What problem does it solve? Creating on-topic, well-composed images for blog posts requires prompt engineering skill and manual resizing, and raw prompts sent directly to an image API produce generic results. This Skill interprets intent, builds optimized 6-component prompts, and orchestrates Gemini through an MCP server to produce hero images, inline illustrations, and OG cards. ## Core Features & Use Cases - Structured Prompt Engineering: Builds 6-component Reasoning Briefs (Subject, Action, Context, Composition, Lighting, Style) across 6 domain modes including Editorial, Product, and Infographic. - Generation and Editing: Creates new images with gemini_generate_image and refines existing ones with gemini_edit_image or multi-turn gemini_chat sessions. - Blog-Ready Output: Sets correct aspect ratios, provides ImageMagick post-processing commands for WebP/AVIF conversion, and delivers alt text plus frontmatter snippets. - Use Case: While writing a blog post about AI search trends, invoke the skill to generate a 16:9 hero image, receive the file path, alt text, and a coverImage frontmatter snippet ready to paste into the article. ## Quick Start Ask the assistant to generate a hero image for your blog post about a specific topic, and run /blog image setup first if the MCP server is not yet configured.
