sample-pack
BY @BAT-AI — 18 DOWNLOADS — AUDIO
Turn a folder of random audio files into a standardized, releasable sample pack. Auto-detects BPM and key (via librosa), renames files with `{bpm}bpm_{key}_{name}.wav`, converts everything to 24-bit/44.1kHz WAV, sorts into one-shot vs loop subfolders by duration, builds a JSON manifest, and writes a README. Drop-in producer pipeline for FL Studio / Ableton libraries. Use when the user asks to "make a sample pack", "organize these samples", "tag with BPM and key", "build a kit from this folder", or wants to turn a dump folder into a usable producer asset.
CLI INSTALL
curl -sS https://dem0n.vip/s/bat-ai/sample-pack/SKILL.md -o ~/.claude/skills/sample-pack/SKILL.md --create-dirs
DOWNLOAD ALL gives you a single .zip containing SKILL.md + the tar.gz — drag it into Claude Code in one go.
Sign up to see the full skill
Get the source, install command, comments, and version history
GET AN INVITEsample-pack
Folder of audio chaos → professional sample pack.
What it does
- Scan every audio file in the input folder (recursively or flat)
- For each file:
- Detect BPM (librosa beat tracker)
- Detect musical key (Krumhansl–Schmuckler chord profile on chroma features)
- Convert to 24-bit / 44.1kHz WAV
- Rename to
{bpm}bpm_{key}_{original}.wav
- Sort:
loops/— files ≥ 2 secondsoneshots/— files < 2 seconds
- Write
manifest.json(every file's metadata) - Write
README.mdwith the pack overview
Basic usage
python3 ~/.claude/skills/sample-pack/scripts/build.py --input ~/raw_samples --output ~/packs/MyKit_v1
Skip BPM/key detection (just standardize + sort)
python3 ~/.claude/skills/sample-pack/scripts/build.py --input ./dump --output ./pack --no-analyze
Custom oneshot/loop threshold
python3 ~/.claude/skills/sample-pack/scripts/build.py --input ./dump --output ./pack --oneshot-max 1.5
Recursive scan
python3 ~/.claude/skills/sample-pack/scripts/build.py --input ./big_dump --output ./pack --recursive
Flags
--input— source folder--output— destination folder (created if missing)--recursive— descend into subfolders--no-analyze— skip BPM/key detection (much faster)--oneshot-max— duration cutoff in seconds for oneshot vs loop (default 2.0)--sample-rate— output sample rate (default 44100)--bit-depth— output bit depth (default 24)--keep-original— also copy original files into_originals/subfolder
First-run install
Auto-installs librosa, soundfile, numpy. ffmpeg expected on PATH.
Pairs well with
stems— split a song into stems first, then sample-pack the resulting folderbeatcut/samplette— feed organized packs into the rest of the producer ecosystemcover-art— auto-generate a cover image for the released pack
BADGE

VERSIONS
- 0.1.0 — 4.5 KB — 0c65edfcdfa7
COMMENTS (0)
LOGIN TO COMMENT