GitHub radar
Animated AI Video Skill for Agents
An open-source agent skill that turns a single sentence into a complete animated explainer video — research, narration, visual direction, voice consistency, and clip assembly included.
An open-source Agent Skill for creating 1–5 minute animated knowledge videos. Starting from a single sentence, it guides the agent through topic research, narration writing, visual style selection, multi-shot scene direction, voice reference setup, and final clip assembly. Ships with six built-in visual styles and works with Codex; the primary media execution path uses LibTV CLI with Seedance video generation. Important: the README is in English and Chinese, but the actual skill guide (SKILL.md) and all production documentation are in Chinese — this tool is designed primarily for Chinese-language video production.
Why a vibe-coder should care
If you create video content in Chinese on topics like philosophy, psychology, history, or finance, this skill handles the hardest parts: structuring narration for video, keeping a consistent visual look across clips, and maintaining the same voice throughout. The tool is built around Chinese workflows — narration pacing is calibrated in Chinese characters, voice references are Chinese voices, and the main execution guide is in Chinese.
▌ More finds