---
title: "Audio analysis skill for Claude"
canonical: "https://agenticfeats.com/c/audio-analysis-skill-for-claude-claude-code/"
agents: ["Claude Code"]
evidence_grade: "A"
outcome: "completed"
job_type: "Creative"
roles: ["Personal"]
source_url: "https://mager.co/blog/2026-03-26-loooom-identify-song-parts"
evidence_url: "https://www.loooom.xyz/s/mager/how-to-identify-parts-of-a-song"
published: "2026-10-13"
updated: "2026-10-13"
---

> From Agentic Feats: sourced examples of AI agents at work. Map of the site for agents: https://agenticfeats.com/llms.txt

# Audio analysis skill for Claude

Claude Code loaded a Gemini-written Loooom skill, installed its own audio libraries, and analyzed a 37-second piano recording of Espresso.

Result: It produced a report with BPM (~91), 4/4 meter, B major key, 50 onsets, a four-section timeline, and a stem list.

## How it was set up

Skill drafted with Gemini and published on loooom.xyz, then loaded in Claude Code. Claude installed ffmpeg and librosa mid-session and ran a Python analysis on a local MP3.

## Worth knowing

Estimated BPM of ~91 differed from the song's canonical ~104. The author says accuracy was not the goal.

## Prompt

```text
Use the song-structure skill to analyze [audio file path]. Report BPM, meter, key, a section timeline with timestamps, frequency brackets, and likely stems. Install any missing audio tools first.
```

## Evidence note

Blog post with terminal screenshots, published skill page, and SoundCloud clip

## Sources

- Original: https://mager.co/blog/2026-03-26-loooom-identify-song-parts
- Evidence: https://www.loooom.xyz/s/mager/how-to-identify-parts-of-a-song

## The receipt

- Agent: Claude Code
- Evidence: Shown (grade A, artifact linked)
- Outcome: Completed
- Job type: Creative
- Roles: Personal
- Source: Blog
- Added: 2026-10-13