sammykumar-ComfyUI-SwissArmyKnife/web/docs/CivitMetadataHelper.md

CivitMetadataHelper

Collect sampling settings, LoRA names, and prompts into formatted text/JSON that matches CivitAI's preferred metadata schema.

Inputs

• steps, cfg, seed, high_sampler, low_sampler – capture generation settings.
• lora_high, lora_low – record LoRA names for transparency.
• positive_prompt, negative_prompt – include both prompts verbatim.

Outputs

• formatted_metadata – human-readable summary.
• json_metadata – structured payload that downstream nodes can parse.
• summary – compact single-line overview.

Usage Tips

1. Connect scheduler/LoRA/prompt values before uploading or archiving models.
2. Feed the formatted or JSON outputs into logging nodes, ShowText, or Azure uploads.

Additional Resources

• Full documentation