A major, full rewrite of linear mode, now under the name "Simple Mode".
- Fixes widget styling
- Adds a new simplified history
- Adds support for non-image outputs
- Supports right sidebar
- Allows and panning on the output image preview
- Provides support for drag and drop zones
- Moves workflow notes into a popover.
- Allows scrolling through outputs with Ctrl+scroll or arrow keys
The primary means of accessing Simple Mode is a toggle button on the
bottom right. This button is only shown if a feature flag is enabled, or
the user has already seen linear mode during the current session. Simple
Mode can also be accessed by
- Using the toggle linear mode keybind
- Loading a workflow that that was saved in Simple Mode workflow
- Loading a template url with appropriate parameter
<img width="1790" height="1387" alt="image"
src="https://github.com/user-attachments/assets/d86a4a41-dfbf-41e7-a6d9-146473005606"
/>
Known issues:
- Outputs on cloud are not filtered to those produced by the current
workflow.
- Output filtering has been globally disabled for consistency
- Outputs will load more items on scroll, but does not unload
- Performance may be reduced on weak devices with very large histories.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7734-linear-v2-2d16d73d3650819b8a10f150ff12ea22)
by [Unito](https://www.unito.io)
## Summary
Setup the variants and migrate existing uses of
TextButton/TextIconButton/IconButton to a single Button component.
Still a work in progress.
## Changes
- **What**: Add a new Button
- **What**: Migrate old buttons
- **What**: Delete old buttons
- **Dependencies**: CVA, upgrade Storybook
## Review Focus
<!-- Critical design decisions or edge cases that need attention -->
<!-- If this PR fixes an issue, uncomment and update the line below -->
<!-- Fixes #ISSUE_NUMBER -->
## Screenshots (if applicable)
<!-- Add screenshots or video recording to help explain your changes -->
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7537-WIP-Component-Button-migration-2cb6d73d36508156a81bfc7bbddb36e9)
by [Unito](https://www.unito.io)
---------
Co-authored-by: GitHub Action <action@github.com>
## Summary
Removes the user.css put at top of the index.html when building for
cloud.
On local, now compiles to this (pictured):
<img width="1909" height="184" alt="image"
src="https://github.com/user-attachments/assets/be03beea-35e9-47d6-a293-08f2971b04be"
/>
Formatted, that looks like:
```html
<!doctype html>
<html lang="en">
<head>
<link rel="stylesheet" href="user.css">
<link rel="stylesheet" href="api/userdata/user.css">
<meta charset="UTF-8">
<title>ComfyUI</title>
<meta name="viewport" content="width=device-width,initial-scale=1,user-scalable=no">
<link rel="stylesheet" href="materialdesignicons.min.css"/>
<meta name="mobile-web-app-capable" content="yes">
<meta name="apple-mobile-web-app-status-bar-style" content="black">
<link rel="manifest" href="./assets/manifest-CebUEmtR.json">
<script type="module" crossorigin src="./assets/index-DuwpHar_.js"></script>
<link rel="modulepreload" crossorigin href="./assets/vendor-other--dOoND1c.js">
<link rel="modulepreload" crossorigin href="./assets/vendor-primevue-BPXiTI_h.js">
<link rel="modulepreload" crossorigin href="./assets/vendor-vue-RrbnUvXR.js">
<link rel="modulepreload" crossorigin href="./assets/vendor-xterm-BZLod3g9.js">
<link rel="modulepreload" crossorigin href="./assets/vendor-three-aR6ntw5X.js">
<link rel="modulepreload" crossorigin href="./assets/vendor-tiptap-BVGjFCxT.js">
<link rel="stylesheet" crossorigin href="./assets/vendor-other-DODGPXtn.css">
<link rel="stylesheet" crossorigin href="./assets/vendor-xterm-BKlWQB97.css">
<link rel="stylesheet" crossorigin href="./assets/index-CX9dQXxD.css">
</head>
<body class="litegraph grid">
<div id="vue-app"></div>
</body>
</html>
```
On cloud, this:
<img width="1911" height="1106" alt="image"
src="https://github.com/user-attachments/assets/bbf6046b-e2fd-4e02-bb71-cba27f579271"
/>
## Context
On the cloud distribution, there are currently 401 errors appearing in
the console from requests attempting to load custom user stylesheets:
- `https://cloud.comfy.org/user.css` (returns 200)
- `https://cloud.comfy.org/api/userdata/user.css` (returns 401)
This is a feature inherited from local ComfyUI that allows users to add
custom stylesheets. The implementation naively requests the stylesheet
from the server, and if the user has added one, it gets loaded;
otherwise, the request fails.
