discovers candidates from Slack/git, analyzes dependencies, resolves conflicts via worktree, and logs results. Use when asked to backport, cherry-pick to stable, manage release branches, do stable branch maintenance, or run a backport session.
---
# Backport Management
Cherry-pick backport management for Comfy-Org/ComfyUI_frontend stable release branches.
## Quick Start
1.**Discover** — Collect candidates from Slack bot + git log gap (`reference/discovery.md`)
| `core/*` | Local/self-hosted ComfyUI only | Core editor, local workflows, node system |
**⚠️ NEVER backport cloud-only PRs to `core/*` branches.** Cloud-only changes (app mode, cloud auth, cloud billing UI, cloud-specific API calls) are irrelevant to local users and waste effort. Before backporting any PR to a `core/*` branch, check:
- Does the PR title/description mention "app mode", "cloud", or cloud-specific features?
- Does the PR only touch files like `appModeStore.ts`, cloud auth, or cloud-specific components?
- If yes → skip for `core/*` branches (may still apply to `cloud/*` branches)
The `pr-backport.yaml` action reports more conflicts than reality. `git cherry-pick -m 1` with git auto-merge handles many cases the automation can't. Always attempt manual cherry-pick before skipping.
### Never Skip Based on Conflict File Count
12 or 27 conflicting files can be trivial (snapshots, new files). **Categorize conflicts first**, then decide. See Conflict Triage below.
## Conflict Triage
**Always categorize before deciding to skip. High conflict count ≠ hard conflicts.**
| **MUST** | User-facing bug, crash, data corruption, security. Clear breakage that users will hit. | Backport (with deps if needed) |
| **SHOULD** | UX improvement, minor bug, small dep chain. No user-visible breakage if skipped, but improves experience. | Backport if clean cherry-pick; defer if conflict resolution is non-trivial |
| **SKIP** | CI/tooling, test-only, lint rules, cosmetic, dep refresh | Skip with documented reason |
| **NEEDS DISCUSSION** | Large dep chain, unclear risk/benefit, touches core types | Flag for human |
### MUST vs SHOULD Decision Guide
When unsure, ask: "If a user on this stable branch reports this issue, would we consider it a bug?"
- **Yes** → MUST. The fix addresses broken behavior.
- **No, but it's noticeably better** → SHOULD. The fix is a quality-of-life improvement.
- **No, and it's cosmetic or internal** → SKIP.
For SHOULD items with conflicts: if conflict resolution requires more than trivial accept-theirs patterns (content conflicts in business logic, not just imports), downgrade to SKIP or escalate to NEEDS DISCUSSION.
| `cloud/*` | Local-only features not present on cloud branch |
Cloud-only PRs backported to `core/*` are wasted effort — `core/*` branches serve local/self-hosted users who never see cloud features. Check PR titles, descriptions, and files changed for cloud-specific indicators.
## Features Not on Stable Branches
Check before backporting — these don't exist on older branches:
- **Painter** (`src/extensions/core/painter.ts`) — not on core/1.40
- **GLSLShader** — not on core/1.40
- **App builder** — check per branch
- **appModeStore.ts** — not on core/1.40
## Dep Refresh PRs
Always SKIP on stable branches. Risk of transitive dependency regressions outweighs audit cleanup benefit. If a specific CVE fix is needed, cherry-pick that individual fix instead.
## Revert Pairs
If PR A is reverted by PR B:
- Skip BOTH A and B
- If a fixed version exists (PR C), backport only C
**⚠️ Human review for conflict resolutions:** When admin-merging a PR where you manually resolved conflicts (especially content conflicts beyond trivial accept-theirs), pause and present the resolution diff to the human for review before merging. Trivial resolutions (binary snapshots, modify/delete, locale key additions) can proceed without review.
