Skip to main content
Glama

SPOTTER

AI-assisted sound design spotting for macOS. Analyze any video → get a complete cue sheet with timecodes → generate or search SFX → export DAW-synchronized stems.

Built by Lemö Labs.


What it is

Two things in one repo:

  • Electron app — visual timeline for manual spotting, clip editing, gain, fades, and export

  • MCP server — automated pipeline for use directly from Claude Code (no GUI needed)

Both share the same project format (analysis.json + media/ + export/).


Related MCP server: soundgrep

Electron app (macOS)

Requirements

  • macOS (Apple Silicon)

  • Node.js 18+

Install & run

git clone https://github.com/GuilleMcDugan/spotter.git
cd spotter
npm install
npm run dev

Build DMG

npm run dist
# → dist/SPOTTER-1.0.0-arm64.dmg

Features

  • AI video analysis (extracts frames → Claude → cue sheet)

  • Multi-track timeline with drag, trim, vertical clip movement, split (S key)

  • Per-clip gain and fade in/out (drag handles)

  • ElevenLabs SFX generation + Freesound search, built in

  • Export: WAV 16-bit / 24-bit / MP3 320k at 44.1kHz or 48kHz

  • DAW-synchronized stems (each file padded to full video length)

  • Auto-save timeline between sessions

  • J/K/L transport + frame-accurate scrubbing


MCP server (Claude Code integration)

See mcp-server/README.md for setup and tool reference.

Quick start:

cd mcp-server
npm install && npm run build

Add to ~/.claude/settings.json:

{
  "mcpServers": {
    "spotter": {
      "command": "node",
      "args": ["/absolute/path/to/spotter/mcp-server/dist/index.js"]
    }
  }
}

Guía de instalación

Documentación completa en Notion (español): SPOTTER — Guía de instalación


API keys

Keys are stored locally — never in the repo.

  • Electron app: Settings panel (⚙) → paste keys → stored in system keychain via electron-store

  • MCP server: configure_provider({ provider: "elevenlabs", api_key: "..." }) → stored in ~/.spotter/config.json

Required:

Claude API key: handled by the Claude CLI (claude auth), not by SPOTTER.


License

MIT

Related MCP Connectors

Related MCP Servers

  • F
    license
    Not graded
    quality
    C
    maintenance
    Enables Claude to classify audio across 7,000+ sound classes, including bird species and urban noise, with tools for batch processing, webhooks, and usage tracking.
    1
    -
  • A
    license
    Not graded
    quality
    D
    maintenance
    Enables semantic search of local audio samples by describing sounds, MIDI generation, stem separation, and other music production tools from Claude Desktop.
    2
    MIT
  • A
    license
    Not graded
    quality
    B
    maintenance
    Enables chat-driven audio analysis and enhancement using local Claude, including denoising, EQ, compression, and loudness normalization, with an A/B viewer for synchronized comparison.
    1
    MIT