"Routes Gradio Interface launch, example app adaptation, and UI
Scanned 9/9/2026
Install to Claude Code
npx -y skills add VectorSpaceLab/AREX-Skill --skill ui --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Ui?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/vectorspacelab-arex-skill-a6c7efbc)More formats (shields.io, HTML) on the badges page.
---
name: ui
description: "Routes Gradio Interface launch, example app adaptation, and UI
troubleshooting workflows for stable_diffusion_videos."
disable-model-invocation: true
metadata:
disco-role: operating
license: Apache 2.0
---
# ui
Use this sub-skill when the task is about the `Interface` class, the image/video
Gradio tabs, or adapting the repository's app launch examples.
## What this route owns
- `Interface(pipeline, params=None)`
- `Interface.launch(...)`
- image-tab inputs that route to `generate_images(...)`
- video-tab inputs that route to `pipeline.walk(...)`
- adapting the simple app launcher into a safe, configurable script
- understanding why the experimental music-video UI is reference-only
## What this route excludes
- Low-level walk, image-generation, audio-timing, and upsampling details, which
belong to `generation`.
- Generic Gradio app design that does not use `stable_diffusion_videos`.
- Full production deployment, auth, or queue infrastructure.
## Read these bundled files
- `references/workflows.md` for launch recipes and UI input mappings.
- `references/troubleshooting.md` for Gradio, model-loading, and experimental
music-app issues.
- `scripts/launch_interface.py` for a configurable launcher with a safe dry-run
mode.
- `../../references/api-reference.md` for exact `Interface` and generation API
signatures.
- `../../references/repo-provenance.md` for source snapshot and evidence paths.
## Common tasks
### Launch the basic UI
Use `scripts/launch_interface.py --dry-run` first to check model/device choices.
Pass `--run` only when the environment is ready to download or load the model
and start a long-running Gradio process.
### Explain the UI tabs
The image tab passes prompt, batch, inference, guidance, size, upsampling, and
output-directory settings to `generate_images(...)`. The video tab parses
newline-separated prompts/seeds and passes them to `pipeline.walk(...)`.
### Adapt the music-video UI
Treat the experimental music-video app as reference material. It includes audio
slicing, timestep plotting, image preview, and a downloader example, but it also
loads GPU models and can use network access. Prefer the generation helper
scripts when you only need audio timing or a command-line recipe.
## Route decision
Choose this sub-skill when the request names `Interface`, `Gradio`, `run_app`,
`run_music_video_app`, "image tab", "video tab", or "launch the demo".
Choose `generation` when the request is primarily about `walk(...)`,
`generate_images(...)`, audio weights, frame generation, or video encoding.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!