## Step 4: Wave Verification
After completing all PRs in a wave for a target branch:
When merging multiple PRs to the same branch, later PRs may conflict with earlier merges:
```bash
git fetch origin TARGET_BRANCH
git rebase origin/TARGET_BRANCH
# Resolve new conflicts
git push --force origin backport-$PR-to-TARGET
sleep 20# Wait for GitHub to recompute merge state
gh pr merge $PR --squash --admin
```
## Lessons Learned
1.**Automation reports more conflicts than reality** — `cherry-pick -m 1` with git auto-merge handles many "conflicts" the automation can't
2.**Never skip based on conflict file count** — 12 or 27 conflicts can be trivial (snapshots, new files). Categorize first: binary PNGs, modify/delete, content, add/add.
3.**Modify/delete goes BOTH ways** — if the PR introduces new files (not on target), `git add` them. If target deleted files the PR modifies, `git rm`.
4.**Binary snapshot PNGs** — always `git checkout --theirs && git add`. Never skip a PR just because it has many snapshot conflicts.
5.**Batch label additions need 2s delay** between API calls to avoid rate limits
6.**Merging 6+ PRs rapidly** can cause later PRs to become unmergeable — wait 20-30s for GitHub to recompute merge state
7.**appModeStore.ts, painter files, GLSLShader files** don't exist on core/1.40 — `git rm` these
8.**Always validate JSON** after resolving locale file conflicts
9.**Dep refresh PRs** — skip on stable branches. Risk of transitive dep regressions outweighs audit cleanup. Cherry-pick individual CVE fixes instead.
10.**Verify after each wave** — run `pnpm typecheck` on the target branch after merging a batch. Catching breakage early prevents compounding errors.
11.**Cloud-only PRs don't belong on core/\* branches** — app mode, cloud auth, and cloud-specific UI changes are irrelevant to local users. Always check PR scope against branch scope before backporting.
- Were there clusters of related PRs that should have been backported together?
- Any PRs that should have been backported sooner (continuous backporting candidates)?
- Feature branches that need tracking for future sessions?
```
## Final Deliverable: Visual Summary
At session end, generate a **mermaid diagram** showing all backported PRs organized by target branch and category (MUST/SHOULD), plus a summary table. Present this to the user as the final output.
```mermaid
graph TD
subgraph branch1["☁️ cloud/X.XX — N PRs"]
C1["#XXXX title"]
C2["#XXXX title"]
end
subgraph branch2must["🔴 core/X.XX MUST — N PRs"]
M1["#XXXX title"]
end
subgraph branch2should["🟡 core/X.XX SHOULD — N PRs"]
description: 'Ships performance fixes with CI-proven improvement using stacked PRs. PR1 adds a @perf test (establishes baseline on main), PR2 adds the fix (CI shows delta). Use when implementing a perf optimization and wanting to prove it in CI.'
---
# Performance Fix with Proof
Ships perf fixes as two stacked PRs so CI automatically proves the improvement.
## Why Two PRs
The `ci-perf-report.yaml` workflow compares PR metrics against the **base branch baseline**. If you add a new `@perf` test in the same PR as the fix, that test doesn't exist on main yet — no baseline, no delta, no proof. Stacking solves this:
1.**PR1 (test-only)** — adds the `@perf` test that exercises the bottleneck. Merges to main. CI runs it on main → baseline established.
2.**PR2 (fix)** — adds the optimization. CI runs the same test → compares against PR1's baseline → delta shows improvement.
If the bottleneck needs a specific workflow (e.g., 50+ nodes, many DOM widgets), add it to `browser_tests/assets/`. Keep it minimal — only the structure needed to trigger the bottleneck.
### Step 4: Verify locally
```bash
pnpm exec playwright test --project=performance --grep "<test name>"
```
Confirm the test runs and produces reasonable metric values.
### Step 5: Create PR1 (test-only)
```bash
pnpm typecheck:browser
pnpm lint
git add browser_tests/
git commit -m "test: add perf test for <bottleneck description>"
git push -u origin perf/test-<name>
gh pr create --title "test: add perf test for <bottleneck>"\
--body "Adds a @perf test to establish a baseline for <bottleneck>.