This PR removes the custom stylesheet loading from the cloud
distribution by removing it from teh index.html and only re-injecting it
on non-cloud builds.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-7442-remove-user-css-on-cloud-to-prevent-failed-requests-on-startup-2c86d73d3650813d82a0deb3b01cee74)
by [Unito](https://www.unito.io)
## Summary
- Adds [Oxc linter](https://oxc.rs/docs/guide/usage/linter) as a dev
dependency
- Creates minimal `.oxlintrc.json` configuration file
- Integrates oxlint into the lint workflow (runs before ESLint)
- Adds `pnpm oxlint` script for standalone usage
- **NEW**: Adds
[eslint-plugin-oxlint](https://github.com/oxc-project/eslint-plugin-oxlint)
to disable redundant ESLint rules
- Updates `CLAUDE.md` documentation with oxlint command
## Motivation
Oxc is a high-performance Rust-based linter that is 50-100x faster than
ESLint. By integrating it into our lint workflow, we get:
- **Faster CI/CD pipelines** (5% improvement in this codebase)
- **Quicker local development feedback**
- **Additional code quality checks** that complement ESLint
- **Reduced duplicate work** by disabling ESLint rules that oxlint
already checks
## Changes
- **package.json**: Added `oxlint` and `eslint-plugin-oxlint` to
devDependencies, integrated into `lint`, `lint:fix`, and `lint:no-cache`
scripts
- **pnpm-workspace.yaml**: Added `eslint-plugin-oxlint` and
`mixpanel-browser` to catalog
- **eslint.config.ts**: Integrated `eslint-plugin-oxlint` to
automatically disable redundant ESLint rules
- **.oxlintrc.json**: Created minimal configuration file with schema
reference
- **CLAUDE.md**: Added `pnpm oxlint` to Quick Commands section
- **.gitignore**: Added `core` dump files
## CI/CD Performance Benchmark
Real-world CI/CD timing from GitHub Actions workflow runs:
### Baseline (ESLint only) - [Run
#18718911051](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/18718911051)
- Run ESLint with auto-fix: **125s**
- Final validation (lint + format + knip): **16s**
- **Total: 141s**
### With Oxlint (oxlint + ESLint) - [Run
#18719037963](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/18719037963)
- Run ESLint with auto-fix (includes oxlint): **118s**
- Final validation (includes oxlint + lint + format + knip): **16s**
- **Total: 134s**
### Results
✅ **7 seconds faster (5.0% improvement)** despite running an additional
linting pass
### Analysis
The oxlint integration actually **improves** CI/CD performance by ~5%.
This unexpected improvement is likely because:
1. **Oxlint catches issues early**: Some code that would have slowed
down ESLint's parsing/analysis is caught by oxlint first
2. **ESLint cache benefits**: The workflow uses `--cache`, and oxlint's
fast execution helps populate/validate the cache more efficiently
3. **Parallel processing**: Modern CI runners can overlap some of the
I/O operations between oxlint and ESLint
Even if oxlint added overhead, the value proposition would still be
strong given its additional code quality checks and local development
speed benefits. The fact that it actually speeds up the pipeline is a
bonus.
