image-gen
by RamazanKara
Generate or edit raster images (PNG/JPG/WebP) from a text description using the local OpenAI Codex CLI's built-in image generator — photos, illustrations, icons, logos, mascots/characters, hero/OG/social images, favicons, textures, sprites, UI mockups, product shots, and transparent-background cutouts, plus edits to an existing image (recolor, restyle, add/remove/replace an element, change the lighting, remove the background, make variations). Use this WHENEVER the user wants to create, generate, make, draw, render, design, paint, or produce an image, picture, icon, logo, illustration, graphic, visual, asset, hero image, OG/social card, favicon, sprite, texture, mascot, or mockup — even when they don't say "Codex", don't name a file format, or just describe the picture they want ("I need a header image of …"). Do NOT use for: vector/SVG icon systems or logos meant to be authored as code; charts/graphs built from data; diagrams or flowcharts better expressed as Mermaid/SVG/HTML; or purely mechanical operations on an existing image file — resizing, super-resolution "upscaling", compression/optimization, or format conversion such as PNG→ICO — because this skill creates and edits image content, it does not transform files.