This is PR 1 of 2. The fix will follow in a separate PR once this baseline is established on main.
## What
Adds \`<test-name>\` to the performance test suite measuring <metric> during <action>.
## Why
Needed to prove the improvement from the upcoming fix for backlog item #<N>."\
--base main
```
### Step 6: Get PR1 merged
Once PR1 merges, CI runs the test on main → baseline artifact saved.
If Δ is negative for the target metric, the fix is proven.
## Test Design Guidelines
1.**Stress the specific bottleneck** — don't measure everything, isolate the hot path
2.**Use enough iterations** — the test should run long enough that the metric difference is clear (100+ frames for idle tests, 50+ interactions for event tests)
3.**Keep it deterministic** — avoid timing-dependent assertions; measure counts not durations when possible
4.**Match the backlog entry** — reference the backlog item number in the test name or PR description
## Examples
**Testing DOM widget reactive mutations (backlog #8):**
description: 'Bug fix workflow that proves test validity with a red-then-green CI sequence. Commits a failing test first (CI red), then the minimal fix (CI green). Use when fixing a bug, writing a regression test, or when asked to prove a fix works.'
---
# Red-Green Fix
Fixes bugs as two commits so CI automatically proves the test catches the bug.
## Why Two Commits
If you commit the test and fix together, the test always passes — reviewers cannot tell whether the test actually detects the bug or is a no-op. Splitting into two commits creates a verifiable CI trail:
1.**Commit 1 (test-only)** — adds a test that exercises the bug. CI runs it → test fails → red X.
2.**Commit 2 (fix)** — adds the minimal fix. CI runs the same test → test passes → green check.
The red-then-green sequence in the commit history proves the test is valid.
## Input
The user provides a bug description as an argument. If no description is given, ask the user to describe the bug before proceeding.
Bug description: $ARGUMENTS
## Step 0 — Setup
Create an isolated branch from main:
```bash
git fetch origin main
git checkout -b fix/<bug-name> origin/main
```
## Step 1 — Red: Failing Test Only
Write a test that reproduces the bug. **Do NOT write any fix code.**
| `test: ...` | :red_circle: Red | Proves the test catches the bug |
| `fix: ...` | :green_circle: Green | Proves the fix resolves the bug |
## Test Plan
- [ ] CI red on test-only commit
- [ ] CI green on fix commit
- [ ] Added/updated E2E regression under `browser_tests/` or explained why not applicable
- [ ] Manual verification (if applicable)
EOF
)"
```
## Gotchas
### CI fails on test commit for unrelated reasons
Lint, typecheck, or other tests may fail — not just your new test. Check the CI logs carefully. If the failure is unrelated, fix it in a separate commit before the `test:` commit so the red X is clearly attributable to your test.
### Test passes when it should fail
The bug may only manifest under specific conditions (e.g., Windows paths, external model directories, certain workflow structures). Make sure your test setup matches the actual bug scenario. Check that you're not accidentally testing the happy path.
### Flaky Playwright tests
If your e2e test is intermittent, it doesn't prove anything. Use retrying assertions (`toBeVisible`, `toHaveText`) instead of `waitForTimeout`. See the `/writing-playwright-tests` skill for anti-patterns.
### Pre-existing CI failures on main
If main itself is red, branch from the last green commit or fix the pre-existing failure first. A red-green proof is meaningless if the baseline is already red.
it('returns same user on repeated calls',async()=>{
constservice=newUserService()
constuser1=awaitservice.getUser(1)
constuser2=awaitservice.getUser(1)
expect(user1).toBe(user2)// Behavior, not implementation
})
```
Why this matters for red-green: if your test is coupled to internals, a valid fix that changes the implementation may break the test — even though the bug is fixed. The green commit should only require changing source code, not rewriting the test.
## Assertion-Free Tests
Every test must assert something meaningful. A test without assertions always passes — it cannot produce the red X needed in Step 1.