## eslint-plugin-oxlint Performance Impact
Benchmark comparing ESLint performance with and without
eslint-plugin-oxlint:
### Baseline (ESLint without plugin) - [Run
#18723242157](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/18723242157)
- Run ESLint with auto-fix: **122s** (2m 2s)
- Final validation: **17s**
### With eslint-plugin-oxlint - [Run
#18723675903](https://github.com/Comfy-Org/ComfyUI_frontend/actions/runs/18723675903)
- Run ESLint with auto-fix: **129s** (2m 9s)
- Final validation: **12s**
### Results
**Performance: +7 seconds ESLint, -5 seconds validation (net +2
seconds)**
The eslint-plugin-oxlint integration has a **minimal performance
impact** (+2 seconds total). The slight increase in ESLint time is
likely due to the additional plugin configuration overhead, while the
validation step is faster because fewer redundant lint warnings need to
be processed.
### Benefits
The small performance cost is outweighed by important benefits:
1. **Prevents duplicate work**: Disables ~50 ESLint rules that oxlint
already checks (e.g., `no-constant-condition`, `no-debugger`,
`no-empty`, etc.)
2. **Reduces noise**: Eliminates redundant lint warnings from two tools
checking the same thing
3. **Cleaner workflow**: One authoritative source for each type of lint
check
4. **Best practice**: Recommended by the Oxc project for ESLint + oxlint
integration
5. **Consistent results**: Ensures both tools don't conflict or give
contradictory advice
## Usage
```bash
# Run oxlint standalone
pnpm oxlint
# Run full lint workflow (oxlint + ESLint)
pnpm lint
pnpm lint:fix
```
## Notes
- Oxlint now runs as part of the standard `pnpm lint` workflow
- The configuration uses minimal rules by default (Oxc's philosophy is
"catch erroneous or useless code without requiring any configurations by
default")
- Oxlint provides fast feedback while ESLint provides comprehensive
checks
- eslint-plugin-oxlint automatically manages rule conflicts between the
two tools
- Both tools complement each other in the linting pipeline
🤖 Generated with [Claude Code](https://claude.com/claude-code)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6197-chore-Add-Oxc-linter-to-project-2946d73d3650818cbb55ef9c0abdb9b9)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: GitHub Action <action@github.com>
Co-authored-by: DrJKL <DrJKL0424@gmail.com>
## Summary
- Replace hardcoded `<base href="/">` in index.html with dynamic vite
base config
- Set `base: DISTRIBUTION === 'cloud' ? '/' : ''` in vite.config.mts
- Ensures proper asset loading across different deployment contexts
## Test plan
- [ ] Verify cloud distribution builds work correctly
- [ ] Verify localhost/desktop distributions work correctly
- [ ] Test asset loading in both contexts
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6562-fix-vite-use-dynamic-base-URL-based-on-cloud-distribution-2a06d73d365081c8b5d2e58870ebd14d)
by [Unito](https://www.unito.io)
## Summary
- Add meta tags plugin for social media previews (Twitter, Facebook,
LinkedIn, etc.)
- Include SEO meta tags (title, description, keywords)
- Only applies to cloud distribution (`DISTRIBUTION === 'cloud'`)
## Changes
- Added Open Graph meta tags for better social media link previews
- Added Twitter Card meta tags for Twitter sharing
- Added SEO meta tags (title, description, keywords)
- Added `og-image.png` for preview image
- Meta tags only inject when `DISTRIBUTION === 'cloud'` to avoid
affecting other distributions
## Test plan
- [x] Build with `DISTRIBUTION=cloud pnpm build`
- [x] Verify meta tags appear in built HTML
- [ ] Test link sharing on Twitter, Facebook, Slack
- [ ] Verify meta tags don't appear in localhost/desktop builds
🤖 Generated with Claude Code
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6571-feat-Add-Open-Graph-and-Twitter-meta-tags-for-cloud-distribution-2a16d73d365081e58c73f6009513b2bb)
by [Unito](https://www.unito.io)
## Summary
Backport outputs from new cloud history endpoint
Does:
1. Show history in the Queue
2. Show outputs from prompt execution
Does not:
1. Handle appending latest images generated to queue history
2. Making sure that workflow data from images is available from load
(requires additional API call to fetch)
Most of this PR is:
1. Test fixtures (truncated workflow to test).
2. The service worker so I could verify my changes locally.
## Changes
- Add `history_v2` to `history` adapter
- Add tests for mapping
- Do branded validation for promptIds (suggestion from @DrJKL)
- Create a dev environment service worker so we can view cloud hosted
images in development.
