Save 4 hours! We auto-setup your workflow! Free!

Drop your workflow.json — we handle every dependency, custom node, and model. Just open the link and run.

Auto-Setup Workflow Json (Free) Now!
ComfyUI > Nodes > ComfyUI-MossPanelDataset > Panel content (scene, characters, dialogue)

ComfyUI Node: Panel content (scene, characters, dialogue)

Class Name

MossMangaPanel

Category
Manga panel dataset
Author
sepiablue-ai (Account age: 467days)
Extension
ComfyUI-MossPanelDataset
Latest Updated
2026-11-05
Github Stars
0.04K

How to Install ComfyUI-MossPanelDataset

Install this extension via the ComfyUI Manager by searching for ComfyUI-MossPanelDataset
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter ComfyUI-MossPanelDataset in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

Panel content (scene, characters, dialogue) Description

Organizes manga panel scenes, characters, dialogue, and composition into validated structured data.

Panel content (scene, characters, dialogue):

MossMangaPanel prepares the content of a single manga panel so you can combine it with other panels in a manga-page workflow. It brings together a scene description, the characters who appear, and up to three dialogue entries, with optional composition guidance for arranging characters. Rather than generating an image itself, it organizes your creative decisions into structured panel data for later processing. Its built-in checks help prevent common storytelling inconsistencies, such as assigning dialogue to a character who does not appear or choosing a composition designed for a different number of characters. This makes it easier to build clearly defined panels while keeping scene direction, dialogue, and character placement consistent.

Panel content (scene, characters, dialogue) Input Parameters:

場面

Required multiline STRING describing what happens in this panel. Write the scene in English, for example, “A manga artist draws at a desk while an editor watches.” The node removes whitespace from the beginning and end and requires the remaining description to contain text; an empty or whitespace-only description stops execution. If you connect a composition, its prompt is appended to this description after “Composition: ” rather than replacing your scene. The default is the first entry in the source’s scene preset list, SCENES[0]; its actual text is not included in the supplied context. No maximum length is specified.

登場人物

Required COMBO selecting which character slots appear in this panel. This determines the panel’s character list, which characters are allowed to speak, and how many character boxes a connected composition must contain. The supplied context confirms choices corresponding to “No characters” and “Character 1 only,” including the internal Japanese values 人物なし and 人物1のみ; the complete list is supplied through options(ACTORS) and is not shown, so additional choices cannot be listed reliably. Character slots are identifiers that can be associated with names and appearances elsewhere in the workflow, rather than names you enter here. No explicit default is declared in this node’s source, and numerical minimum and maximum values do not apply.

話者1

Required COMBO assigning the speaker for the first dialogue entry, or disabling that entry with “No dialogue” (セリフなし). A selected speaker must be included in this panel’s character selection and must have nonempty text in セリフ1; otherwise, execution stops with an explanatory error. The speaker choices come from options(SPEAKERS) and refer to character slots, with “No dialogue” also available; the complete option list is not shown in the supplied source. No explicit input default is declared. If this argument is absent when calling the underlying method directly, the method treats it as “No dialogue.”

セリフ1

Required multiline STRING containing the first dialogue entry. Its default is an empty string, and no maximum length is specified. Leave it empty when 話者1 is “No dialogue,” or enter non-whitespace text when a character is selected as the speaker. The node checks whether the text is blank but preserves the original text when storing a valid entry, including leading or trailing spaces, line breaks, punctuation, and quotation marks. Unlike the scene description, this field is not required to be English; the supplied tests include Japanese dialogue and mixed Japanese/English text.

話者2

Required COMBO assigning the speaker for the second dialogue entry, with the same choices and validation rules as 話者1. Choose a character who appears in this panel and provide text in セリフ2, or choose “No dialogue” (セリフなし) and leave that text field empty. The second entry is processed after the first, so active entries retain their numbered order in the resulting dialogue list. Choices are supplied through options(SPEAKERS), whose complete list is not included in the context. No explicit input default is declared; the underlying method uses “No dialogue” if the argument is absent.

セリフ2

Required multiline STRING containing the second dialogue entry. The default is an empty string, and no maximum length is specified. A nonempty entry requires a matching character selection in 話者2, while selecting “No dialogue” requires this field to remain empty or contain only whitespace. Valid text is saved exactly as supplied, without trimming or translation. This entry is independent of the first, so you can leave it disabled when the panel needs only one spoken line.

構図

Optional MOSS_COMPOSITION connection providing a composition template, normally from the “Choose a composition (preview images)” node. The connected data must contain a nonempty text prompt and a list of character boxes. The number of boxes must exactly match the number of characters selected for this panel. When accepted, the template’s prompt is appended to the scene description and its boxes are retained as composition guidance in the panel data. This helps coordinate scene direction with character placement without replacing your dialogue. No explicit default or numeric range is defined; leaving it disconnected produces panel data without the additional composition prompt or boxes. This node checks the presence and basic types of the prompt and box list, but does not itself validate individual box coordinates.

話者3

Optional COMBO assigning the speaker for a third dialogue entry. Its explicit default is “No dialogue” (セリフなし). It uses the same speaker choices as 話者1 and 話者2, supplied through options(SPEAKERS); the complete option list is not included in the provided context. If you enable this entry, choose a character included in 登場人物 and supply nonempty text in セリフ3. A valid third entry is placed after the other active dialogue entries. Leave the default selected when you do not need a third line.

