HeyGen Skills (v1
April 14, 2026 · View on GitHub
Looking for the current HeyGen Skills? Go to heygen-com/skills.
git clone https://github.com/heygen-com/skills.git
This repository preserves the original v1 HeyGen skills for reference. It is no longer maintained.
The current HeyGen Skills replaces this repo with two purpose-built skills — heygen-avatar and heygen-video — powered by the v3 API with best practices from 80+ production test videos.
What Was Here
These v1 skills were individual API wrappers — each mapped to a HeyGen endpoint or local tool:
| Skill | What it did |
|---|---|
| create-video | Prompt-based video creation via Video Agent |
| avatar-video | Precise avatar/scene video control |
| text-to-speech | Standalone TTS audio generation |
| video-translate | Video translation and dubbing |
| ai-video-gen | AI video generation (VEO, Kling, Sora, Runway) |
| faceswap | Face swap in videos |
| visual-style | Portable visual design systems |
| video-download | Download video/audio from URLs (yt-dlp) |
| video-edit | Local video editing (ffmpeg) |
| video-understand | Local video understanding (ffmpeg + Whisper) |
The new skills take a different approach — instead of wrapping individual endpoints, they encode the full production workflow: discovery, scripting, prompt engineering, frame correction, and delivery.
License
MIT