MOSS-Image Character Sheet Designer is a community ComfyUI node that generates character reference sheets from text descriptions, supporting up to 13 views, with illustrated view selection, part-specific prompts, styles, and automatic/manual layout options. The 1.0.1 version introduces editable view prompts: the base instruction for each view can be modified and reset, and these changes are saved with the workflow. Image input is also supported, but images are first converted into text descriptions by the MOSS text encoder before generation; original pixels are not preserved, and character consistency is not guaranteed.
Why It Matters Saves you from manually writing prompts for each view when creating multi-view character reference sheets; layout and part descriptions can be adjusted directly within the node.
GitHub Stars 5 · Registry Downloads 14
License: Commercial Use Allowed · ComfyUI: Community Node · Scan Passed · VRAM: Not Specified
Node Audit: N1 Passed · N2 Passed · N3 Pending Manual Review (Version 1.0.1)
Rationale
- Both the Comfy Registry automated scan and static code scan have passed, and the license permits commercial use
Usage
- Download a single workflow JSON from the author's GitHub Releases, and describe only one character's appearance and clothing in the Character Profile
Risks
- Manual audit (N3) is not yet complete, so installation commands are not currently displayed; Registry download counts are still very low
- Image input is not reference image editing; facial features, clothing, and colors may change