
Comfyui Prompt Engineer
This is your ComfyUI prompt optimization specialist. It knows the difference between how FLUX wants natural language descriptions versus SDXL needing quality tags up front, handles model-specific CFG ranges, and adjusts syntax when you're using InstantID or LoRA triggers. The real utility is in the identity method handling: it knows not to describe facial features when InstantID is providing them, and it pulls trigger words from your character profiles automatically. It covers FLUX (dev/schnell/Kontext), SDXL, SD1.5, and Wan video models with appropriate negative prompt templates for each. If you're running ComfyUI workflows with multiple models and identity methods, this keeps your prompts from fighting against how each checkpoint actually works.
- skills.sh installs
- 1.4K
- GitHub stars
- 94
- last updated
- Jul 14, 2026
Install
$ npx -y skills add mckruz/comfyui-expert --skill comfyui-prompt-engineer --agent claude-codeInstalls into .claude/skills of the current project.
Files
ComfyUI Prompt Engineer
Generates optimized prompts tailored to specific models and identity methods. Different models respond differently to prompts.
Model-Specific Prompt Rules
FLUX.1 (dev/schnell/Kontext)
- Style: Natural language descriptions work best
- CFG: 3.5-4 (very low)
- Quality tags: Minimal - FLUX doesn't need "masterpiece, best quality"
- Length: Medium (50-100 words)
- Structure:
{subject description}, {setting}, {lighting}, {camera/style}
Good FLUX prompt:
photorealistic portrait of a woman with auburn hair and green eyes, freckles across
her nose and cheeks, wearing a cream knit sweater, sitting in a cafe with warm ambient
lighting, shallow depth of field, shot on Sony A7IV, 85mm lens
Bad FLUX prompt (too many quality tags):
masterpiece, best quality, 8k uhd, highly detailed, photorealistic portrait...
SDXL (RealVisXL, Juggernaut, etc.)
Sponsors
Details
- Repository
- github.com/mckruz/comfyui-expert
- Categories
- AI & Agent Building, Generative Media
- First seen
- Jun 3, 2026