A backup now keeps each model's scan file (.info, small; on by default)
so names, trigger words and base models survive Civitai removing a
model, and on request Civitai's covers (stored once as the civitai_bak
copy; they can be hundreds of MB). Putting back writes a scan file only
where the model has none (replaced only with "replace") and Civitai's
cover only where it has no cover, with its civitai_bak copy so it stays
Civitai's and a later user cover still wins.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A backup also keeps a model list (each model's folder, path, size,
SHA-256 and link). Inspecting it names the models this computer lacks
(two copies of one file are one download); "Find downloads..." looks
them up like Model Check (Civitai by hash, the saved link, Hugging
Face), shows size, source and free space, and downloads the ticked ones
one after another into the path they had, each checked against its hash
and scanned after. A file found by name only starts unticked. Verified
on the isolated ComfyUI: an 18 MB Civitai LoRA came back to its folder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Export packs the stores (recipes, combos, materials, parameters, and
optionally ComfyUI's saved workflows), the plugin settings without the
Civitai key, and each model's user layer and own cover into one .zip to
keep on a cloud drive. Import lists what is new, the same or different
and the models it finds again (by SHA-256, else folder, path and size),
adds only what is missing by default, replaces differing files only on
request with this computer's going to the Recycle Bin first, merges a
model's name and notes field by field and puts the user's cover in
place of a Civitai one. Unsafe or foreign zips are refused. Both are
activity entries. tests/test_backup.py covers the rules; testing.md
names a single test file the way unittest finds it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The recurring traps (a :has() rule slowing every canvas frame, fixes
left on an unpushed local main, release checks, imports lost in a module
split) are now docs/architecture/lessons.md, so the links from
ARCHITECTURE.md, AGENTS.md and the docs indexes open for everyone.
AGENTS.md: tests live in tests/, .agents stays a local folder for
screenshots and logs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
tests/ (Python backend tests, .mjs frontend rule tests) is no longer
ignored. tools/run_tests.mjs runs the structure check, every
tests/*.mjs and the Python tests with $PYTHON, the portable
python_embeded or python. docs/guides/testing.md says how to run and
write them; README links it. Developer notes no longer point at 18 test
files that were removed earlier (AGENTS.md table, backend, frontend and
update-guide notes).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Seven short "see it in action" steps in English and Chinese, each with a
still or a few-second animation (WebP, about 500 KB in all) from the
isolated demo ComfyUI: scan results, Model Check putting models back,
gallery search and drag to canvas, a combo dragged onto the canvas,
Prompt Studio tags, the activity log with Undo, character voices; the
model cover wall under the title. .gitignore keeps docs/images/*.webp.
The Activity row no longer says it only records: canvas changes have Undo.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each slot names its node (and node pack when not ComfyUI's own). A small
"译" button beside a prompt shows its translation in a read-only panel
(ui_translation_peek.js). On nodes with several models, saved values go
to the row with the same number (strength_02 under lora_02) and more
than four fold behind "+N"; empty optional rows (None) start unticked in
the save dialog so a ten-row LoRA stack does not make ten empty slots.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Models, LoRAs and prompts in their own groups, in the order they flow
(LoRA 1 next to the model, positive before negative); made-up names
renumbered along the chain. Covers looked up for every model, a whole
row opens the picker, names wrap to two lines, each model node's saved
values (strength, CLIP strength, type) shown read-only, prompt boxes
grow with their text. New captures keep nodes in flow order. The combo
editor's sticky top bar sat 20 px down with the page showing above it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The canvas's node menu (the frontend's getNodeMenuItems hook) offers
"Save as combo" for the picked nodes, or the right-clicked node when it
is not among them, when any of them is a model or text node; it opens
the same save dialog without opening the browser first, and the browser
shows the new combo afterwards.
