How leeguooooo/image-use works
image-use is a small, single-file Python CLI and AI-agent skill for generating images through existing subscriptions. The main workflow is a user prompt and options entering the CLI, which selects a backend and saves the generated image; optional capabilities include animation, reusable/community styles, uploads, diagnostics, and self-upgrade. The README documents these capabilities, but the CLI source was not sampled, so internal calls and unshown implementation details are not mapped.
Main entry points: image-use image-use CLI · SKILL.md Agent skill
Components 10
Other components
- User
- image-use CLI
image-use - Agent skill
SKILL.md - ChatGPT web
- Codex backend
- Gemini web
- Antigravity CLI
- Style gallery
- Image tools
- GitHub releases
Connections 13
- User → image-use CLI: runs prompt
README.md - image-use CLI → ChatGPT web: drives browser
README.md - image-use CLI → Codex backend: uses backend
README.md - image-use CLI → Gemini web: drives browser
README.md - image-use CLI → Antigravity CLI: uses backend
README.md - User → Agent skill: installs skill
README.md - Agent skill → image-use CLI: invokes requests
- image-use CLI → Style gallery: searches styles
README.md - image-use CLI → Style gallery: publishes styles
README.md - image-use CLI → Style gallery: uploads results
README.md - image-use CLI → Image tools: post-processes animation
README.md - image-use CLI → GitHub releases: checks updates
README.md - image-use CLI → GitHub releases: installs upgrade
README.md