## Review Focus
1. Is the dev-only service work the right way to do it? It was the
easiest I could think of.
4. Are the validation changes too heavy? I can rip them out if needed.
## Screenshots 🎃https://github.com/user-attachments/assets/1787485a-8d27-4abe-abc8-cf133c1a52aa
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6288-Feat-history-v2-outputs-2976d73d365081a99864c40343449dcd)
by [Unito](https://www.unito.io)
---------
Co-authored-by: bymyself <cbyrne@comfy.org>
## Summary
Implements server-side remote configuration to decouple runtime behavior
from build artifacts, enabling dynamic configuration updates without
redeployment.
## Technical Changes
- **Replaced** build-time constants (`__MIXPANEL_TOKEN__`,
`__BUILD_FLAGS__`) with runtime configuration loaded from
`/api/features`
- Configuration now sourced from `window.__CONFIG__` (hydrated from
`/api/features` endpoint)
- **Added** `src/platform/remoteConfig/` service that polls server
configuration every 30 seconds
- **Modified** application bootstrap sequence in `main.ts` to load
remote config before module initialization (required for cloud builds)
- **Removed** global constants: `__BUILD_FLAGS__`, `__MIXPANEL_TOKEN__`.
Runtime subscription enforcement toggle via `subscription_required` flag
- Server health alerts with variant-based severity rendering
(info/warning/error) via topbar badges
## Rationale
- **Build-once-deploy-anywhere**: Single immutable artifact promoted
through environments (staging → production)
- **Zero-downtime configuration**: Update behavior without rebuilding or
redeploying the application
- **Incident response**: Disable features or display alerts dynamically
in response to outages or degraded service
- **Instant rollback**: Revert configuration changes server-side without
artifact redeployment
- **Progressive delivery**: Enable A/B testing, canary releases, and
user/region-based configuration
- **Environment parity**: Eliminate configuration drift between staging
and production builds
- Decouples deployment cadence from configuration changes
- Enables GitOps workflows for configuration management separate from
code deployments
- Supports real-time operational control of client behavior
- Reduces build matrix complexity (no per-environment builds)
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6246-change-cloud-feature-flags-to-be-loaded-dynamically-at-runtime-rather-than-set-in-build-2966d73d3650811cbb41c9093961037a)
by [Unito](https://www.unito.io)
## Summary
This code is entirely excluded from open-source, local, and desktop
builds. During minification and dead-code elimination, the Mixpanel
library is fully tree-shaken -- meaning no telemetry code is ever
included or downloaded in those builds. Even the inline callsites are
removed during the build (because `isCloud` becomes false and the entire
block becomes dead code and is removed). The code not only has no
effect, is not even distributed in the first place. We’ve gone to great
lengths to ensure this behavior.
Verification proof:
https://github.com/user-attachments/assets/b66c35f7-e233-447f-93da-4d70c433908d
Telemetry is *enabled only in the ComfyUI Cloud environment*. Its goal
is to help us understand and improve onboarding and new-user adoption.
ComfyUI aims to be accessible to everyone, but we know the learning
curve can be steep. Anonymous usage insights will help us identify where
users struggle and guide us toward making the experience more intuitive
and welcoming.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6154-add-telemetry-provider-for-cloud-distribution-2926d73d3650813cb9ccfb3a2733848b)
by [Unito](https://www.unito.io)
---------
Co-authored-by: Claude <noreply@anthropic.com>
## Summary
Added a `manualChunks` strategy in `vite.config.mts` that splits
primevue, tiptap, chart.js, three/@xterm, core Vue/Pinia, and the
remaining dependencies into dedicated vendor bundles. This reduces the
main application chunk size and allows browsers to cache heavy
third-party code across releases, improving load times when those
libraries stay unchanged.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6137-perf-manually-chunk-vendored-code-2916d73d36508140a44ec0de228ef9cc)
by [Unito](https://www.unito.io)
## Summary
Introduces a `GENERATE_SOURCEMAP` environment flag in `vite.config.mts`
that defaults to enabled (`true` unless set to `'false'`). This keeps
source maps on by default, while allowing opt-out for lean production
artifacts.