Values other than the slots stay as they were saved (agreed): a combo
always puts down exactly what was saved; other values mean another combo.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A combo can now hold its own node structure: pick nodes on the canvas and
New -> From the nodes picked on the canvas keeps the model nodes (a
drop-down of model files) and the text nodes (a multiline text box by the
node's definition) with the links between them; other picked nodes and
links to them are left out, their ports stay open on the canvas.
Slots are per box, so a node with several model drop-downs has several; a
box fed by a link is no slot. Each model slot's folder is the one whose
files hold all its drop-down's model-file options (ComfyUI's /models
lists), never guessed from names. Every other box is stored by name and
put back untouched, in order (a count box adds its rows first).
The save dialog shows what is kept and dropped and lets slots be renamed
or unticked. The editor lists the slots: a picker of that folder's models
with covers, filtered by base model, and text boxes. Put on canvas makes
the nodes (missing node types and their packs are named before anything
is placed), fills the slots by name, links by port name and type, as one
Ctrl+Z step; a box or link that cannot be restored is reported, never
guessed. New -> Same structure as ... starts another combo from a saved
structure. The default combo is unchanged. MCP list_combos lists node
combos by slot and place_combo places them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Downloads: with the China mirror on, a Civitai match whose very file (same
SHA-256) is on Hugging Face too, by the workflow's link or ComfyUI-Manager's
list, is fetched from Hugging Face; Civitai's name, base model and page stay.
- A file downloaded from Hugging Face gets its model card's first example
image as cover and its base model, trigger words and example prompts as
notes, with the repository as link (api/hf_card.py), never over what the
user or a scan set.
- Text encoders, CLIP vision and upscalers are shown (and scanned) by
default, so shared workflows carry their fingerprints; older configs get
them turned on once (folder_defaults_version 2).
- The models page's search sits in the header (ui_model_search.js) instead
of inside the folder list; it searches the whole type by name, file name,
folder, trigger words, base model and notes, every word; Esc clears it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Models not on Civitai (Flux's VAE, T5 / CLIP text encoders, upscalers,
ControlNets...) are looked up by file name in the model-list.json of the
ComfyUI-Manager installed here (api/manager_catalog.py), keeping entries
that fit the node's folder type. With the workflow's fingerprint only the
listed file with that SHA-256 counts (same-name entries are each asked);
without one, a single entry is offered as "Download by name" with a
warning, never in Download all or the MCP's download_missing_models.
A site that stops answering is skipped for the rest of a lookup while the
others are still asked, so Civitai failing in mainland China no longer
stops Hugging Face through its mirror. GitHub raw file links count as
sources; a Hugging Face file behind a login offers its repository page.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three MCP tools: download_missing_models (Model Check's Download all for
the open workflow, through the page so the placement rules and the put-in
afterwards are the UI's own), download_model (one model from a Civitai
version link, a Hugging Face / GitHub file link or a Civitai hash, into a
folder of its type) and download_status. The default models folder of a
type is shared by the dialog and the tools.
A Civitai version link without a matching file name now takes the
version's primary model file (people rename their files); a hash, when
there is one, still decides.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Settings > Model downloads can send Hugging Face downloads through
hf-mirror.com, on by default for the Chinese interface. The lookup and the
download use the chosen host whichever one the link names; when the mirror
sends a network outside China back to huggingface.co, the lookup follows
once for the size and SHA-256. A failed download from Hugging Face itself
suggests the mirror.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Model Check offers Download (and Download all) for a missing model whose
source names one exact file: Civitai by the workflow's fingerprint, or a
Civitai version / Hugging Face / GitHub file link the workflow carries.
A dialog shows each file's size, source and full destination, with the
models folder (when there are several) and subfolder changeable; Settings
> Model downloads sets the default: where the workflow says (the node
needs no change) or a set folder, {base} sorting by base model and
matching an existing folder in any spelling.
Downloads run one at a time into <file>.part, continue after an
interruption, are checked against the SHA-256 before they get their name,
never replace a file, and send the Civitai key to Civitai only. Finished
files are put into their nodes (one Ctrl+Z step each), scanned for info
and covers, and recorded in the activity log.
