ChatVideo Documentation
ChatVideo is ChatArch's video workflow CLI/API package. The current focus is turning practical image-to-video workflows into reusable command designs: ordered keyframes, especially a three-image storyboard, become adjacent first/last-frame segments that are reviewed, assembled, and delivered as one video.
Site entry: https://arch.gh.wzhecnu.cn/ChatVideo/
Choose An Entry Point
-
See Implemented Commands
Start with the implemented
chatvideocommand tree to confirm that the CLI currently keeps only real tool entry points. -
Understand The Image-To-Video Model
Three ordered keyframes become adjacent first/last-frame segments and then one final video.
-
Plan First/Last-Frame Segments
Use this when the provider accepts a first frame plus a last frame to generate one video segment.
-
Separate Review From Final
Temporary review artifacts stay separate from durable final delivery and reusable docs.
Reading Map
| Question | Page |
|---|---|
| What is actually implemented? | CLI Tree |
| How do three keyframes become one video? | Workflow Blueprint |
| How is ordered image input recorded? | Workflow Blueprint |
| How are text-to-video and editing capabilities planned? | Workflow Blueprint |
| Which capabilities are still planned? | CLI Tree: Planned Boundaries |
Quick Commands
chatvideo --help
chatvideo --version
chatvideo --tree
chatvideo --tree-brief
The current CLI keeps only real tool entry points. The shared ChatStyle runtime generates both full and brief trees from the Click registry. Workflow planning belongs in the workflow blueprint, not in a chatvideo subcommand.