@audiolabtools/mcp-server

@audiolabtools/mcp-server

MCP (Model Context Protocol) server that gives any MCP-capable AI — Claude Desktop, Claude Code, Cursor, and others — eight audio-analysis tools, backed by the hosted AudioLab API. It is a thin HTTP client: no local audio engine, no ffmpeg, nothing to compile.

Install

Point your MCP client at the package via npx (nothing to install globally):

{
  "mcpServers": {
    "audiolab": {
      "command": "npx",
      "args": ["-y", "@audiolabtools/mcp-server"],
      "env": { "AUDIOLAB_API_KEY": "al_live_yourkey" }
    }
  }
}

Get a key: sign in at https://audiolab.tools/account and generate one (free tier available).

Requirements

• Node ≥ 18 — uses the built-in global fetch + AbortSignal.timeout.
• An AUDIOLAB_API_KEY. No ffmpeg, no native dependencies.

Tools

Every tool takes a public https URL to the audio; the API fetches and analyses it server-side and returns numbers only (nothing is uploaded from your machine).

Tool	Input	Returns
analyze_loudness	{ url }	Integrated LUFS (EBU R128 / BS.1770-4), true-peak (dBTP), LRA, crest factor, stereo correlation, mono compatibility, tonal balance
check_target	{ url, target, lufs?, tp? }	Pass/fail vs a delivery target (spotify / apple-music / youtube / tidal / amazon-music / podcast / ebu-broadcast / atsc-broadcast, or target:"custom" + lufs+tp), with per-metric deltas and an ffmpeg loudnorm fix command
analyze_timeseries	{ url, waveformPoints? }	Short-term LUFS over time + downsampled waveform peaks (for graphs/meters)
get_spectrum	{ url }	FFT magnitude data + 7-band energies + dominant band
analyze_voice	{ url }	Voice QA: speech/silence ratio, speaking rate, SNR, noise floor, room echo, sibilance & clipping risk
get_speech_segments	{ url }	Voiced regions with start/end + per-segment RMS (auto-trim, chapters)
index_signal	{ url }	Content-type guess, tags, clipping/silence regions, brightness & dynamics buckets
compare_loudness	{ urlA, urlB }	Runs loudness on both and returns both results for an A/B comparison

Example asks to your AI:

• “Analyze the loudness of https://example.com/track.wav” → analyze_loudness
• “Does https://example.com/mix.wav pass Spotify?” → check_target with target:"spotify"
• “A/B-compare masterA.wav vs masterB.wav for loudness” → compare_loudness

Configuration (env)

Var	Default	Purpose
AUDIOLAB_API_KEY	— (required)	Your API key.
AUDIOLAB_API_BASE	https://audiolab.tools/v1	Override the API base (must be https://).
AUDIOLAB_TIMEOUT_MS	60000	Per-request timeout in milliseconds.

Privacy

The tools send the audio URL and your API key to the AudioLab API (audiolab.tools), which fetches and analyses the audio server-side and returns numbers only — your audio is not stored (see https://audiolab.tools/privacy). This package has no telemetry and writes nothing to disk.

Limits

• One file per call (agents loop for many); one-shot (no streaming/realtime).
• File-size and rate limits are enforced by the API, per key.

Smoke test

node hosted-server.mjs --selftest   # verifies the 8 tools register + the missing-key guard; no network

License

MIT © Nathan Renting