A change of fingerprints alone (after any scan) no longer takes a Ctrl+Z
step that seemed to do nothing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Current node's "Insert LoRA before / after" refused the most common workflows: a
checkpoint whose CLIP feeds both prompts (one link per output only) and UNet workflows
with their own CLIP loader (MODEL and CLIP had to come from one node).
- graph_splice.js: planLoraInsertion decides the links (after a node: every link of its
MODEL output; before one: only its own MODEL input) and, for LoRAs that train the text
encoder, the CLIP line (same node, a UNet workflow's one CLIP source, or before a
sampler the source its prompts share); spliceLora wires it, all or nothing. The old
analyzeModelChainInsertion / spliceModelChainNode / spliceAfterOutputs are gone.
- api/lora_info.py: GET /anomalous/lora_info reads the LoRA's safetensors header; the
picker asks it on confirm and inserts a model-only loader when there are no
text-encoder weights. The MCP add_lora uses the same helper.
- Insertion messages speak of the MODEL line; the fan-out ones are removed.
- tests/graph_splice.mjs (local) covers checkpoint, UNet + CLIP loader, before a
sampler, and the rollback.
Verified on 8199 through the Current node picker: after a checkpoint feeding two prompts
(full LoRA, both prompts moved) and before a sampler (model-only Flux LoRA).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A model-only LoRA loader was used for every UNet workflow, so a LoRA with text-encoder
weights lost them; and inserting after a checkpoint whose CLIP feeds both prompts was
refused (one link per output only).
- api/mcp_actions.py reads the LoRA's safetensors header: lora_te / te1 / te2 /
text_encoder keys mean it trains the text encoder.
- Such a LoRA becomes a LoraLoader taking MODEL from the model line's end and CLIP from
the same node (checkpoint, full LoRA loader) or a UNet workflow's one CLIP source;
others stay model-only.
- graph_splice.js: spliceAfterOutputs puts a node after outputs on different nodes and
moves every link each output had. The picker's spliceModelChainNode is back to its
committed behavior (the `types` option is gone).
Verified on 8199: UNet + CLIP loader with a text-encoder LoRA, a model-only one after it
and another text-encoder one; a checkpoint feeding two prompts; Ctrl+Z after each.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The second batch of MCP tools: they change things, through the same code the UI uses.
- Canvas, done by the open ComfyUI page: describe_canvas, set_prompt, set_model,
add_lora, place_combo, check_workflow_models, fix_workflow_models, run_workflow,
open_in_anomalous. api/mcp_bridge.py sends the action over ComfyUI's websocket and
waits; web/modules/mcp_bridge.js claims it (visible tab first), does it with
Anomalous's canvas code and answers. Each change is one Ctrl+Z step, shows a toast and
is logged as the AI's (activity entries carry via: "mcp", with their undo).
- This server: scan_models (new / look up again / picked models, via /anomalous/scan_all),
list_voices and speak (Anomalous TTS through /prompt and /history).
- graph_splice.js takes the chain's slot types, so a LoRA can go into a model-only (UNet,
Flux) chain as LoraLoaderModelOnly; the default (MODEL and CLIP) is unchanged.
- Nothing deletes or renames files; acting tools are not marked read-only, so AI apps
ask before calling them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An MCP endpoint (Streamable HTTP) inside ComfyUI's server, so Claude, Cursor, Cherry
Studio and other MCP apps can ask about the user's library without anything extra
installed and without memory or VRAM while nobody is connected.
- api/mcp_server.py: one JSON-RPC message per POST, answered with JSON. Serves the
2026-07-28 stateless protocol (headers must match the body) and the earlier
initialize-based versions; no sessions. This computer only: loopback peer, no proxy
headers, loopback Host and Origin (DNS rebinding).
- api/mcp_tools.py: 11 read-only tools: library overview, model search and details
(optional cover), scan status, output-image search and generation settings
(optional thumbnail), combos, workflow recipes, saved prompts, generated audio,
activity log. Ids resolve inside their roots; tool failures come back as isError.
