What problem does it solve? When asked to create an image, a single generation attempt often misses the mark. This Skill mobilizes every available image source—Codex CLI, Gemini API, HTML/CSS rendering, personal photo libraries, and existing brand assets—to produce multiple candidates, compare them, and recommend the best fit for the intended use. ## Core Features & Use Cases - Multi-source generation: Invokes Codex image generation, the Gemini API via the google-genai SDK, HTML/CSS screenshot rendering, iCloud photo libraries, and existing website assets in one workflow. - Side-by-side comparison: Displays all candidates together and evaluates each against the actual use case (standalone visual, text-overlay banner, data visualization, or real-person photo). - Brand consistency: Enforces brand color codes and restrained composition in prompts so generated images match the brand mood. - Use Case: You need a banner image for a blog post. The Skill generates candidates via Codex and Gemini with matched brand-tone prompts, renders an HTML/CSS stat card as a third option, shows all three, and recommends the one with enough whitespace for overlay text. ## Quick Start Ask the agent to use every available method to create an image for your blog banner and compare the results before picking one.