Allows the choice to be made as part of the distribution pipeline and
changed for different targets.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6134-ci-allow-setting-GENERATE_SOURCEMAP-as-env-var-2916d73d3650815d91b9eff12b6e55fd)
by [Unito](https://www.unito.io)
## Summary
Establishes distribution-specific code pattern using compile-time
constants and dead code elimination. Demonstrates with Help Center by
hiding extension manager and update buttons in cloud distribution.
Below commentary makes assumption that minifcation and tree-shaking is
enabled (which isn't true yet, but will be eventually).
## Changes
- **What**: Added `src/platform/distribution/types.ts` with distribution
detection via `__DISTRIBUTION__` variable
- **Build**: Vite replaces `__DISTRIBUTION__` at build time using
environment variables
- **Tree-shaking**: All code not relevant to target distribution is
DCR'd and eliminated from bundle
- **Example**: Help Center hides "Manager Extension" menu item and
"Update" buttons in cloud builds
## Pattern
This PR defines a `__DISTRIBUTION__` variable which gets replaced at
build time by Vite using environment variables. All code not relevant to
the given distribution is then DCR'd and tree-shaken.
For simple cases (like this Help Center PR), import `isCloud` and use
compile-time conditionals:
```typescript
import { isCloud } from '@/platform/distribution/types'
if (!isCloud) {
items.push({
key: 'manager',
action: async () => {
await useManagerState().openManager({ ... })
}
})
}
```
The code is DCR'd at build time so there's zero runtime overhead - we
don't even incur the `if (isCloud)` cost because Terser eliminates it.
For complex services later, we'll add interfaces and use an index.ts
that exports different implementations under the same alias per
distribution. It will resemble a DI container but simpler since we don't
need runtime discovery like backend devs do. This guarantees types and
makes testing easier.
Example for services:
```typescript
// src/platform/storage/index.ts
import { isCloud } from '@/platform/distribution/types'
if (isCloud) {
export { CloudStorage as StorageService } from './cloud'
} else {
export { LocalStorage as StorageService } from './local'
}
```
Example for component variants:
```typescript
// src/components/downloads/index.ts
import { isCloud } from '@/platform/distribution/types'
if (isCloud) {
export { default as DownloadButton } from './DownloadButton.cloud.vue'
} else {
export { default as DownloadButton } from './DownloadButton.desktop.vue'
}
```
## Implementation Details
Distribution types (`src/platform/distribution/types.ts`):
```typescript
type Distribution = 'desktop' | 'localhost' | 'cloud'
declare global {
const __DISTRIBUTION__: Distribution
}
const DISTRIBUTION: Distribution = __DISTRIBUTION__
export const isCloud = DISTRIBUTION === 'cloud'
```
Vite configuration adds the define:
```typescript
const DISTRIBUTION = (process.env.DISTRIBUTION || 'localhost') as
| 'desktop'
| 'localhost'
| 'cloud'
export default defineConfig({
define: {
__DISTRIBUTION__: JSON.stringify(DISTRIBUTION)
}
})
```
## Build Commands
```bash
pnpm build # localhost (default)
DISTRIBUTION=cloud pnpm build # cloud
DISTRIBUTION=desktop pnpm build # desktop
```
## Future Applications
This pattern can be used with auth or telemetry services - which will
guarantee all the telemetry code, for example, is not even in the code
distributed in OSS Comfy whatsoever while still being able to develop
off `main`.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-6028-Add-distribution-detection-pattern-28a6d73d365081b08767d395472cd1bc)
by [Unito](https://www.unito.io)
## Summary
Restricts Vite's dependency optimization scanning entry points to the
root `index.html`, preventing excessive error messages from scanning
unintended files in other packages.
