# ai-avatar-video

A skill for making AI avatar and talking-head videos through the inference.sh CLI. Useful for presenter clips, UGC ads and dubbing.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 860700 installs · ★ 12
- Author: 101-skills
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/101-skills/superpowers/tree/HEAD/tools/video/ai-avatar-video/SKILL.md
- Updated: 2026-08-31
- HTML: https://affpapa.org/en/nevibecode/skills/ai-avatar-video

## What it does

The skill documents the `belt` CLI from inference.sh: turning a portrait plus a text script into an avatar video, syncing lips to an existing audio track, dubbing footage into another language. It ships as a set of commands and parameters covering several models. The recommended one is `pruna/p-video-avatar`, which handles speech synthesis itself, offers 30 voices across 10 languages and supports 1080p. Other options are `bytedance/omnihuman-1-5` for scenes with several people, `falai/fabric-1-0` and `falai/pixverse-lipsync`.

Models without built-in TTS — OmniHuman and PixVerse — need speech generated as a separate step, using `inworld/text-to-speech-2`, `ElevenLabs` or `infsh/kokoro-tts`. Dubbing takes a longer chain: transcribe with `infsh/fast-whisper-large-v3`, translate, synthesize the new voice, then sync lips with `infsh/latentsync-1-6`. The portrait itself can come from `pruna/p-image` at whatever aspect ratio you need.

You have to log in before anything runs: `belt login`. Without that, `belt app run` calls go nowhere. The skill is scoped to `Bash(belt *)`, so the agent drives the CLI directly rather than through a separate API client.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add 101-skills/superpowers --skill ai-avatar-video -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/ai-avatar-video/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/ai-avatar-video/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add 101-skills/superpowers --skill ai-avatar-video -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/ai-avatar-video/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add 101-skills/superpowers --skill ai-avatar-video -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/ai-avatar-video/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install 101-skills/superpowers ai-avatar-video
```

2. **Manually: put the files in place**
- File: `.github/skills/ai-avatar-video/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add 101-skills/superpowers --skill ai-avatar-video -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/101-skills/superpowers --path tools/video/ai-avatar-video
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/ai-avatar-video/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add 101-skills/superpowers --skill ai-avatar-video -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/101-skills/superpowers/tree/HEAD/tools/video/ai-avatar-video/SKILL.md

## When to use it
- turn a headshot and a script into a talking avatar clip
- dub a video into another language with matching lips
- generate a presenter portrait and animate it in one go
- produce a first-person UGC-style ad
- swap the audio in existing footage while keeping facial motion

## Example prompts
- Create a talking avatar video from `portrait.jpg` with the line "Hi, this is our product demo"
- Dub this clip into English and sync the lips with `infsh/latentsync-1-6`
- Generate a UGC-style portrait with `pruna/p-image` and animate it speaking in first person

## Similar
- [skill-creator](https://affpapa.org/en/nevibecode/skills/skill-creator.md) — skill · 414180 installs
- [writing-for-agents](https://affpapa.org/en/nevibecode/skills/writing-for-agents.md) — skill · 466719 installs
- [media-use](https://affpapa.org/en/nevibecode/skills/media-use.md) — skill · 1100000 installs
- [mcp-builder](https://affpapa.org/en/nevibecode/skills/mcp-builder.md) — skill · 130949 installs
- [subagent-driven-development](https://affpapa.org/en/nevibecode/skills/subagent-driven-development.md) — skill · 237746 installs
- [orchestration](https://affpapa.org/en/nevibecode/skills/orchestration.md) — skill · 349800 installs

## Other versions with the same name
- [ai-avatar-video · prime-skills](https://affpapa.org/en/nevibecode/skills/prime-skills-ai-avatar-video.md) — 379900 installs

From the same set: [101-skills/superpowers](https://affpapa.org/en/nevibecode/by/101-skills/superpowers.md) — 22 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md