Appndo
by 101-skillsDesign

AI Video Generation

Generates and edits AI videos from text, images, references, video, and audio using models available through the inference.sh CLI.

What this skill does

The skill runs video models through inference.sh for text-to-video, image-to-video, and reference-to-video generation. It supports models such as Google Veo, Seedance, HappyHorse, Wan, and Grok. It can also create talking avatars and lip-synced clips, edit existing footage with natural-language instructions, upscale video, add Foley effects, and merge clips with transitions. The agent selects an appropriate model and passes the requested inputs to it through the command line.

Try it with a task

Example request to give your agent after installation.

Create a 10-second 1080p video of a golden retriever running through autumn leaves in slow motion, then add matching natural sound effects.

Before you start

Requires the inference.sh command-line tool (`belt`) and authentication through `belt login`. The reference also directs users to install the belt CLI skill. Some workflows require publicly accessible URLs for source images, audio, or video.

Original SKILL.md

Version bfd37efb · fetched Oct 19, 2026 · No open license is stated, so the original text is not reproduced here.

Related skills

  • AI Image Generationby 101-skills · Design

    Generates and edits images with dozens of AI models through the inference.sh command-line interface.

    805.1Kinstalls

  • Twitter Automationby 101-skills · Automation

    Automates X posts, media publishing, likes, reposts, direct messages, follows, profile lookups, and post retrieval or deletion.

    804.9Kinstalls

  • AI Avatar Videoby 101-skills · Design

    Creates talking-head and avatar videos from portraits paired with scripts or audio, with options for voice, language, style, and lip sync.

    804.6Kinstalls