- docs/guides/mcp.md: setup for each app, what can be asked, privacy note.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- The bar starts with "Model Check:" so it is clear it comes from the browser.
- "Don't show again" turns the check on opening off and says where to turn it
back on; Settings → Workflows has the same switch (on by default), and turning
it on checks the open workflow right away. While off, the bar never shows.
- The tour's Model Check step mentions the button.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- While a combo group follows the pointer, Esc removes it again; the activity log
holds back until it is put down, so a cancelled group leaves no entry.
- The positive and negative prompt nodes are always laid out (positive empty when
the combo has no prompt); the negative takes Prompt Studio's common negative card
(starter cards moved to prompt_composition.js as PROMPT_PRESETS).
- The UNet "connect clip yourself" message shows once the group is down; it used to
pop up while the group followed the pointer, and closing it put the group down.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
"Use" tried to put a combo into the open workflow (main loader, LoRA
chain rebuilt, positive prompt) with a list of what changed and what was
skipped and why, which read as guesswork. A combo is a shortcut: its one
button, Put on canvas (放到画布), and dragging its card both lay out the
main model loader, LoRAs and prompt nodes as a new wired group, and never
change nodes already on the canvas. An existing node's values are changed
in Current node.
notebook_apply.js, the Use confirmation and its strings are removed; the
editor's More menu keeps only Delete.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A combo card dropped on a main model loader does Use for that loader
(same confirmation), and on empty canvas builds its new group of nodes
where it was dropped; other nodes refuse it with a hint.
The star's keep menu has Remove (取消收藏) beside each kept row: after a
confirmation the recipe, combo or saved prompt goes to the Recycle Bin
through its own delete route, and the star (and the gallery's cached
list) goes back to empty.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
"Use draft" (改草稿) read as "edit the draft". The node's name now has a ✕
that stops editing it, and the draft is labelled as not in any node.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Tag meanings are brighter and slightly larger, and the top bar picks
their language (Simplified/Traditional Chinese, Japanese, Korean,
Russian, Spanish, French, German, Portuguese, or none) for every
interface language; the old Chinese switch carries over. Typed text in
any language that is not plain English is translated to English, and
translating a box's foreign tags drops one whose English is already
there. The backend skips DeepL for targets it does not take instead of
falling back to English, and lets MyMemory detect other sources.
Unused prompt helpers in translation_service.js are removed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The assembly board, its write-to-node step and the positive/negative tabs
are gone. The studio now shows the prompt boxes of the prompt node last
selected on the canvas (or a draft) as tags that are edited in place:
weight, edit, remove, reorder, type to add with Chinese translated to
English, and an optional Chinese meaning under each tag. Every change is
one node write with Ctrl+Z, and the studio has its own Undo.
Cards: three common ones instead of six, plus saved prompts; a saved
prompt's text is read once instead of every 30 seconds. Current node
lists the prompt boxes with Edit in Prompt Studio instead of its own
translation. The inspector modal, starter-card data, plan loading and
about 2,000 lines of CSS for the old board are removed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The page held four unrelated things, so each now lives where it is used:
the rail's Prompts entry (and the Ctrl+Shift+L shortcut) opens Prompt
Studio beside the canvas, where saved prompts are managed; the browser
comes back when the studio closes. Saved node values stay in Current
node, which can now delete them; moved whole workflows are no longer
offered there twice. Recent results, combos and whole workflows already
have their own homes.
Deleted ui_materials.js, ui_material_cards.js, ui_material_detail.js,
ui_material_application.js (its receipt is ui_apply_receipt.js) and
material_feedback.js, the recipe page's copy-to-library buttons and the
save_parameter_material route, with about 2,000 lines of CSS and 200 locale
keys. Words that said "material library" now say where things go.
