egaki

Every time you work with egaki, you MUST fetch the latest README:

curl -s https://raw.githubusercontent.com/remorses/egaki/main/README.md # NEVER pipe to head/tail, read the full output

The README is the single source of truth for all user-facing docs: CLI usage, MDX video syntax, animation primitives, easing presets, LayoutTransition, captions, voice cloning, media components, CSS rules, and project setup.

Always check help first

egaki --help # NEVER pipe to head/tail, read the full output

For subcommand details: egaki <command> --help (e.g. egaki image --help, egaki video --help, egaki login --help)

Auth options

Two authentication modes:

1. Egaki subscription key (recommended, all models, one key)
2. Provider API keys (Google, OpenAI, Fal, xAI, etc.) via egaki login

If using Egaki subscription, set it up with egaki subscribe, then store the key with egaki login --provider egaki --key egaki_....

Login behavior for remote agents

When login requires a URL flow, run login in the background and send the login URL to the user so they can complete auth interactively.

Model selection

The --model / -m flag is optional on both egaki image and egaki video.

• Interactive (TTY): omitting --model shows a picker
• Non-interactive (piped/scripted): uses a sensible default

Agents should always pass -m explicitly to avoid the interactive picker.

Preferred image models

1. gpt-image-1 (or gpt-image-2 / chatgpt-image-latest if available)
2. grok-imagine-image (xAI Grok)
3. nano-banana-pro-preview

Preferred video models

1. grok-imagine-video-1.5 (xAI Grok)

Extracting small assets from a reference

Crop to the target object first with an image tool, leaving a small margin, then pass that crop to egaki image --input. This makes small details easier for the model to see and reduces ambiguity about which object to preserve. In a switch-extraction comparison, a tight crop preserved the design much better than the full instrument image with the same prompt.

Keep the prompt short: "Extract this switch exactly as it is, with a transparent background. Do not redesign it or change its shape, materials, or details."

Inspect the result: generation can still redraw details. Use a direct crop and background masking when exact original pixels are required.

Video generation note

Video generation can be very slow (1-3 minutes per request). Always use a command timeout of at least 5 minutes from automation or agent workflows.