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 > MOSS-Image Character Sheet Designer > MOSS-Image Character Sheet Designer v1.0

ComfyUI Node: MOSS-Image Character Sheet Designer v1.0

Class Name

MOSSImageCharacterSheetDesigner

Category
MOSS-Image/Character Sheet
Author
ukr8b3g-cmyk (Account age: 400days)
Extension
MOSS-Image Character Sheet Designer
Latest Updated
2026-11-01
Github Stars
0.01K

How to Install MOSS-Image Character Sheet Designer

Install this extension via the ComfyUI Manager by searching for MOSS-Image Character Sheet Designer
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter MOSS-Image Character Sheet Designer 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

MOSS-Image Character Sheet Designer v1.0 Description

Plans character-sheet prompts and canvas layouts for single-view or multi-panel image generation.

MOSS-Image Character Sheet Designer v1.0:

MOSSImageCharacterSheetDesigner helps you plan illustrated character sheets for a ComfyUI image-generation workflow. It combines a character description with selected views and sheet settings to prepare a coordinated composition, supporting a single-view presentation or a multi-panel layout. This is useful when you want to organize different views of the same character rather than write every panel description and layout instruction yourself. For sheets that include clothing views or at least three face views, it also performs an automatic appearance-analysis step to help tailor the descriptions to those views. The node is a preparation tool, not an image renderer: it structures the instructions and canvas layout that other nodes use to generate the final artwork. Its main benefit is making character-sheet planning more consistent and manageable, although the final visual consistency still depends on the image-generation model and the rest of your workflow.

MOSS-Image Character Sheet Designer v1.0 Input Parameters:

prompt

A required STRING containing the description of your character. This is a connection-only input, so you supply it from another node that produces text rather than through an ordinary text box on this node. The description becomes the character identity used when compiling the sheet; it replaces any character identity and separate outfit information already present in the sheet settings, with the separate outfit field reset to an empty string. For that reason, include important clothing details in this description along with recognizable features such as hairstyle, facial characteristics, and accessories. For example, you could supply “An adult explorer with short silver hair, round glasses, a navy field jacket, tan trousers, and brown boots.” When the selected views require appearance analysis, this same description is also analyzed before the sheet is composed. The text is validated before compilation, but the supplied source does not reveal the validator’s exact length or content restrictions. Default: none declared. Minimum, maximum, and preset options: not specified in the exposed interface.

state_json

A required STRING containing the structured sheet configuration in JSON format. It supplies the settings the compiler uses to organize the selected views and resolve the sheet layout and canvas dimensions. The provided code also shows that the state contains character and negative-prompt information, although the character information is replaced by the connected description during execution and the negative prompt is not exposed as a node output. Your view selections affect the processing path: any view identifier beginning with clothing_, or at least three view identifiers beginning with face_, enables automatic appearance analysis. The configuration is parsed and validated against the node’s state rules and ComfyUI’s MAX_RESOLUTION before compilation. Default: a compact JSON serialization of the package’s DEFAULT configuration; the actual JSON contents are not included in the supplied source. The interface provides a multiline text field with dynamic prompt expansion disabled, so the JSON is handled as configuration text rather than expanded as a dynamic prompt. Minimum, maximum, and selectable configuration options: not fully specified in the provided excerpts; use the supplied default or settings produced by the designer interface instead of inventing field names or values.

MOSS-Image Character Sheet Designer v1.0 Output Parameters:

prompt

A STRING containing the compiled character-sheet instructions for your positive text encoder. This is not simply a copy of your original character description: the compiler builds a sheet description using the character information, selected views, and resolved layout. The supporting compiler code constructs a layout draft with a canvas size, shared caption, and individual items, then renders that draft into the final prompt. If appearance analysis is required, the internal composition step applies the analyzed features before returning the result. Connect this output directly to the positive CLIPTextEncode node intended for your MOSS workflow. A direct connection is especially important for analysis-enabled sheets because the designer inspects that downstream encoder connection to locate the text encoder it must use.

width

An INT giving the resolved horizontal canvas size in pixels. This value is calculated from the sheet configuration rather than entered as a separate input on this node. Connect it to the width input of EmptyFlux2LatentImage, as recommended by the node description, so the generated canvas matches the dimensions used to prepare the sheet layout. If you use a different width elsewhere in the workflow, the composition instructions and actual image dimensions may no longer agree. The supplied excerpts do not specify a fixed default width, numerical minimum, or numerical maximum; state validation uses ComfyUI’s MAX_RESOLUTION.

height

An INT giving the resolved vertical canvas size in pixels. Use it together with the width output to create the canvas that corresponds to the compiled sheet layout. Connect it to the height input of EmptyFlux2LatentImage so the intended panel arrangement and canvas proportions remain aligned. This value describes the overall sheet height, not the height of an individual character or panel. As with width, it is derived from the validated configuration, and the supplied excerpts do not provide a fixed default or explicit numerical range.