セリフ3

Optional multiline STRING containing the third dialogue entry, with an empty-string default and no specified maximum length. Enter text only when 話者3 identifies a character who appears in the panel. If 話者3 remains “No dialogue,” this field must be empty or contain only whitespace; if you select a character, whitespace-only text is rejected. As with the other dialogue fields, valid text is preserved exactly, including spacing and line breaks.

Panel content (scene, characters, dialogue) Output Parameters:

Panel content

A MOSS_MANGA_PANEL value containing the structured description of one panel, rather than an image or a finished page. It includes scene, the trimmed scene description; actors, the selected character-slot list; and dialogue, the active dialogue entries in numbered order, each recording its speaker slot and original text. Entries set to “No dialogue” are omitted. When a composition is connected successfully, scene also includes the appended composition prompt and composition_boxes contains the template’s character boxes. You can pass this output to a panel input on MossPanelDatasetFlexible to incorporate the panel into a larger manga-page dataset. Character names and appearances are handled separately in that workflow rather than being defined by this output.

Panel content (scene, characters, dialogue) Usage Tips:

  • Write 場面 in English and focus on the panel’s action and setting. For example, “An artist pauses over a manuscript while an editor points to the page” gives a clear description without embedding dialogue in the scene.
  • Set 登場人物 before assigning speakers. Every active speaker must belong to that selection, even if the dialogue text itself mentions other characters.
  • For a silent panel, set all available speaker fields to “No dialogue” and clear their corresponding text fields. You can still include characters and describe their actions.
  • For an establishing shot with no characters, choose “No characters,” disable every dialogue entry, and describe the environment in 場面.
  • Use the dialogue fields in the order you want the entries stored. The node processes entries 1, 2, and 3 in sequence, skipping disabled entries.
  • Connect 構図 when you need template-based placement guidance, and match its character count to 登場人物. A two-character template cannot be used with a one-character panel.
  • Keep the scene description focused on what is happening when using a composition template. The node automatically appends the template’s composition prompt, so you do not need to copy it into 場面.
  • Review dialogue spacing and line breaks before continuing. The node preserves valid dialogue exactly; it does not clean up indentation, remove quotation marks, or translate the text.
  • Use separate MossMangaPanel instances for separate panels, then connect their Panel content outputs to the corresponding panel inputs of MossPanelDatasetFlexible.

Panel content (scene, characters, dialogue) Common Errors and Solutions:

"Scene" is empty. Describe the scene of this panel in English.

  • Explanation: 場面 is empty or contains only spaces or line breaks. A composition connection does not replace the requirement for a scene description.
  • Solution: Enter a meaningful English description in 場面, such as “An editor waits beside a crowded drawing desk.”

"Line 1" has text, but "Speaker 1" is "No dialogue". Choose who says line 1, or clear line 1.

  • Explanation: A dialogue field contains non-whitespace text while its speaker is set to “No dialogue.” The same message can appear for entry 2 or 3. - Solution: Select the intended character in the matching speaker field, or clear the dialogue text if that entry should be disabled.

A panel with "No characters" cannot have dialogue, but "Speaker 1" is Character 1.

  • Explanation: 登場人物 is set to “No characters,” but a speaker field selects a character. This check applies even before the node checks whether that speaker’s text is empty. The full message also asks you to choose characters or set the speaker to “No dialogue”; the entry number and character vary with your settings.
  • Solution: Add the speaking character through 登場人物, or set every speaker to “No dialogue” and clear the dialogue fields. This node does not support dialogue assigned to absent characters.

"Speaker 1" is Character 1, but this panel's characters are a different selection.

  • Explanation: The selected speaker is not included in 登場人物. The actual error identifies the selected speaker and panel character selection, and explains that a character who is not in the panel cannot speak.
  • Solution: Change 登場人物 to include that character, or assign the line to a character who already appears in the panel. If the line is unnecessary, disable the entry and clear its text.

"Speaker 1" is Character 1, but "Line 1" is empty. Enter the line, or set speaker 1 to "No dialogue".

  • Explanation: A character has been selected as the speaker, but the matching dialogue field is empty or contains only whitespace. The entry number and character name vary depending on the affected field.
  • Solution: Enter the intended dialogue in the corresponding セリフ field, or set its 話者 field to “No dialogue” when no line is needed.

Invalid composition data. Connect the output of "Choose a composition (preview images)" to "Composition".

  • Explanation: The connected composition data lacks a nonempty string prompt or a list of character boxes. This indicates that the supplied data does not have the structure expected by this node.
  • Solution: Connect the MOSS_COMPOSITION output from the composition-selection node to 構図 and select a valid template. Disconnect 構図 if you do not need template-based composition guidance.

The connected composition is for a different number of characters than this panel.

  • Explanation: The composition’s box count does not match the number of selected characters. The actual error includes the composition name, its supported character count, and this panel’s character count, followed by “Choose a composition with the same number of characters, or change the characters.”
  • Solution: Choose a composition designed for the selected number of characters, or change 登場人物 to match the template. Check this again whenever you change either the character selection or the composition.

Panel content (scene, characters, dialogue) Related Nodes

Go back to the extension to check out more related nodes.
ComfyUI-MossPanelDataset
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

Panel content (scene, characters, dialogue)