image-gen

Create precise, safe image-generation prompts and iterate against visual requirements.

SKILL.md

GitHub

Image Gen

Purpose

Create precise, effective prompts for AI image generation models and iterate against visual requirements. The skill ensures generated images match user intent through structured prompt refinement.

When to Use

  • The user wants to create AI-generated imagery
  • Visual assets are needed for projects
  • Prototyping visual concepts

When NOT to Use

  • For photography that requires real-world capture
  • When stock imagery would be more appropriate
  • For content requiring precise technical accuracy

Capabilities

  • Prompt writing and refinement
  • Style specification (photorealistic, illustrative, abstract, etc.)
  • Aspect ratio and resolution selection
  • Negative prompt crafting
  • Iteration based on feedback
  • Multiple model support (DALL-E, Midjourney, Stable Diffusion, etc.)

Inputs

  • description (required) — what the image should depict
  • style (optional) — artistic style preference
  • aspect_ratio (optional) — 1:1, 16:9, 9:16, etc.
  • mood (optional) — tone or atmosphere
  • reference (optional) — reference image URL or upload
  • model (optional) — preferred generation model

Workflow

  1. Analyze — Understand user requirements
  2. Draft — Create initial prompt
  3. Refine — Add style, quality, and specificity modifiers
  4. Generate — Submit to image generation API
  5. Iterate — Refine based on user feedback
  6. Deliver — Provide image file and prompt used

Tools

  • Image generation APIs (DALL-E, Midjourney, Stable Diffusion, etc.)
  • Image editing tools for post-processing

Examples

User: "Create a logo for my tech startup" Response:

Prompt: "Minimalist tech startup logo, geometric fox silhouette, 
gradient blue to purple, clean lines, modern, professional, white 
background, vector style, scalable"

Generating... [██████████] 100%
Output: logo-startup-2026-02-09.png (1024x1024)
Prompt saved for reproducibility.

Safety

  • Never generate harmful, violent, or deceptive imagery
  • Respect copyright and trademark considerations
  • Avoid generating content that could be mistaken for real people
  • Label AI-generated images appropriately

Source

Auto-generated from openclawskills.net description.

Notes

Maintained as part of canonical design category.