generate-image

The generate-image skill provides a powerful interface for creating and modifying high-quality visual content using state-of-the-art AI models, including Google's Gemini 3 Pro and Black Forest Labs' FLUX.2. Designed for researchers, academics, and creators, this tool integrates directly into your scientific workflow, allowing for the rapid generation of photorealistic images, artistic illustrations, and concept art to enhance papers, presentations, posters, and general documentation. By utilizing OpenRouter's API, the tool ensures high-performance model access with flexible options for quality, speed, and cost-effectiveness.

• Support for advanced image generation and editing using models like Gemini 3 Pro and FLUX.2 Pro.

• Ability to generate new imagery from text prompts or perform specific edits on existing local images.

• Configurable model selection to balance aesthetic quality with computational costs, including fast modes for quick drafts.

• Seamless integration with existing file structures, enabling output to custom paths and support for batch generation patterns.

• Professional-grade visual asset creation for scientific figures, conceptual art, and presentation backgrounds.

• Requires an OpenRouter API key configured via a .env file or environment variable to function correctly.

• Use this skill for general-purpose visual content; for technical schematics, flowcharts, pathways, or circuit diagrams, please utilize the scientific-schematics skill instead.

• Always verify generated content for accuracy and relevance; ensures compliance with research standards by excluding AI meta-instructions from final image outputs.

• Standard inputs include descriptive text prompts and optional input file paths for editing; outputs are saved as local PNG files by default.

• Users should ensure appropriate attribution and ethical consideration when using AI-generated visual content in formal academic publications.