Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add cosmicstack-labs/mercury-agent-skills --skill audio-extractiongit clone --depth 1 https://github.com/cosmicstack-labs/mercury-agent-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cosmicstack-labs/mercury-agent-skills/audio-extraction)<a href="https://agentmods.dev/skills/cosmicstack-labs/mercury-agent-skills/audio-extraction"><img src="https://agentmods.dev/badge/skills/cosmicstack-labs/mercury-agent-skills/audio-extraction/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cosmicstack-labs/mercury-agent-skills/audio-extraction"><img src="https://agentmods.dev/badge/skills/cosmicstack-labs/mercury-agent-skills/audio-extraction.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00020 | $0.04863 |
| Opus 5 | $0.00010 | $0.02431 |
| Sonnet 5 | $0.00004 | $0.00973 |
| Haiku 4.5 | $0.00002 | $0.00486 |
Grade A, and why
audio-extraction scanned grade A with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
from urllib.parse import urlparse Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
subprocess.run(cmd, capture_output=True) How it starts
The opening of the file, as written. The whole thing — 572 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Audio Extraction
Extract high-quality audio from video files, convert between formats, manage metadata, and build organized audio collections. This skill covers everything from one-off audio rips to batch processing pipelines.
Core Principles
1. Source Quality Determines Output Quality
You cannot create quality that wasn't captured. Start with the highest quality source available — lossy-to-lossy transcoding degrades audio further. Always extract from the best original source.
2. Choose the Right Format for the Use Case
- MP3 (lossy): Universal compatibility, great for music players and portable devices
- FLAC (lossless): Archival quality, for listening on quality equipment or future transcoding
- AAC/M4A: Better quality than MP3 at the same bitrate, native to Apple ecosystem
- OGG/Opus: Best quality-per-bitrate, perfect for streaming and podcasts
- WAV (uncompressed): Editing and production, not for everyday listening
3. Metadata Is Not Optional
Untagged audio files are unmanageable at scale. Proper ID3 tags, cover art, and consistent naming conventions turn a pile of files into a browsable music library.
4. Preserve the Original
Always keep a copy of the original file or at minimum log what source was used. Once you transcode, you lose information. Archival means keeping the best available original plus a convenient playback copy.
Audio Extraction with yt-dlp
Basic Audio Extraction
# Simplest audio extraction (best quality)
yt-dlp -x "https://youtube.com/watch?v=VIDEO_ID"
# Specific audio format
yt-dlp -x --audio-format mp3 "https://youtube.com/watch?v=VIDEO_ID"
# Best quality with metadata
yt-dlp -x --audio-format mp3 --audio-quality 0 \
--embed-thumbnail --embed-metadata "URL"
Format Conversion Options
# MP3 at various quality levels
yt-dlp -x --audio-format mp3 --audio-quality 0 "URL" # 320kbps (best)
yt-dlp -x --audio-format mp3 --audio-quality 2 "URL" # ~256kbps
yt-dlp -x --audio-format mp3 --audio-quality 5 "URL" # ~192kbps (good)
yt-dlp -x --audio-format mp3 --audio-quality 9 "URL" # ~128kbps (acceptable)
# FLAC (lossless)
yt-dlp -x --audio-format flac --audio-quality 0 "URL"
# AAC/M4A
yt-dlp -x --audio-format m4a "URL"
# Opus (best quality-per-bitrate)
yt-dlp -x --audio-format opus "URL"
# WAV (uncompressed)
yt-dlp -x --audio-format wav "URL"
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 572 lines · 20 tokens per session scan A 5840d9caa6fd
audio-extraction is a skill published in the GitHub repository cosmicstack-labs/mercury-agent-skills (471 stars, last pushed 17d ago), licensed MIT. It adds 20 tokens to every session and 4,863 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 2 findings (makes network calls, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
muapi-seedance-2
Expert Cinema Director skill for Seedance 2.0 (ByteDance) — high-fidelity video generation across Chinese, Global, and VIP tiers. Supports text-to-video, image-to-video, first-last-frame, omni reference, character training, omni-reference training, video editing, and watermark removal.
muapi-workflow
Build, run, and visualize multi-step AI generation workflows. The AI architect translates natural language descriptions into connected node graphs — chain image generation, video creation, enhancement, and editing into automated pipelines.
muapi-media-editing
Edit and enhance images and videos with AI via muapi.ai — prompt-based editing, upscaling, background removal, face swap, lipsync, video effects, and more.
muapi-media-generation
Generate AI images, videos, music, and audio from the terminal via muapi.ai — supports 100+ models including Flux, Midjourney v7, Kling 3.0, Veo3, and Suno V5.
wechat-article-writer
A Chinese-language content-creation workflow for public-account and social-media writing. It supports article drafting, cover images, illustrations, process diagrams, style analysis, and adapting writing to selected styles.
gemini-watermark-remover
Remove visible Gemini image watermarks from local image files by calling the project's CLI. Use when the user wants an agent to clean one or more local Gemini-generated images and save de-watermarked output files.