Mock only system boundaries (network, filesystem, Electron APIs). If you mock the module under test, you are testing your mocks — the test will not detect the real bug.
**Bad** — mocking everything:
```typescript
vi.mock('./pathResolver')
vi.mock('./validator')
vi.mock('./downloader')
it('downloads model',()=>{
// This only tests that mocks were called, not that the bug exists
| `waitForTimeout` instead of assertions | Use `toBeVisible()`, `toHaveText()` |
| Shared state between tests | Isolate with `afterEach` / `beforeEach` |
| Timing-dependent logic | Use `expect.poll()` or `toPass()` |
## Gaming the Red-Green Process
These are ways the red-green proof gets invalidated during Step 2 (the fix commit). The test from Step 1 is immutable — if any of these happen, restart from Step 1.
// Step 2 (green) — weakened to pass without a real fix
expect(result).toBeDefined()// This proves nothing
```
**Updating snapshots to bless the bug:**
```bash
# Instead of fixing the code, just updating the snapshot to match buggy output
pnpm test:unit -- --update
```
If a snapshot needs updating, the fix should change the code behavior, not the expected output.
**Adding mocks in Step 2 that hide the failure:**
```typescript
// Step 2 adds a mock that didn't exist in Step 1
vi.mock('./pathResolver',()=>({
resolve:()=>'/expected/path'// Hardcoded to pass
}))
```
Step 2 should only change source code — not test infrastructure.
## Testing the Happy Path Only
The red-green pattern specifically requires the test to exercise the **broken path**. If you only test the case that already works, the test will pass (green) on Step 1 — defeating the purpose.
description: 'Parse ticket URL (Notion or GitHub), extract all data, initialize pipeline run. Use when starting work on a new ticket or when asked to pick up a ticket.'
---
# Ticket Intake
Parses a ticket URL from supported sources (Notion or GitHub), extracts all relevant information, and creates a ticket in the pipeline API.
> **🚨 CRITICAL REQUIREMENT**: This skill MUST register the ticket in the Pipeline API and update the source (Notion/GitHub). If these steps are skipped, the entire pipeline breaks. See [Mandatory API Calls](#mandatory-api-calls-execute-all-three) below.
**⚠️ These three API calls are the ENTIRE POINT of this skill. Without them, the ticket is invisible to the pipeline, downstream skills will fail, and Notion status won't update.**
**You MUST make these HTTP requests.** Use `curl` from bash — do not just read this as documentation.
**If `state` is not `RESEARCH`, go back to Step 4 and complete the missing calls.**
### Step 6: Output Summary and Handoff
Print a clear summary:
```markdown
## Ticket Intake Complete
**Source:** Notion | GitHub
**Title:** [Ticket title]
**ID:** abc12345
**Status:** In Progress (queued)
**Priority:** High
**Area:** UI
### Description
[Brief description or first 200 chars]
### Acceptance Criteria
- [ ] Criterion 1
- [ ] Criterion 2
### Links
- **Ticket:** [Original URL]
- **Slack:** [Slack thread content fetched via slackdump] (Notion only)
### Pipeline
- **API Ticket ID:** abc12345
- **State:** RESEARCH
- **Verified:** ✅ (via verify-intake.sh or public API)
```
**After printing the summary, immediately handoff** to continue the pipeline. Use the `handoff` tool with all necessary context (ticket ID, source, title, description, slack context if any):
> **Handoff goal:** "Continue pipeline for ticket {ID} ({title}). Ticket is in RESEARCH state. Load skill: `research-orchestrator` to begin research phase. Ticket data: source={source}, notion_page_id={pageId}, priority={priority}. {slack context summary if available}"
**Do NOT wait for human approval to proceed.** The intake phase is complete — handoff immediately.
## Error Handling
### Unsupported URL
```
❌ Unsupported ticket URL format.
