# add-voice

Adds two-way Grok voice conversations to an app. Use it when replacing an STT-LLM-TTS chain or an OpenAI Realtime integration.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 319 installs · ★ 14099
- Author: cursor
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/cursor/plugins/tree/HEAD/grok-voice/skills/add-voice/SKILL.md
- Updated: 2026-10-25
- HTML: https://affpapa.org/en/nevibecode/skills/add-voice

## What it does

This skill wires an app to Grok voice over WebSocket: it captures microphone audio, sends audio chunks, and plays returned speech as it arrives. It covers session setup, server-side VAD, `grok-transcribe`, and interruption when the user starts speaking.

For browser and mobile clients, it directs the app to request a temporary token from its backend; server connections use `XAI_API_KEY`. Do not ship a long-lived key in a client bundle, so review the existing auth path before adding voice. TypeScript is the main example, with a Python version using `websockets`.

The skill also sets composer behaviour. A waveform button starts voice mode, while the microphone icon remains reserved for dictation. Use `/add-dictation` for speech-to-text only, `/add-read-aloud` for spoken text replies, and `/debug-voice` for development logging and log-based fixes.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add cursor/plugins --skill add-voice -a claude-code
```

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

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add cursor/plugins --skill add-voice -a codex
```

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

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add cursor/plugins --skill add-voice -a cursor
```

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

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install cursor/plugins add-voice
```

2. **Manually: put the files in place**
- File: `.github/skills/add-voice/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add cursor/plugins --skill add-voice -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/cursor/plugins --path grok-voice/skills/add-voice
```

### Windsurf (Devin Desktop)

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

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add cursor/plugins --skill add-voice -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/cursor/plugins/tree/HEAD/grok-voice/skills/add-voice/SKILL.md

## When to use it
- add a two-way Grok voice conversation
- replace an STT-LLM-TTS chain with one connection
- keep long-lived API keys out of client bundles
- interrupt queued audio when the user speaks
- separate voice mode from dictation controls

## Example prompts
- \`/add-voice` Add Grok voice mode to this app.
- Add a Grok speech-to-speech session and wire the app microphone.
- \`/add-voice` Replace the current STT-LLM-TTS flow with Grok voice.

## Similar
- [realtime](https://affpapa.org/en/nevibecode/skills/realtime.md) — skill · 634 installs
- [bun-websocket-server](https://affpapa.org/en/nevibecode/skills/bun-websocket-server.md) — skill · 394 installs
- [azure-messaging-webpubsubservice-py](https://affpapa.org/en/nevibecode/skills/azure-messaging-webpubsubservice-py.md) — skill · 1 install
- [twilio-voice-twiml](https://affpapa.org/en/nevibecode/skills/twilio-voice-twiml.md) — skill · ★ 7596
- [neon-realtime](https://affpapa.org/en/nevibecode/skills/neon-realtime.md) — skill · 6000 installs
- [speech-engine](https://affpapa.org/en/nevibecode/skills/speech-engine.md) — skill · 6500 installs

From the same set: [cursor/plugins](https://affpapa.org/en/nevibecode/by/cursor/plugins.md) — 47 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