Material files are untouched.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Library cards in the studio were read-only. Their preview now renames a
saved prompt (tags kept) or deletes it to the Recycle Bin, and any card
can be dragged out of the drawer onto a canvas prompt box (outlined,
previewed, one undo step) or onto empty canvas for a prompt node; over
the drawer the drag passes through to the assembly board.
The preview moved to ui_prompt_card_popover.js (the deck goes from 751 to
480 lines) and the canvas drag is prompt_card_drag.js; material_drag.js
takes passThrough and effectAllowed, prompt_drop.js a known-text drag.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every PNG in the gallery has a star that asks what to keep: the whole
workflow as a Workflow Recipe, its main model, LoRAs and positive prompt
as a combo (models looked up on this computer, missing ones left out), or
its prompts as a saved prompt. What is kept remembers its image, so
GET /anomalous/kept_images fills the star and the menu opens what was
kept. The image workbench's Keep opens the same menu instead of its name
and tags form. Things are named after the prompt's first tags.
The Material Library's Recent shelf and its route are gone: the gallery
lists the same images. Gallery cards moved to ui_gallery_card.js, and
deleting asks in the usual dialog. Two missing locale keys are fixed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A note was always a main model, LoRAs and a prompt kept together, so it is
now a combo (搭配) with its own rail page and Home card: a card per combo
with its model's cover, search and New; a card opens the editor, Back
returns to the list. Use puts the combo into the open workflow when the
canvas has a main model loader, builds a new group of nodes when it has
none, and asks when several loaders leave it unclear. The notes column,
quick-jump list and the Material Library's notes shelf are gone, with
their CSS and words; note files are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ui_material_notes.js lists the notes above the saved prompts, each with
its model, LoRAs and prompt, Put into this workflow and New group of
nodes; a card opens the note's editor, which now has a way back to the
library, and New note starts one. The separate Prompt Notes button goes,
and so does the editor's "save to library" card with the route only it
used (save_prompt_note_material): a note is in the library already.
Notes keep their own files; prompt materials saved from notes before
still list and open.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Recent's star and the image workbench's whole-workflow save now make a
Workflow Recipe with the image as cover (recipe_save.js lays the image's
workflow on a canvas of its own and saves it through the same path as the
canvas save, now shared). The Material Library's "all" lists the pieces
only and loses its Workflow pill. Whole workflows kept as materials
before can be moved from the Workflows page: each becomes a recipe and
its material gains moved_to_recipe, which keeps it out of every listing
while the file stays. Recent marks images kept as recipes and opens them;
partial recipes are "Modules" everywhere.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
notebook_apply.js plans and makes a note's changes to the canvas: the main
loader takes the note's model (same kind only), the LoRA chain right after
it is rebuilt with the note's LoRAs, rewiring MODEL and, for LoraLoader,
CLIP to what the old chain fed (a kept LoRA keeps its strengths), and the
positive prompt box of the sampler that model reaches takes the prompt. A
selected loader or prompt node decides when there are several. The note
editor's "Put into this workflow" shows each change and what is left as
is before writing, as one Ctrl+Z step; "New group of nodes" (formerly
Send to canvas) now wires a UNet note's prompts to the canvas's only CLIP
loader, and says so when there is none.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The overview's models are one compact list (preview, name, node, here
under the saved name or missing, size, note, download page). The recipe's
own match / apply-match flow, identity badges, advanced path and hash
blocks, origin editor and recipe fingerprint are gone; Model Check finds
renamed or moved models once the workflow is on the canvas, and saving
the recipe again keeps it. ui_recipe_model_matching.js becomes
ui_recipe_models.js; ui_recipe_detail.js drops from 776 to 292 lines.
