What problem does it solve? It lets you generate and edit images through ChatGPT's Codex backend API without manually managing API keys, since the session token is fetched automatically from your browser login. ## Core Features & Use Cases - Text-to-Image Generation: Create images from text prompts via the gpt-image-2 tool on the Codex responses endpoint. - Image-to-Image Editing: Transform a local reference image into a new image using multimodal input (input_text + input_image). - Automatic Authentication: Tokens are resolved from cache, an existing browser session, or an interactive login flow, and are never printed to stdout. - Use Case: Ask for a cinematic portrait based on your photo, and the skill encodes the reference image, calls the Codex API, and saves the result as a PNG. ## Quick Start Generate an image of a cute rabbit on a light background using the codex image skill and save it as rabbit.png.