Nodes/Eric's Prompt Enhancers for ComfyUI/Video Prompt Expander (Advanced)
ComfyUI Node

Video Prompt Expander (Advanced)

A ComfyUI node in Eric Prompt Enhancers with 29 inputs and 7 outputs.

By EricRollei·Created 10 months ago·Updated 8 months ago· 38
Video Prompt Expander (Advanced)
  • reference_image
  • positive_prompt_1
  • positive_prompt_2
  • positive_prompt_3
  • negative_prompt
  • breakdown
  • status
  • vision_caption
basic_promptA cat playing piano in a cozy room
operation_modeexpand_from_idea
presetcinematic
detail_leveldetailed
creativity_modebalanced
reference_moderecreate_exact
shot_structure3_shot_structure
light_sourceauto
lighting_qualityauto
time_of_dayauto
shot_sizeauto
compositionauto
lensauto
camera_angleauto
camera_movementauto
color_toneauto
art_styleauto
scene_detailauto
visual_styleauto
visual_effectauto
character_emotionauto
llm_backendlm_studio
api_endpointhttp://localhost:1234/v1
positive_keywords
negative_keywords
num_variations1
save_to_filefalse
filename_basevideo_prompt_advanced
CategoryEric Prompt Enhancers

Inputs (29)

NameTypeDefaultDescription
basic_promptSTRINGA cat playing piano in a cozy roomEnter your prompt. Supports emphasis (keyword:1.5) and alternations {opt1|opt2}
operation_modeCOMBOexpand_from_ideaexpand_from_idea: Take a short concept and expand it fully refine_existing: Polish and improve an existing prompt modify_style: Change the style/aesthetic of existing prompt add_details: Add more descriptive details to existing prompt
presetCOMBOcinematic7 options: custom, cinematic, surreal, action, stylized, noir, +1
detail_levelCOMBOdetailedconcise: Brief, essential details only moderate: Good balance of detail detailed: Rich, comprehensive description exhaustive: Maximum detail for cinematic quality
creativity_modeCOMBObalancedConservative: Focused, predictable (temp 0.5) Balanced: Good variety (temp 0.7) Creative: More experimental (temp 0.85) Highly Creative: Maximum variety (temp 1.0)
reference_modeCOMBOrecreate_exactrecreate_exact: Use image as exact reference for character, costume, and setting subject_only: Keep character identity, ignore background and lighting style_only: Match aesthetic and mood, create new subject color_palette_only: Extract and apply color scheme only action_only: Use the pose/action, change everything else character_remix: Keep character, place in new scenario reimagine: Loosely inspired by image, creative reinterpretation
shot_structureCOMBO3_shot_structurecontinuous_paragraph: Single flowing description (no shot breaks) 2_shot_structure: Two distinct shots (Opening + Final Reveal) 3_shot_structure: Three shots (Setup, Development, Finale) [Recommended] 4_shot_structure: Four shots (Intro, Build, Climax, Resolution)
light_sourceCOMBOautoPrimary source of illumination in the scene
lighting_qualityCOMBOautoQuality and style of lighting
time_of_dayCOMBOauto9 options: auto, none, sunrise time, dawn time, daylight, daytime, +3
shot_sizeCOMBOauto10 options: auto, none, extreme close-up shot, close-up shot, medium close-up shot, medium shot, +4
compositionCOMBOauto9 options: auto, none, center composition, balanced composition, left-weighted composition, right-weighted composition, +3
lensCOMBOauto7 options: auto, none, wide-angle lens, medium lens, long-focus lens, telephoto lens, +1
camera_angleCOMBOauto12 options: auto, none, eye-level shot, high angle shot, low angle shot, dutch angle shot, +6
camera_movementCOMBOautoHow the camera moves through the scene (Wan 2.2 optimized)
color_toneCOMBOauto8 options: auto, none, warm colors, cool colors, saturated colors, desaturated colors, +2
art_styleCOMBOautoApply the distinctive style of famous artists or art movements
scene_detailCOMBOautoLevel of detail and complexity in the scene composition
visual_styleCOMBOauto15 options: auto, none, photorealistic, cinematic, 3D cartoon style, 2D anime style, +9
visual_effectCOMBOauto11 options: auto, none, tilt-shift photography, time-lapse, slow motion, motion blur, +5
character_emotionCOMBOauto14 options: auto, none, angry, fearful, happy, sad, +8
llm_backendCOMBOlm_studiolm_studio: Uses currently loaded model in LM Studio ollama: Uses currently loaded model in Ollama qwen3_vl: Auto-detects local Qwen3-VL model (no API server needed)
api_endpointSTRINGhttp://localhost:1234/v1lm_studio/ollama: API endpoint URL qwen3_vl: Leave default, or specify custom model path like 'local:A:\path\to\model'
positive_keywordsSTRING
negative_keywordsSTRING
num_variationsINT11–3
save_to_fileBOOLEANfalse
filename_baseSTRINGvideo_prompt_advanced
reference_imageoptIMAGEOptional: Provide an image to analyze and incorporate into the prompt using Qwen3-VL

Outputs (7)

NameTypeDescription
positive_prompt_1STRING
positive_prompt_2STRING
positive_prompt_3STRING
negative_promptSTRING
breakdownSTRING
statusSTRING
vision_captionSTRING