AnimaV3PromptAssemble:
AnimaV3PromptAssemble combines generated prompts with your handwritten additions, helping you refine an AI artwork’s description without rebuilding it from scratch. You can simply append extra wording, merge comma-separated tags while removing duplicates and resolving recognized conflicts, or preserve the main prompt without adding anything. The merge method keeps tags in their original order of first appearance, normalizes them to lowercase, and uses the plugin’s conflict-resolution rules to reduce incompatible combinations. It can also remove tags above a selected content level when merging. This is a text-processing tool, not an AI prompt writer or a complete prompt checker: it does not generate new descriptions or comprehensively diagnose prompt problems. In the supplied V4 plugin, AnimaV3PromptAssemble is a backward-compatible alias of AnimaV4PromptAssemble, so older workflows use the same implementation.
AnimaV3PromptAssemble Input Parameters:
main
A required STRING containing your primary prompt, typically the generated description or existing text you want to keep as the foundation. It supports multiline text and defaults to an empty string; no minimum or maximum length is specified. Before processing, the node removes surrounding whitespace and trailing commas. In merge mode, it reads this text first, retaining the order of the first occurrence of each tag before applying conflict resolution and any selected level filter. In append mode, the primary text remains before the supplement. In pass-through mode, this is the only text returned, although the initial whitespace and trailing-comma cleanup still occurs.
supplement
A required STRING containing your additions, such as clothing details, background tags, lighting choices, or artist tags. It supports multiline text and defaults to an empty string; no minimum or maximum length is specified. “Required” means the parameter must exist, not that you must enter text. The node removes surrounding whitespace and trailing commas before using it. In append mode, it is placed after the main text, separated by a comma and a space. In merge mode, its comma-separated tags are processed after those from the main text, with duplicate tags removed and recognized conflicts resolved. If this field is empty after cleanup, the node immediately returns the cleaned main text without deduplication, conflict resolution, or level filtering.
mode
A required selection controlling how the two text fields are handled. Its exact options, in interface order, are 追加到末尾 (“Append to end”), 合并去重 (“Merge and deduplicate”), and 仅透传 main (“Pass through main only”). The source does not explicitly declare a default; 追加到末尾 is the first listed option. Append mode joins the cleaned text fields without removing duplicate tags, resolving conflicts, changing capitalization, or applying the content-level limit. Merge mode splits both fields on commas, trims each tag, converts it to lowercase, removes empty and repeated tags, applies recognized conflict-resolution rules, and then applies the optional level filter. Pass-through mode ignores the supplement and returns only the cleaned main text. These are discrete choices, so numerical minimum and maximum values do not apply.
max_level
An optional selection specifying the maximum content level allowed during merge processing. Its default is 不限 (“No limit”), and its declared options are 不限 followed by the plugin’s LEVEL_CHOICES list. The supplied source does not show the full contents of that list, so its exact labels and level range cannot be enumerated here. When you select a limit, the node checks each remaining tag using the plugin’s calibrated tag-level classification and removes tags above the selected level; it does not rewrite them into alternatives. This filtering happens after conflict resolution and applies to tags from both text fields. It is skipped in append mode, pass-through mode, and whenever the supplement is empty after cleanup. Treat it as a tag-level filter rather than a guarantee about the content of a generated image.
AnimaV3PromptAssemble Output Parameters:
text
A STRING containing the resulting prompt, ready to connect to another text-processing node or a compatible prompt encoder. Its contents depend on the selected mode: cleaned main text alone, the two cleaned text fields joined together, or a lowercase, deduplicated, conflict-resolved tag list with optional level filtering. In merge mode, surviving tags are separated by a comma and a space. The result can be empty if no usable text remains, including when filtering removes every remaining tag. This output contains only prompt text: it does not include a report identifying removed duplicates, resolved conflicts, or filtered tags, and it does not certify that the prompt meets a particular content rating.
AnimaV3PromptAssemble Usage Tips:
- Choose
追加到末尾when you want a straightforward addition without changing tag capitalization or removing repeated wording. For example,portrait, soft lightplusgardenbecomesportrait, soft light, garden. - Choose
合并去重when combining tag lists that may overlap. For example,Portrait, soft lightplusportrait, gardenbecomesportrait, soft light, garden, assuming those tags are unaffected by conflict rules or filtering. - Separate individual tags with commas when merging. Line breaks alone are not tag separators, so multiline text still needs commas between the tags you want processed individually.
- Use
仅透传 mainto temporarily bypass your additions while comparing results. Remember that surrounding whitespace and trailing commas are still removed. - To apply
max_level, select合并去重and provide a nonempty supplement. If you need to check or repair a standalone prompt without adding text, use the plugin’s dedicated prompt-checking node instead. - Inspect the assembled text when working with potentially conflicting details. Merge mode can remove tags according to the plugin’s rules, and the node does not explain which tags were removed or why.
- Keep sentences, capitalization-sensitive names, or other text that must remain unchanged in append or pass-through mode. Merge mode lowercases every comma-separated item rather than preserving its original capitalization.
AnimaV3PromptAssemble Common Errors and Solutions:
Node type “ AnimaV3PromptAssemble” is missing
- Explanation: This is a possible workflow-loading issue, not an error raised by the assembly method. The supplied V4 plugin registers the V3 node name as an alias, so a missing-node warning suggests that the plugin is unavailable, failed to load, or differs from the supplied implementation.
- Solution: Confirm that Comfyui-Anima-Wardrobe-V4 is installed and enabled, restart ComfyUI, and check its startup log for plugin-loading failures. The supplied version supports the V3 node name without requiring you to replace it manually.
The selected max_level does not remove any tags
- Explanation: This is a behavior symptom, not a node-generated error message. Level filtering runs only in merge mode with a nonempty supplement. It is bypassed in append mode, pass-through mode, and the empty-supplement shortcut;
不限also disables filtering. - Solution: Select
合并去重, choose a level other than不限, and check that the supplement contains meaningful text. Use the dedicated prompt-checking node if you need standalone prompt checking or repair.
Duplicate tags remain in the output
- Explanation: Append and pass-through modes do not remove duplicates. Merge mode also does not run when the supplement is empty. During merging, duplicate detection compares trimmed, lowercase text, so differently written phrases are not necessarily treated as the same tag.
- Solution: Use
合并去重with a nonempty supplement, separate tags with commas, and standardize wording where two phrases are intended to represent the same detail.
Capitalization changed or tags disappeared
- Explanation: This is expected behavior in merge mode, not an execution error. Tags are converted to lowercase, duplicates are removed, recognized conflicts are resolved, and tags above the selected level may be filtered out. The node produces no removal report.
- Solution: Compare the merged result with append mode to identify changes. Select
不限to disable level filtering, or use append mode when preserving the original wording matters more than deduplication and conflict handling.
The output is empty
- Explanation: Empty output is a valid result rather than a dedicated error. It can occur when the main text is empty in pass-through mode, both text fields are empty after cleanup, or merge processing leaves no surviving tags.
- Solution: Check the incoming text and selected mode first. If you are merging populated fields, temporarily select
不限and inspect the result to determine whether the level filter removed the remaining tags.
Conflicting details remain after assembly
- Explanation: Append and pass-through modes do not resolve conflicts. Merge mode applies only the plugin’s recognized conflict rules, not a general understanding of every possible contradiction in natural-language descriptions.
- Solution: Use
合并去重with comma-separated tags, review incompatible details manually, and use the dedicated prompt-checking node when you need broader diagnostics rather than assembly alone.
