AI product Open source
Travel Photo Abstraction is a source-available Codex skill that analyzes travel and everyday photographs as visual systems and converts their distinctive shapes, counts, positions, proportions, colors, directions, depth, and spatial rhythm into sparse editorial abstractions. It inventories observable evidence, maps retained source facts to minimal marks, and preserves relationships such as asymmetry, occlusion, count groups, negative space, horizons, focal coordinates, and directional lines. The skill generates a clean abstract panel with Codex's built-in image-generation tool, then uses a fail-closed finalizer to compose it beside the untouched original photograph and verify the result at the pixel level. The repository includes a bundled reference library, showcase outputs, installation and validation scripts, and requires Codex with skill support, image generation, Python with Pillow for deterministic composition and checks, and a photograph the user has permission to use.
1 use taken from transcripts — each links to the moment in the video.
A Codex skill that analyzes a travel photograph as a visual system and turns its measured elements into a sparse editorial panel beside the original. It inventories what is present and verifies the untouched source with a fail-close script.
1 in the library.