@apilium/image-gen

AI image generation skill for Mayros -- generate images with OpenAI DALL-E 3 and Google Gemini.

Installation

Copy this skill directory into your Mayros skills folder:

cp -r skills/image-gen ~/.mayros/skills/image-gen

Or install via the Mayros CLI:

mayros skill install @apilium/image-gen

Configuration

Set the following environment variables:

| Variable | Description | Required | |----------|-------------|----------| | OPENAI_API_KEY | OpenAI API key (starts with sk-) | Yes | | GOOGLE_API_KEY | Google AI API key (for Gemini) | No (alternative) |

Usage

Once configured, the agent can:

• Generate images: Text-to-image with DALL-E 3 or Gemini
• Edit images: Inpainting with masks (OpenAI edits endpoint)
• Batch variations: Generate multiple style variations of a concept
• Gallery: Create HTML galleries for browsing generated images
• Prompt engineering: Iterate from draft (standard) to final (HD) quality

Example prompts

"Generate an image of a futuristic cityscape at sunset"
"Create 4 variations of a cat astronaut in different art styles"
"Edit this image to replace the background with a beach"
"Generate a logo concept for a coffee shop called 'Brew Lab'"
"Create an HD landscape image in 1792x1024"

Semantic Predicates

| Predicate | Scope | Description | |-----------|-------|-------------| | image:generated | agent | Images generated by the agent | | image:history | agent | Generation history for context enrichment |

Pricing

| Model | Quality | Size | Price | |-------|---------|------|-------| | DALL-E 3 | standard | any | $0.040/image | | DALL-E 3 | hd | any | $0.080/image |

License

MIT