Plain wording for the save dialog, empty page and status; no per-node
save-to-library buttons on the parameters page. Recipe CSS rules for
classes no script uses (107 rules) are removed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When a workflow is serialized (save, export, image metadata, share code),
a model present here without a link kept in the workflow gets its own
information's link (the editor's, else Civitai's), marked auto so a newer
one replaces it, and links for names no node holds any more are dropped;
the live graph is untouched. So after Model Check swaps in a local model,
the export names that file, its fingerprint and link. Fingerprints now
cover GGUF, .pth and .sft too. The share-code export shows how many models
carry a fingerprint, warns about those without, and can leave links out.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
For a workflow without fingerprints, a missing ordinary model took the
local same-name file's hash as its evidence, so it showed as "the same
file under another name" and counted among those put in at once. Only the
workflow's own record is evidence now; the one same-name file the node can
load is a candidate, and a size candidate with the same name is shown as
likely, first. Foundation components still go by fingerprint only. Model
Check also reads GGUF values and resolves upscale models.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Models page, the folder counts, search and lookups only matched
.safetensors, .ckpt and .pt in lower case, so quantized GGUF models and
most upscalers never showed. One list, MODEL_EXTENSIONS, and
is_model_file() now decide what a model file is everywhere.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Folder management was a pop-up listing ComfyUI's raw folder names with
eye/cross emoji, "Category mode / Physical mode" radios and a Save and
reload button. It is now a view of the settings page with Back: grouping as
two plain choices (by model type, by folders on disk), the shown types
first with their usual names, folder names, a switch and up/down, then the
ones that are off, the rarely used ComfyUI folders folded away. Each change
is saved at once and the folder list reloads.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The manual was a long block of text that had fallen behind the window (it
still called recipes and materials "in testing") and was reachable only
from Settings. Pages explain themselves, Home lists what you can do and !
shows what changed, so Settings -> Help and feedback now offers the
interface tour, what's new, the full guide (the README on GitHub, at its
Chinese half in Chinese) and feedback. Removes ui_help.js, its strings and
styles, and the recipe "beta" notice code that no longer had an element.
The README's backup note no longer calls the features a test.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Card thumbnails lived in ComfyUI's temp folder, which ComfyUI empties on
every start, so each first visit after a restart remade them all. They now
live in <user>/anomalous/cache/card_thumbnails (still capped at 256 MB).
Listing a model type queues the missing ones for one background thread;
the first screen of cards loads before the rest. A video cover gets a
first-frame poster (PyAV) and fetches nothing until it plays; hover-play
is now the default, as the setting already recommended.
The gear opens a settings page instead of a pop-up with a dialog inside:
look and language, model cards and memory with the thumbnail cache (size,
empty), folders, workflow fingerprints, opening mode and window, help.
Display preferences live behind owner.displayPrefs. Also removes a dead
overlay rule that shared the new page's id.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Voices has two views switched at the top, Characters and Voice-over, sharing
the character list (clicking a character on Voice-over makes it speak); the
rail's Voices entry reopens the view used last, and the separate Voice-over
rail entry is gone. Generation settings: language as choices and speed as a
slider in view, the sampling numbers folded under fine-tuning with plain
names, no sideways scroll. A finished take says how long it took and links
to the audio gallery. Line cards show editing buttons on hover, the emotion's
reference clip sits beside its chips, and a character with only its main
voice shows no chips. The audio gallery shows each file's lines once and
can open them on Voice-over for another take; the list no longer repeats
the gallery. Also fixes the audio list staying empty after a reload.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Script Director was a drawer behind one of four equal buttons on
Voices, over the character cards, and its name did not say what it does.
