# AudioKit

> AudioKit is a freemium MCP server for design and media. Early-access MCP server for audio jobs: tempo, key, fingerprint, tags and loudness. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Freemium
- **Platforms:** API
- **Website:** https://audiokit.ai/docs/mcp/
- **MCP Registry name:** io.github.epsilon-records/audiokit
- **Categories:** design and media
- **Listing updated:** 2026-11-15

## About

Because the MCP server is the same API as `/v1`, it uses the same keys, validation and jobs. Audio is passed as an https URL plus its SHA-256, and URLs must resolve to public addresses. Four analysis shortcuts submit a job and wait: `tempo` (BPM and beat grid, search range 40-300), `key`, `fingerprint` (AcoustID-compatible Chromaprint) and `tags` (genre, mood and vocals via CLAP, listed only when the instance has the model). `analyze` bundles loudness, tempo, key, fingerprint and tags in one pass.

Other operations go through `submit_job`, `get_job` and `cancel_job`, with a `wait_seconds` setting of up to 300 and an idempotency `job_id`. If time runs out the job comes back queued or running, which is not an error. A failed job is a result with `status: failed`, not a tool error.

Pricing is in credits at $0.01 each: analysis costs 1 credit a track, the analyze bundle 3, stems 3 per audio minute with a 10-credit minimum, and transcoding or normalising 1 per minute. The service is in beta, so packs cannot be bought yet. No operation currently needs consent, though future voice tools will.

## Features

- **Tempo and beat grid:** Returns BPM within a configurable range.
- **Key detection:** Reports musical key.
- **Fingerprinting:** AcoustID-compatible Chromaprint output.
- **One-pass analyze:** Loudness, tempo, key, fingerprint and tags together.
- **Job control:** Submit, poll and cancel jobs with an idempotency ID.
- **Credit pricing:** Per-operation credits with a monthly free allowance.

## Good for

- Detect BPM and key of a track library
- Fingerprint audio for identification
- Tag tracks by genre, mood and vocals
- Run stem separation or cleanup jobs through submit_job

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http audiokit https://api.audiokit.ai/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://api.audiokit.ai/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "audiokit": {
      "url": "https://api.audiokit.ai/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "audiokit": {
      "type": "http",
      "url": "https://api.audiokit.ai/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "audiokit": {
      "serverUrl": "https://api.audiokit.ai/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "audiokit": {
      "type": "http",
      "url": "https://api.audiokit.ai/mcp"
    }
  }
}
```

## Tools

- `health`: Service status and version.
- `list_operations`: Operations this instance can run.
- `submit_job`: Submits any job.
- `get_job`: Gets a job's status and result.
- `cancel_job`: Cancels a job.
- `tempo`: BPM and beat grid.
- `key`: Musical key.
- `fingerprint`: AcoustID-compatible Chromaprint.
- `analyze`: Loudness, tempo, key, fingerprint and tags in one pass.
- `tags`: Genre, mood and vocals.

## FAQ

### How do I authenticate?

With an AudioKit API key as a Bearer header.

### What does it cost?

One credit is $0.01; accounts get 100 free credits a month.

### Can I buy credits now?

Not yet; packs open after the beta.

### What happens if a job runs long?

The call returns the job as queued or running and you poll with get_job.

### Is AudioKit free?

AudioKit has a free plan, with paid plans for more usage or features.

### How do I add AudioKit to Claude Code?

Run this command in your terminal: claude mcp add --transport http audiokit https://api.audiokit.ai/mcp

### What are the best alternatives to AudioKit?

The closest alternatives to AudioKit are HyperFrames by HeyGen, ffmpeg-skill, AvatarSay — Fictional Character & Talking Video MCP, ChartAnt and VfxSmith.

## Alternatives

- [HyperFrames by HeyGen](https://ai-powered.com/mcp/hyperframes-by-heygen): HyperFrames by HeyGen is a freemium MCP server for design and media.
- [ffmpeg-skill](https://ai-powered.com/mcp/ffmpeg-skill): ffmpeg-skill is an open-source MCP server for design and media.
- [AvatarSay — Fictional Character & Talking Video MCP](https://ai-powered.com/mcp/avatarsay-fictional-character-and-talking-video-mcp): AvatarSay — Fictional Character & Talking Video MCP is a freemium MCP server for design and media.
- [ChartAnt](https://ai-powered.com/mcp/chartant): ChartAnt is a freemium MCP server for design and media.
- [VfxSmith](https://ai-powered.com/mcp/vfxsmith): VfxSmith is a freemium MCP server for design and media.
- [CalderFlow](https://ai-powered.com/mcp/calderflow): CalderFlow is a free MCP server for design and media.
- [Eikino](https://ai-powered.com/mcp/eikino): Eikino is a freemium MCP server for security.
- [Hourlore](https://ai-powered.com/mcp/hourlore): Hourlore is a freemium MCP server for design and media.

---
Source: https://ai-powered.com/mcp/audiokit