Supported formats:
- Notion: https://notion.so/... or https://www.notion.so/...
description: 'Write or update Storybook stories for Vue components in ComfyUI_frontend. Use when adding, modifying, reviewing, or debugging `.stories.ts` files, Storybook docs, component demos, or visual catalog entries in `src/` or `apps/desktop-ui/`.'
---
# Write Storybook Stories for ComfyUI_frontend
## Workflow
1. !!!!IMPORTANT Confirm the worktree is on a `feat/*` or `fix/*` branch. Base PRs on the local `main`, not a fork branch.
2. Read the component source first. Understand props, emits, slots, exposed methods, and any supporting types or composables.
4. If a Figma link is provided, list the states you need to cover before writing stories.
5. Co-locate the story file with the component: `ComponentName.stories.ts`.
6. Add each variation on separate stories, except hover state. this should be automatically applied by the implementation and not require a separate story.
7. Run Storybook and validation checks before handing off.
## Match Local Conventions
- Copy the closest neighboring story instead of forcing one universal template.
- Most repo stories use `@storybook/vue3-vite`. Some stories under `apps/desktop-ui` still use `@storybook/vue3`; keep the local convention for that area.
- Add `tags: ['autodocs']` unless the surrounding stories in that area intentionally omit it.
- Use `ComponentPropsAndSlots<typeof Component>` when it helps with prop and slot typing.
- Keep `render` functions stateful when needed. Use `ref()`, `computed()`, and `toRefs(args)` instead of mutating Storybook args directly.
- Use `args.default` or other slot-shaped args when the component content is provided through slots.
- Use `ComponentExposed` only when a component's exposed API breaks the normal Storybook typing.
- Add decorators for realistic width or background context when the component needs it.
## Title Patterns
Do not invent titles from scratch when a close sibling story already exists. Match the nearest domain pattern.
- Prefer explicit prop-driven stories such as `Disabled`, `Loading`, `Invalid`, `WithPlaceholder`, `AllSizes`, or `EdgeCases`.
- Add an aggregate story such as `AllVariants`, `AllSizes`, or `EdgeCases` when side-by-side comparison is useful.
- Use pseudo-state parameters only if the addon is already configured in this repo.
- If a Figma state cannot be represented exactly, capture the closest prop-driven version and explain the gap in the story docs.
## Component-Specific Notes
- Widget components often need a minimal `SimplifiedWidget` object. Build it in `setup()` and use `computed()` when `args` change `widget.options`.
- Input and search components often need a width-constrained wrapper so they render at realistic sizes.
- Asset and platform cards often need background decorators such as `bg-base-background` and fixed-width containers.
- Desktop installer stories may need custom `backgrounds` parameters and may intentionally keep the older Storybook import style used by neighboring files.
- Use semantic tokens such as `bg-base-background` and `bg-node-component-surface` instead of `dark:` variants or hardcoded theme assumptions.
## Checklist
- [ ] Read the component source and any supporting types or composables
- [ ] Match the nearest local title pattern and story style
- [ ] Include a baseline story; name it `Default` only when that matches nearby conventions
- [ ] Add focused stories for meaningful states
- [ ] Add `tags: ['autodocs']`
- [ ] Keep the story co-located with the component
- [ ] Run `pnpm storybook`
- [ ] Run `pnpm typecheck`
- [ ] Run `pnpm lint`
## Avoid
- Do not guess props, emits, slots, or exposed methods.
- Do not force one generic title convention across the repo.
- Do not mutate Storybook args directly for `v-model` components.
- Do not introduce `dark:` Tailwind variants in story wrappers.
- Do not create barrel files.
- Do not assume every story needs `layout: 'centered'` or a `Default` export; follow the nearest existing pattern.
Use only PR metadata already available in the review context: the PR title, commit subjects in this PR, the files changed in this PR relative to the PR base (equivalent to `base...head`), and the PR description.
Do not rely on shell commands. Do not inspect reverse diffs, files changed only on the base branch, or files outside this PR. If the changed-file list or commit subjects are unavailable, mark the check inconclusive instead of guessing.
