genimg¶
One simple CLI, made with both humans and agents in mind. The agent-native design is inspired by kenn-io and the tools they put out, especially roborev and agentsview.
Generate images with OpenAI and Google DeepMind models in a unified interface, via API or via Codex with a ChatGPT subscription. It bundles an image review tool and a drawing studio app for image generation, as well as several skills for productive image generation workflows.
One prompt, four takes¶
genimg "a minimal fox logo, NOT a grid" \
--model oai:gi2.5 \
--num 4 \
--deltas "line art, block print, brush stroke" \
--output fox.png \
--grid
Watch it run
A real run. The 30-second wait plays at 6x speed.

Where next¶
- Getting started: install, connect a provider, make a first image.
- For agents: add the genimg skills to your coding agent.
- Diverse images: get takes that really differ.
- Grid and carousel and Draw Studio: review results and sketch prompts.