# assemblyai

A skill for working with AssemblyAI: transcription, streaming STT, dictation and the LLM Gateway. It keeps the agent away from deprecated examples in its training data.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 570 installs · ★ 14
- Author: AssemblyAI
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/AssemblyAI/assemblyai-skill/tree/HEAD/skills/assemblyai/SKILL.md
- Updated: 2026-10-14
- HTML: https://affpapa.org/en/nevibecode/skills/assemblyai

## What it does

The skill covers the whole AssemblyAI surface: async transcription, sync STT for clips up to 120 seconds, dictation with rewriting, streaming v3 and the Voice Agent API. It also spells out the US and EU data zones, so audio and transcripts stay where they belong.

Most of the value is in fixing mistakes. Auth uses the header `Authorization: YOUR_API_KEY` with no `Bearer` prefix — that trips people up constantly. The skill also flags that the Java, Go and C# SDKs were discontinued in April 2025, leaving the REST API as the only route for those languages. Python, JavaScript/TypeScript and Ruby are the maintained ones.

Python 1.0.0 brings breaking changes: `aai.Lemur` and every `Lemur*` name are gone, the `assemblyai[extras]` option no longer installs, and `MicrophoneStream` was removed from the SDK — you capture mic audio with `pyaudio` or `sounddevice` instead. Node 4.37.0 drops `client.lemur` the same way. If your project pins an older version, check it before upgrading.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add AssemblyAI/assemblyai-skill --skill assemblyai -a claude-code
```

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

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add AssemblyAI/assemblyai-skill --skill assemblyai -a codex
```

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

### Cursor

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

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

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install AssemblyAI/assemblyai-skill assemblyai
```

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

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add AssemblyAI/assemblyai-skill --skill assemblyai -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/AssemblyAI/assemblyai-skill --path skills/assemblyai
```

### Windsurf (Devin Desktop)

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

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add AssemblyAI/assemblyai-skill --skill assemblyai -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/AssemblyAI/assemblyai-skill/tree/HEAD/skills/assemblyai/SKILL.md

## When to use it
- transcribe an audio or video file through the async API
- run real-time streaming speech recognition
- turn dictated notes into clean text
- split a recording by speaker and redact personal data
- build a voice agent on LiveKit or Pipecat that needs STT

## Example prompts
- Set up AssemblyAI and transcribe `meeting.mp3` with speaker diarization
- Wire up AssemblyAI streaming STT for my voice agent
- Check which AssemblyAI SDK version this project uses and update it

## Similar
- [videodb](https://affpapa.org/en/nevibecode/skills/videodb.md) — skill · 4500 installs
- [speech-to-text](https://affpapa.org/en/nevibecode/skills/speech-to-text.md) — skill · 11800 installs
- [openai-whisper](https://affpapa.org/en/nevibecode/skills/openai-whisper.md) — skill · 7100 installs
- [faster-whisper](https://affpapa.org/en/nevibecode/skills/faster-whisper.md) — skill · 1800 installs
- [asr](https://affpapa.org/en/nevibecode/skills/asr.md) — skill · 1300 installs
- [azure-speech-to-text](https://affpapa.org/en/nevibecode/skills/azure-speech-to-text.md) — skill · 775 installs

---
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