AI Image Prompt Doctor: Lighting, Lens and Fixes
Diagnose why AI images come out generic or wrong and rewrite the prompt with named light, lens, framing and model-correct syntax.
Before & after
Same prompt, same agent — the only difference is the skill.
You'd want to describe the furniture piece clearly, including its style and color. Then add details about the setting and lighting. For example, you could describe the room it's in, what time of day it looks like, and what mood you want. Including details about the camera angle and any props around the furniture would help too. You might mention the photography style you're going for, like professional product photography or lifestyle shot. The more specific you are about these details, the better the AI can generate what you want.
PROMPT SET: LUXURY FURNITURE PRODUCT PHOTOGRAPHY SHOT 1: PRODUCT HERO (mid-century sofa) STYLE: Minimalist luxury, ultra-sharp COMPOSITION: 3/4 profile angle, sofa centered in frame, 400mm lens equivalent LIGHTING: Key light 45 degrees left, diffused softbox; fill light 50% fill ratio TEMPERATURE: Warm tungsten 3200k, high contrast MATERIALS EMPHASIS: Close focus on linen texture, individual thread definition BACKGROUND: Seamless white gradient, blurred depth PROP: Italian marble side table (corner), sculptural object on surface CAMERA SETTINGS: F2.8 depth of field, sharp focus on fabric, soft bokeh behind SHOT 2: LIFESTYLE CONTEXT (sofa in room) STYLE: Editorial fashion magazine spread, warm modernism COMPOSITION: Wide 35mm equivalent, sofa on left third, natural window light on right LIGHTING: Golden hour window light (5pm, LA apartment, west-facing) TEMPERATURE: Warm 4500k, cinematic grade STYLING: Live plant (fiddle leaf fig), throw blanket (cream cashmere), books stacked BACKGROUND: Mid-century wooden bookshelf (soft focus), tall ceiling visible MOOD: Aspirational, lived-in luxury, approachable COLOR GRADING: Crushed blacks, lifted shadows, warm shadow tone SHOT 3: DETAIL CLOSE-UP (fabric swatch) STYLE: Macro product photography, tactile COMPOSITION: Extreme close-up, fabric fills frame LIGHTING: Raking light 80 degrees (shows weave texture) DEPTH: Single plane sharp, thread definition visible MATERIAL HERO: Show linen weave, natural nubs, undyed fibers REUSABLE ELEMENTS Brand accent color: Warm brass hardware visible in all shots Lighting ratio: 3:1 (key to fill) across series Tone match: Warm, high-end, tactile
About this skill
Stop rerolling. Paste the prompt that keeps failing, or a rough idea, and get the likely cause, a rewrite in your model's syntax, and a short test plan that changes one thing at a time.
What it does
- Diagnoses generic, flat, crowded or garbled results from the prompt and a description of the output
- Rewrites with named light sources, lens, shot size and subject placement that models actually follow
- Uses the right syntax per model: Midjourney parameters, SD weights and negative field, natural-language models
- Removes quality boosters, conflicting style words and negations that summon what you excluded
- Gives a 3-4 variant test plan so the next round converges
What you get
A diagnosis, the rewritten prompt in a copy-ready block, a what-changed list, and a test plan.
Best for
Creators and marketers who generate images regularly and are tired of the stock-photo look.
SKILL.md preview
---
name: ai-image-prompts
description: Diagnose why an AI image keeps coming out wrong or generic and rewrite the prompt with explicit lighting, lens, composition, and model-correct syntax, plus a one-variable test plan so the next round converges. Covers Midjourney, Stable Diffusion and SDXL, Flux, GPT image, Imagen, and Ideogram. Use when the user says "my AI images all look generic", "fix this Midjourney prompt", "why does it keep adding people", "the lighting is flat", "make this prompt more cinematic", "the negative prompt isn't working", or "turn this rough idea into a proper image prompt".
---
# AI Image Prompt Doctor
… (sign up to view the full skill)More video & editing skills
View all Video & Editing skills →Character and Location Reference Sheets for AI Video
Locks faces, wardrobe and sets before AI video: identity blocks, 3-view turnarounds, empty location plates, and reference role lines.
AI Image Realism Pass: Fix Plastic Skin and Flat Depth
Diagnoses why AI characters look plastic or pasted-on and rewrites prompts with gray reference sheets, lit haze, real optics, and depth.
AI Visual Style Reference: Style Card and Drift Test
Turns a look into a reusable AI style reference: measured style card, paste-in style block, reference-image plan, and a drift test.
Cinematic AI Image Prompts and Character Lock
Paste-ready photoreal prompts for Nano Banana Pro, GPT Image, and Higgsfield Soul, plus character sheets and face-locked outfit swaps.