## Changes
- **What**: Adds `entries: ['index.html']` to `optimizeDeps` in
vite.config.mts
- **Breaking**: None
## Review Focus
May require additional entries or inversion - exclude-based approach
instead of include-based - if possible.
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5934-Restrict-Vite-entry-point-to-index-html-2846d73d3650816fbf45fa29493891ae)
by [Unito](https://www.unito.io)
## Summary
Extracts shared formatting and network utilities into dedicated
workspace package.
## Changes
- **What**: Created `@comfyorg/shared-frontend-utils` package containing
formatUtil and networkUtil
- **Breaking**: None - utilities remain accessible via path aliases in
`tsconfig`
Split `createAnnotatedPath` and `electronMirrorCheck` out and left in
frontend, due to their tightly-coupled nature. See [discussion on this
PR](https://github.com/Comfy-Org/ComfyUI_frontend/pull/5843#issuecomment-3344724727).
## Summary
Consolidates Tailwind configuration and styles into a shared
`@comfyorg/design-system` package for reuse across monorepo apps.
The goal was not to make changes to how the design system works; merely
to separate it cleanly. I _would_ strongly recommend some drastic
sweeping changes, however I believe that should be done after the
migration.
## Changes
- **What**: Migrates CSS files, Tailwind config, and custom icons to
design-system package
- **Dependencies**: Moves `@iconify-json/lucide` and `@iconify/tailwind`
to design-system package
## Summary
Prerequisite refactor/cleanup to use a global store instead of having
nodes throw up events to a parent component that stores a reference to a
singleton service that itself bootstraps and synchronizes with a
separate service to maintain a partially reactive but not fully reactive
set of states that describe some but not all aspects of the nodes on
either the litegraph, the vue side, or both.
## Changes
- **What**: Refactoring, the behavior should not change.
- **Dependencies**: A type utility to help with Vue component props
## Review Focus
Is there something about the current structure that this could affect
that would not be caught by our tests or using the application?
┆Issue is synchronized with this [Notion
page](https://www.notion.so/PR-5695-Refactor-Composable-disentangling-2746d73d365081e6938ce656932f3e36)
by [Unito](https://www.unito.io)
* feat: Initial shadcn configuration
* component: Add Slider component from shadcn-vue
* deps: Add tw-animate-css
* component: Align slider with Figma styles
* component: Set the step value for the slider, update styles
* fix: update component tests to work with Array of values
* vite: Don't reload dev server for test changes
* component: Swap text for a number input kept in sync with the slider
* cleanup: Don't need the override if the input isn't type="number"
* test: add step size tests
* cleanup: Don't need cn for these
* css: Update token names to match new Figma Variables
* lint: Fix camelCase vs train-case in passthrough
* feat: If the value is deleted, revert to the slider state cc: @PabloWiedemann
* feat: Improve cursor styles, grabbable thumb, clickable track
* lint: temporarily disable some warnings
* feat: Grabbing while sliding (most of the time)
* temp: move tailwind calls out of the layer
* temp: ts tailwind config
* upgrade: Tailwind v4
This got a little out of hand.
Had to add a relative reference to the stylesheet in any component that uses @apply instead of the utility classes directly.
* upgrade: bg-opacity is now a modifier
* fix: Classic menu buttons assume a border
* Update test expectations [skip ci]
* fix: New preflight removal pattern
* fix: Skeletons don't have skin
* Update test expectations [skip ci]
* fix: Missing @reference
* [auto-fix] Apply ESLint and Prettier fixes
---------
Co-authored-by: github-actions <github-actions@github.com>
Co-authored-by: GitHub Action <action@github.com>
* fix: [@vue/compiler-sfc] defineModel is a compiler macro and no longer needs to be imported.
* fix: Duplicate name conflict/warning from unplugin-vue-components
* fix: enforce correct line endings for the commonjs and esm variants via git
Since litegraph is now integrated as a git subtree rather than an
external npm dependency, it no longer needs to be excluded from
Vite's dependency optimization.