It is now Voice-over (配音), a rail page between Voices and Audio
(ui_script_page.js mounts the director); each character card has a
Voice-over button that opens it with that character chosen, and Home's
"Give a character a voice" goes there. Generate is the main button;
pushing, dragging and copying sit under "Into your own workflow", and a
push or drop is one Ctrl+Z step. Plainer words throughout; the Voices
page keeps import, refresh and search.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Saving parameters asked people to decide, then save, while busy
generating. The library's first category, Recent, now lists the newest
output PNGs that carry their workflow, each with the start of its positive
prompt, the main model and steps / CFG / sampler / size (GET
/anomalous/recent_generations, read from PNG text chunks and cached). A star
keeps one as a whole-workflow material through the existing
save_image_material route, named after the prompt's first tags; a kept
image shows a filled star (matched by the image's SHA256, as the save
route's duplicate check does) and opens its material. The picture opens the
image workbench for keeping single nodes; the search box searches Recent
too. The empty-library guide points here.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Sources chip at the end of the model types now shows, in place of the
cards, where each model is downloaded: for this workflow or all models,
filtered by with / without a link, searchable. Each row's link can be
typed or pasted (any site), found, opened and saved with one button; a
link goes to the model's own information (the editor's download link) and,
for the workflow's models, into the workflow too, as one Ctrl+Z step. The
canvas note and the clipboard list stay. Model Check's "Where to download"
opens the view on this workflow.
The pop-up goes, with its lock-then-edit links and four ways of saving;
ui_model_sources.js (815 lines) splits into the view and
model_source_links.js (data and actions). Old hub CSS and locale keys go.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
"Doctor" said little about what it does: check the opened workflow's
models and put back the ones this computer has. The rail, page, home card,
help, tour, README, the unreleased changelog and the architecture docs now
say Model Check (模型检查); the help paragraph now describes the check as
it works today. Code, CSS and locale keys keep "doctor"; released changelog
entries, audits and decisions keep the old name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Pages made at different times used their own greys (slate here, neutral
there) and accents (sky, violet, indigo), and most ignored the Abyssal
Scarlet theme. About 1,850 literal colours in web/styles and in CSS and
inline styles written from JS now use the --amb-* tokens by what they are
for: surfaces, text (new --amb-text-soft and --amb-text-dim), borders (new
--amb-border-strong), the primary button, and one accent (new --amb-link,
tinted with color-mix). Both themes define the new tokens.
Literal colours stay where the colour is the meaning: status, prompt
roles, voice emotions, canvas node colours, shadows and overlays; theme
override rules are untouched. The share-code dialogs move from inline
styles to .anomalous-share-* classes. Stylesheet versions are bumped.
Checked: text contrast on every rail page, Prompt Studio, Prompt Notes,
Model Sources, settings, model detail and editor, share codes, in both
themes; structure check and JS tests pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
While a prompt is over a prompt box, the text a release writes is laid
over that box and the opposite-role box on the same sampler; leaving the
box, the canvas or the drag removes it. The overlay never touches the
widget's value.
Over a node of a type the material holds values for, the hint lists what
would change ("steps 12 -> 28"), as applyMaterialBlock does it: seeds stay,
and a choice this computer lacks is named. Every dragged material is now
fetched when its drag starts (prepareMaterialDrag).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The rail keeps scan, Model Doctor and Current node (ui_rail_tools.js,
which replaces ui_toolbox.js). The tools the toolbox held now sit where
their things are:
- Workflow share codes (AMB0/AMB1 import / export): the Workflows page's
top bar, in place of the disabled "import" button.
- Prompt Studio and Prompt Notes: the Material Library's top bar.
- Model Sources: the end of the type chips on Models (library) and
"where to download" in Model Doctor (the open workflow).
Removed with it: the toolbox popover, its tool catalog and custom-item
hook (no plugin used it), the generic tool dispatch (Home opens the doctor
directly), the tour step, its styles and the toolbox and old "planned
tools" strings. The scan button's tooltip no longer says "wizard".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Current node lists the selected node's prompt boxes (prompt_boxes.js, so
each is marked positive or negative by its wiring). Chinese text gets
"translate to English", written back only if the box still holds the same
text when the translation arrives (one undo); English text can show its
Chinese meaning without writing.
The standalone Prompt Translator (ui_prompt_translator.js, 838 lines, its
four fallback ways of guessing a text box, its styles and its toolbox
entry) is removed, with Prompt Studio's appendPromptToStudio that only it
used. Prompt Studio keeps its own translate buttons.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
