Skills generate-image
🎨

generate-image

Safe 🌐 Network access📁 Filesystem access🔑 Env variables⚙️ External commands

Generate AI images with FLUX and Gemini

Also available from: davila7

Create high-quality images for presentations, documents, and publications using AI models. Generate original artwork or edit existing images with simple text prompts. Access FLUX and Gemini models through OpenRouter API.

Supports: Claude Codex Code(CC)
🥉 73 Bronze
1

Download the skill ZIP

2

Upload in Claude

Go to Settings → Capabilities → Skills → Upload skill

3

Toggle on and start using

Test it

Using "generate-image". Generate an image of a molecular structure with glowing bonds, scientific illustration style

Expected outcome:

  • Model: google/gemini-3-pro-image-preview
  • Image saved to: generated_image.png
  • Generation time: ~15 seconds

Using "generate-image". Edit this image: Make the sky purple with sunset colors, input: landscape.png

Expected outcome:

  • Model: google/gemini-3-pro-image-preview
  • Input image: landscape.png
  • Image saved to: generated_image.png

Security Audit

Safe
v4 • 1/17/2026

All 153 static findings are FALSE POSITIVES. The skill is a legitimate image generation tool using OpenRouter API. The static scanner flagged standard API key handling, base64 image encoding, and documentation code blocks as suspicious patterns. No malicious intent or exfiltration behavior was found. The code performs exactly as documented - generating images via API calls to https://openrouter.ai/api/v1/chat/completions.

4
Files scanned
966
Lines analyzed
4
findings
4
Total audits

Risk Factors

🌐 Network access (1)
scripts/generate_image.py:144-145
📁 Filesystem access (1)
scripts/generate_image.py:71-72
🔑 Env variables (1)
scripts/generate_image.py:22-35
⚙️ External commands (1)
SKILL.md:36-42
Audited by: claude View Audit History →

Quality Score

45
Architecture
100
Maintainability
85
Content
31
Community
100
Security
91
Spec Compliance

What You Can Build

Create figures for papers

Generate conceptual illustrations for research papers and presentations with scientific visualization style.

Produce visual assets

Create illustrations, artwork, and marketing visuals for documents, websites, and campaigns.

Design presentation graphics

Generate hero images, backgrounds, and visual aids for slides, posters, and educational materials.

Try These Prompts

Basic generation 
Generate an image of [subject] in [style] style, [additional details]
Scientific illustration 
[Scientific concept description], scientific illustration style, clean and professional, suitable for academic presentation
Image editing 
Edit this image: [specific instructions for changes to the input image]
Model selection 
Generate [description] using model black-forest-labs/flux.2-pro, output to [filename].png

Best Practices

  • Be specific in prompts - describe subjects, styles, lighting, and composition details
  • Use gemini-3-pro or flux.2-pro models for image editing tasks
  • Check OpenRouter pricing page to understand model costs before generation

Avoid

  • Do not use for technical diagrams, flowcharts, or schematics - use scientific-schematics skill instead
  • Do not share API keys in prompts or scripts - store in .env file
  • Do not expect photorealistic output from FLUX.2-flex - use Pro models for higher quality

Frequently Asked Questions

Which model should I use for image editing?
Use google/gemini-3-pro-image-preview or black-forest-labs/flux.2-pro. Both support image editing capabilities.
How do I set up my OpenRouter API key?
Create a .env file with OPENROUTER_API_KEY=your-key or set the environment variable before running.
Can I edit images with this skill?
Yes, use the --input flag to provide an existing image and describe your editing instructions in the prompt.
What image formats are supported?
Input images can be PNG, JPEG, GIF, or WebP. Generated images are saved as PNG files.
Why should I not use this for diagrams?
This skill is designed for artistic images and photos. Use the scientific-schematics skill for flowcharts and diagrams.
How long does image generation take?
Generation typically takes 5-30 seconds depending on the model selected and current API load.

Developer Details

Author

K-Dense-AI

License

MIT license

Repository

https://github.com/K-Dense-AI/claude-scientific-skills/tree/main/scientific-skills/generate-image

Ref

main

File structure

📁 scripts/

📄 generate_image.py

📄 evaluation.json

📄 SKILL.md