Audio Analyzer
Let Claude or Cursor analyze uploaded audio files with spectral, harmonic, rhythm, and structural metrics instead of guessing from filenames.
Overview
io.github.JuzzyDee/audio-analyzer is a Build-phase MCP server that runs local spectral, harmonic, rhythm, stereo, and structural analysis on audio files for agent-assisted development.
What is this MCP server?
- Spectral, harmonic, rhythm, stereo, and structural analysis exposed as MCP tools
- Prebuilt mcpb packages for darwin arm64/x64, linux x64, and win32 x64 over stdio
- Turns opaque waveform files into structured signals LLMs can cite in reviews and specs
- Local stdio transport—no hosted API key for the analyzer itself
- Useful for podcast tools, music apps, QA on mixes, and accessibility-oriented audio workflows
- Server version 1.0.0 with 4 platform mcpb packages (darwin arm64/x64, linux x64, win32 x64)
- stdio transport via Model Context Protocol registry schema 2025-12-11
Community signal: 27 GitHub stars.
What problem does it solve?
Agents cannot hear audio, so builders waste time manually inspecting waveforms or describing sound in vague terms.
Who is it for?
Indie builders adding music, podcast, or media features who want quantitative audio feedback inside Claude Code or Cursor.
Skip if: Teams that only need speech-to-text transcription without musical or spectral detail.
What do I get? / Deliverables
After you add the MCP server, your agent can return structured audio metrics you can paste into tickets, specs, and review comments.
- Structured spectral, harmonic, rhythm, stereo, and structural readouts per analyzed file
- Agent-grounded audio QA notes for specs and reviews
- Repeatable local analysis workflow without a separate hosted analyzer API
Recommended MCP Servers
Journey fit
Audio understanding is wired in during product build when you add media features, tooling, or agent workflows that must reason about sound. It sits on integrations because it extends the agent stack with a dedicated stdio MCP binary rather than a one-off script in app code.
How it compares
Signal-processing MCP server, not a cloud speech-transcription API skill.
Common Questions / FAQ
Who is audio-analyzer for?
Solo and indie builders using MCP agents who need harmonic, rhythm, and structural facts from local audio while building media or creative products.
When should I use audio-analyzer?
Use it during build when you are integrating audio uploads, mix checks, or content QA and want the agent to analyze files instead of guessing.
How do I add audio-analyzer to my agent?
Install the mcpb package matching your OS from the GitHub release, register the stdio MCP server in Claude Code or Cursor, then invoke analysis tools on file paths your agent can read.