blog-image

This skill generates and edits images for blog content using Google's Gemini image models accessed through an MCP server. It solves the problem of sourcing on-brand visuals for articles by producing hero images, inline illustrations, social preview cards, and Open Graph images from descriptive prompts, and by editing existing images. It can run standalone or be invoked internally by companion blog-writing skills, and it falls back gracefully when the MCP integration is unavailable.

The skill supports six domain modes (Editorial, Product, Landscape, UI/Web, Infographic, and Abstract) that map blog use cases to appropriate aspect ratios and prompt focus. Its bundled prompt-engineering reference teaches a six-component reasoning brief (subject, action, context, composition, lighting, and style) written as natural narrative paragraphs rather than keyword lists, with good-versus-bad examples for each component. Additional references document the available Gemini models and MCP tools, and setup and validation scripts help configure and verify the image MCP connection.

Target users are bloggers, content teams, and technical writers who need illustrations without leaving their authoring workflow. Typical use cases include generating a 16:9 hero image for a new post, creating a social share card, producing an infographic-style diagram, or editing a previously generated illustration. The skill depends on an external Gemini MCP being configured; its included Python scripts are for setup and validation of that integration.