Extract Audio
- 32 installs
- 269 repo stars
- Updated June 11, 2026
- gupsammy/claudest
Strip or export soundtrack from video files with ffmpeg/ffprobe format rules instead of guessing codec flags.
About
extract-audio is an agent skill that turns natural-language requests like “get the mp3” or “strip audio from video” into a repeatable ffmpeg workflow. Solo builders shipping courses, podcasts, or short-form content install it when they need dependable audio exports without memorizing codec tables. The skill opens with a format decision tree—music archive vs voice podcast vs DAW editing—then walks through probing streams with ffprobe and applying the matching ffmpeg flags. It sits in the build phase as a CLI integration pattern: the agent runs constrained Bash, interprets stream metadata, and can ask clarifying questions when multiple audio tracks exist. Compared to one-off chat answers, the procedural structure reduces wrong-bitrate exports and accidental re-encodes when copy would suffice.
- Format decision tree maps use cases to FLAC, MP3 VBR/CBR, AAC, WAV, or stream copy
- ffprobe JSON probe step lists audio streams before choosing encode flags
- Explicit Bash allowlist for ffprobe and ffmpeg only plus AskUserQuestion for ambiguous inputs
- Re-encode vs copy guidance avoids unnecessary quality loss when source already matches target
Extract Audio by the numbers
- 32 all-time installs (skills.sh)
- Ranked #343 of 550 CLI & Terminal skills by installs in the Skillselion catalog
- Security screen: LOW risk (skills.sh audit)
- Data as of Aug 4, 2026 (Skillselion catalog sync)
npx skills add https://github.com/gupsammy/claudest --skill extract-audioAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 32 |
|---|---|
| repo stars | ★ 269 |
| Security audit | 3 / 3 scanners passed |
| Last updated | June 11, 2026 |
| Repository | gupsammy/claudest ↗ |
What it does
Strip or export soundtrack from video files with ffmpeg/ffprobe format rules instead of guessing codec flags.
Files
Format Decision Tree
| User wants | Format | Flags | Why |
|---|---|---|---|
| Music, archive quality | FLAC | -c:a flac | Lossless, no quality loss |
| Music, small + transparent | MP3 VBR | -c:a libmp3lame -q:a 0 | ~200kbps avg, perceptually lossless |
| Podcast / voice | MP3 128k CBR | -c:a libmp3lame -b:a 128k | Sufficient for speech, universally compatible |
| Mobile / streaming | AAC 192k | -c:a aac -b:a 192k | Better than MP3 at equivalent bitrate |
| DAW / editing | WAV | -c:a pcm_s16le -ar 44100 | No encoding loss, widest DAW support |
| Source already target format | Copy | -c:a copy | No re-encode, instant, lossless |
Process
1. Probe audio streams
ffprobe -v quiet -print_format json -show_streams "$INPUT" | \
python3 -c "
import json, sys
streams = [s for s in json.load(sys.stdin)['streams'] if s['codec_type']=='audio']
for i, s in enumerate(streams):
print(f'Stream {i}: {s[\"codec_name\"]} {s.get(\"bit_rate\",\"?\")} bps {s.get(\"channel_layout\",\"?\")}')
"2. Determine format
Apply the decision tree above if the user didn't specify. If the source audio codec already matches the target, use -c:a copy to avoid transcoding.
If multiple audio streams exist, ask the user which to extract — or use -map 0:a to extract all. Once the user responds, apply -map 0:a:N (where N is the zero-based stream index they chose) or -map 0:a for all streams in the Phase 3 command.
3. Construct command
# General pattern (-vn drops the video stream entirely):
ffmpeg -i "$INPUT" -vn [FORMAT_FLAGS] "$OUTPUT"
# Examples:
ffmpeg -i video.mp4 -vn -c:a libmp3lame -q:a 0 audio.mp3 # MP3 VBR best quality
ffmpeg -i video.mp4 -vn -c:a libmp3lame -b:a 128k podcast.mp3 # MP3 128k CBR
ffmpeg -i video.mp4 -vn -c:a flac archive.flac # FLAC lossless
ffmpeg -i video.mp4 -vn -c:a aac -b:a 192k mobile.aac # AAC
ffmpeg -i video.mp4 -vn -c:a pcm_s16le -ar 44100 edit.wav # WAV for DAW
ffmpeg -i video.mp4 -vn -c:a copy original.m4a # Copy audio stream4. Confirm and run
Show: detected source codec and bitrate, chosen output format, output path. Wait for approval, then run.
Report output file size and duration: ffprobe -v quiet -show_format "$OUTPUT" | grep -E "duration|size"
Key Decisions
Preserve generation quality: avoid transcoding chains that degrade source fidelity. Each decision below is an application of this principle.
- Lossy-to-lossy warning: if the source is already lossy (MP3, AAC, OGG) and the user wants a different lossy format, warn them that re-encoding degrades quality. Recommend keeping the source format or using
-c:a copywhere container compatibility allows. - For files >1 hour, ask whether the user wants the full file or a specific range — trimming can be added with
-ssand-tobefore-vn. - M4A vs AAC: AAC is the codec, M4A is the container. Use
.m4aextension for Apple device compatibility; use.aacfor a raw stream.
Related skills
Lark SharedWire Feishu/Lark Open Platform auth, tokens, and shared API helpers into agent-built bots and internal tools.514k
Runcomfy CliLet their coding agent call any RunComfy model from the terminal or scripts without managing separate SDKs.298k30
Skills CliDiscover, install, update, and manage reusable agent capabilities from the open Skillselion ecosystem directly inside their coding agent.270k71TzstCreate, extract, list, and script `.tzst` / `.tar.zst` archives from the terminal with the right compression, streaming, and conflict flags.267k71XdropPull files from an Xdrop share URL, decrypt them locally, and land outputs in a chosen directory with optional JSON for automation.266k71
WizardAuthor interactive bash setup wizards that open URLs, capture secrets, write .env upserts, and set GitHub Actions secrets step by step.104k202kFAQ
Is Extract Audio safe to install?
skills.sh reports 3 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.