MOSS-Image Character Sheet Designer v1.0 Usage Tips:

  • Write one clear character description that includes the visual details you want preserved across the sheet. Include outfit information there rather than relying on character or outfit values stored in state_json, because execution replaces those stored character fields.
  • Connect the compiled prompt directly to your positive MOSS CLIPTextEncode node. For sheets requiring appearance analysis, do not insert text-processing nodes between them: the connection lookup only recognizes a direct link to a node whose class is CLIPTextEncode.
  • Connect both dimension outputs to EmptyFlux2LatentImage. Keeping the generated canvas synchronized with the compiled dimensions helps avoid a mismatch between the planned sheet layout and the actual image proportions.
  • Keep your negative prompt in a separate CLIPTextEncode, as the node description recommends. This node exposes only the positive sheet prompt and canvas dimensions; it does not provide a negative-prompt output.
  • Start with the supplied default configuration or settings generated by the designer interface. If you edit state_json manually, preserve valid JSON syntax and existing field structures, because the node validates the configuration before producing results.
  • For a simple workflow without the additional analysis stage, choose no clothing views and fewer than three face views. Any clothing view or three or more face views activates the internal TextGenerate step, so those configurations require a working analysis-capable setup.
  • When analysis is enabled, make sure all directly connected positive CLIPTextEncode nodes reference the same encoder source. The code requires exactly one distinct encoder connection; multiple direct text encoders with different encoder sources will cause an error.
  • If an analysis-enabled workflow rebuilds its internal expansion on execution, that behavior is intentional. The node explicitly requests that rebuild to check its downstream encoder connection, while ComfyUI can still cache the internal TextGenerate node.

MOSS-Image Character Sheet Designer v1.0 Common Errors and Solutions:

E_ANALYSIS_ENCODER: Connect this prompt output directly to one MOSS CLIPTextEncode encoder. / prompt出力をMOSSのCLIPTextEncodeへ直接接続してください。

  • Explanation: This is an explicit error from the supplied source. When appearance analysis is needed, the node searches for a direct connection from its prompt output to a CLIPTextEncode node and collects the encoder connections used there. It fails unless it finds exactly one distinct encoder source. A missing connection, an intervening node, a different encoder node class, or multiple distinct encoder sources can trigger this error.
  • Solution: Connect the designer’s prompt output directly to the text input of the positive MOSS CLIPTextEncode node, and make sure that node has its encoder input connected. Remove conflicting direct connections to encoders using different sources. If you need to process the compiled text further, first establish a workflow that satisfies this direct-connection requirement.

State validation error in state_json — exact message depends on the validator

  • Explanation: The node parses and validates the sheet configuration before execution. Invalid JSON or configuration values that do not satisfy the state rules can stop compilation. The supplied excerpts do not include the individual validation messages or the complete configuration schema.
  • Solution: Restore the supplied default configuration or regenerate the settings through the designer interface. If you edited the JSON manually, check quotation marks, commas, and brackets, then undo unsupported changes. Follow any field name identified in the actual error message rather than guessing a replacement structure.

Prompt validation error — exact message depends on the validator

  • Explanation: The character description is checked by validate_prompt before the sheet is compiled. The validator’s implementation is not included, so its exact accepted lengths, rejected content, and error wording cannot be documented from the supplied code.
  • Solution: Confirm that the required prompt input is connected to a node that returns a STRING. Use a straightforward character description and follow any specific restriction reported in the actual validation message. If necessary, test with a short description to separate a text-validation problem from a sheet-configuration problem.

Resolution validation error — exact message depends on the state settings

  • Explanation: The configuration is parsed with ComfyUI’s MAX_RESOLUTION limit. Resolution settings that the state parser does not accept can prevent the node from compiling. The provided code does not show the numerical limit or the exact resolution-related validation messages.
  • Solution: Return to the default canvas settings or select smaller supported dimensions through the designer interface. Do not assume an arbitrary resolution is valid simply because another node accepts it; use dimensions accepted by this node’s configuration validator.

TextGenerate unavailable or unable to execute — descriptive troubleshooting label

  • Explanation: Clothing views and configurations containing at least three face views cause the node to create an internal TextGenerate node. If that node type is unavailable in your installation, or the connected encoder cannot support the requested operation, the analysis branch may fail. The exact error text comes from the surrounding ComfyUI setup and is not provided in these excerpts.
  • Solution: Check the execution error for TextGenerate or encoder-related details, and confirm that your MOSS workflow includes the required text-generation support. As a diagnostic test, select no clothing views and fewer than three face views; this avoids the analysis branch and helps identify whether the problem is specific to that stage.

Appearance-analysis validation error in appearance_analysis — exact message depends on the returned data

  • Explanation: The internal analysis result must be parseable structured data with the expected fields. The provided parser checks the response text against a 12,000-character limit and each expected field against a 1,500-character limit. Invalid structure, unexpected fields, or values that fail those checks can prevent the analyzed sheet from being composed; the exact validation messages are not shown.
  • Solution: Confirm that the workflow uses the intended MOSS encoder and compatible TextGenerate implementation. Simplify the character description if it is unusually long or complicated, then retry. To isolate the issue, test a sheet without clothing views and with fewer than three face views; if that succeeds, the failure is associated with the appearance-analysis path rather than basic sheet compilation.

MOSS-Image Character Sheet Designer v1.0 Related Nodes

Go back to the extension to check out more related nodes.
MOSS-Image Character Sheet Designer
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.

MOSS-Image Character Sheet Designer v1.0