>>27400I use an amalgamation of things I've seen in positive and negative prompts of images that I like, so this is in no way the "correct" way to do things but here it is:
parameters
score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up,
<style>, 1girl, solo, asian, pale skin, pretty face, chubby, curvy, looking at viewer, all fours, from the front, crawling, inside military facility, stuck, ass squeezed inside small opening, (fat hips stuck), struggling, straining, exerting, grunting, (huge ass, wide hips, fat hips, hyper ass, ass expansion, fat ass:1.2) spy jumpsuit outfit, ripped clothes, thick thighs, action shot
Negative prompt: censored, blindfold, furry, bad anatomy, bad, sketch, low quality, lowres, text, watermark, patreon, heterochromia,
Steps: 35, Sampler: Euler a, Schedule type: Automatic, CFG scale: 7, Seed: 2474606618, Face restoration: CodeFormer, Size: 1152x864, Model hash: 059934ff58, Model: ponyRealism_v21VAE, VAE hash: 235745af8d, VAE: sdxl_vae.safetensors, Clip skip: 2, ADetailer model: face_yolov8n.pt, ADetailer confidence: 0.3, ADetailer dilate erode: 4, ADetailer mask blur: 4, ADetailer denoising strength: 0.4, ADetailer inpaint only masked: True, ADetailer inpaint padding: 32, ADetailer version: 24.6.0, Version: 1.9.4
If you're using automatic1111 you can also download these images and drop them into the PNG Info tab to extract this information, then send it to text2img or img2img