Pass if at least one of the following is true:
1. Neither the PR title nor any commit subject in the PR uses bug-fix language such as `fix`, `fixed`, `fixes`, `fixing`, `bugfix`, or `hotfix`.
2. The PR changes at least one file under `browser_tests/`.
3. The PR description includes a concrete, non-placeholder explanation of why an end-to-end regression test was not added.
Fail otherwise. When failing, mention which bug-fix signal you found and ask the author to either add or update a Playwright regression test under `browser_tests/` or add a concrete explanation in the PR description of why an end-to-end regression test is not practical.
echo "::error::Linting/formatting issues found. Since this PR is from a fork, auto-fix cannot be applied automatically."
echo ""
echo "Please run these commands locally and push the changes:"
echo " pnpm lint:fix"
echo " pnpm stylelint:fix"
echo " pnpm format"
echo ""
echo "Or set up pre-commit hooks to automatically format on every commit:"
echo " pnpm prepare"
echo ""
echo "See CONTRIBUTING.md for more details."
exit 1
- name:Final validation
run:|
pnpm lint
@@ -84,16 +100,3 @@ jobs:
repo: context.repo.repo,
body: '## 🔧 Auto-fixes Applied\n\nThis PR has been automatically updated to fix linting and formatting issues.\n\n**⚠️ Important**: Your local branch is now behind. Run `git pull` before making additional changes to avoid conflicts.\n\n### Changes made:\n- ESLint auto-fixes\n- Oxfmt formatting'
body: '## ⚠️ Linting/Formatting Issues Found\n\nThis PR has linting or formatting issues that need to be fixed.\n\n**Since this PR is from a fork, auto-fix cannot be applied automatically.**\n\n### Option 1: Set up pre-commit hooks (recommended)\nRun this once to automatically format code on every commit:\n```bash\npnpm prepare\n```\n\n### Option 2: Fix manually\nRun these commands and push the changes:\n```bash\npnpm lint:fix\npnpm format\n```\n\nSee [CONTRIBUTING.md](https://github.com/Comfy-Org/ComfyUI_frontend/blob/main/CONTRIBUTING.md#git-pre-commit-hooks) for more details.'
if [ "${{ needs.resolve-version.outputs.needs_release }}" = "true" ]; then
RELEASE_NOTE="⚠️ **Release PR must be merged first** - check [release workflow runs](https://github.com/Comfy-Org/ComfyUI_frontend/actions/workflows/release-version-bump.yaml)"
BODY=$''"${RELEASE_NOTE}"$'\n\n'"${BODY}"
fi
PYPI_NOTE="✅ **PyPI package confirmed available** — \`comfyui-frontend-package==${{ needs.resolve-version.outputs.target_version }}\` has been published and verified."
BODY=$''"${PYPI_NOTE}"$'\n\n'"${BODY}"
# Save to file for later use
printf '%s\n' "$BODY" > pr-body.txt
@@ -307,7 +427,11 @@ jobs:
fi
if [ -n "$EXISTING_PR" ] && [ "$EXISTING_PR" != "null" ]; then
echo "PR already exists (#${EXISTING_PR}), updating branch will update the PR"
@@ -17,7 +17,7 @@ Have another idea? Drop into Discord or open an issue, and let's chat!
### Prerequisites & Technology Stack
- **Required Software**:
- Node.js (v24) and pnpm
- Node.js (see `.nvmrc` for the required version) and pnpm
- Git for version control
- A running ComfyUI backend instance (otherwise, you can use `pnpm dev:cloud`)
@@ -87,6 +87,10 @@ navigate to `http://<server_ip>:5173` (e.g. `http://192.168.2.20:5173` here), to
> ⚠️ IMPORTANT:
> The dev server will NOT load JavaScript extensions from custom nodes. Only core extensions (built into the frontend) will be loaded. This is because the shim system that allows custom node JavaScript to import frontend modules only works in production builds. Python custom nodes still function normally. See [Extension Development Guide](docs/extensions/development.md) for details and workarounds. And See [Extension Overview](docs/extensions/README.md) for extensions overview.
## Troubleshooting
If you run into issues during development (e.g. `pnpm dev` hanging, TypeScript errors after pulling, lock file conflicts), see [TROUBLESHOOTING.md](TROUBLESHOOTING.md) for common fixes.
#### Q: `eslint-disable` comment isn't suppressing an oxlint rule
**Symptoms:**
- `// eslint-disable-next-line rule-name` has no effect
- Lint error persists despite the disable comment
**Solution:**
oxlint has its own disable syntax. Use `oxlint-disable` instead:
```ts
// oxlint-disable-next-line no-console
console.log('debug')
```
Check whether the rule is enforced by oxlint (in `.oxlintrc.json`) or ESLint (in `eslint.config.ts`) to pick the right disable comment.
---
#### Q: New lint errors after pulling/upgrading oxlint
**Symptoms:**
- Lint errors in files you didn't change
- Rules you haven't seen before (e.g. `no-immediate-mutation`, `prefer-optional-chain`)
**Solutions:**
1. **Run the auto-fixer first:**
```bash
pnpm lint:fix
```
2. **Review changes carefully** — some oxlint auto-fixes can produce incorrect code. Check the diff before committing.
3. **If a rule seems wrong**, check `.oxlintrc.json` to see if it should be disabled or configured differently.
**Why this happens:** oxlint version bumps often enable new rules by default.
---
#### Q: oxlint fails with TypeScript errors
**Symptoms:**
- `pnpm oxlint` or `pnpm lint` fails with type-related errors
- Errors mention type resolution or missing type information
**Solution:**
oxlint runs with `--type-aware` in this project, which requires valid TypeScript compilation. Fix the TS errors first:
```bash
pnpm typecheck # Identify TS errors
pnpm build # Or do a full build
pnpm lint # Then re-run lint
```
---
#### Q: Duplicate lint errors from both oxlint and ESLint
**Symptoms:**
- Same violation reported twice
- Conflicting auto-fix suggestions
**Solution:**
The project uses `eslint-plugin-oxlint` to automatically disable ESLint rules that oxlint already covers (see `eslint.config.ts`). If you see duplicates:
1. Ensure `.oxlintrc.json` is up to date after adding new oxlint rules
2. Run `pnpm lint` (which runs oxlint then ESLint in sequence) rather than running them individually
---
### Dependency and Package Issues
#### Q: "Package not found" after adding a dependency
**Symptoms:**
- Module not found after `pnpm add`
- Import errors for newly installed packages
**Solutions:**
1. **Ensure you installed in the correct workspace** (see `pnpm-workspace.yaml` for available workspaces):
```bash
# Example: install in a specific workspace
pnpm --filter <workspace-name> add <package>
```
2. **Clear pnpm cache:**
```bash
pnpm store prune
pnpm install
```
---
#### Q: Lock file conflicts after merge/rebase
**Symptoms:**
- Git conflicts in `pnpm-lock.yaml`
- Dependency resolution errors
**Solutions:**
1. **Regenerate lock file:**
```bash
rm pnpm-lock.yaml
pnpm install
```
2. **Or accept upstream lock file:**
```bash
git checkout --theirs pnpm-lock.yaml
pnpm install
```
---
### Testing Issues
#### Q: Tests fail locally but pass in CI
**Symptoms:**
- Flaky tests
- Different results between local and CI
**Solutions:**
1. **Run tests in CI mode:**
```bash
CI=true pnpm test:unit
```
2. **Clear test cache:**
```bash
pnpm test:unit --no-cache
```
3. **Check Node version matches CI** (see `.nvmrc` for the required version):
```bash
node --version
nvm use # If using nvm — reads .nvmrc automatically
```
---
### Git and Branch Issues
#### Q: Changes from another branch appearing in my branch
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.