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 agentmods add skills/webern/mx/mx-commentsnpx skills add webern/mx --skill mx-commentsgit clone --depth 1 https://github.com/webern/mxWhat 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 | $0.00022 | $0.00479 |
| Opus 5 | $0.00011 | $0.00239 |
| Sonnet 5 | $0.00004 | $0.00096 |
| Haiku 4.5 | $0.00002 | $0.00048 |
Grade A, and why
mx-comments scanned grade A with 0 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 2d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
/mx-comments
Code comments are meant to help a developer or coding agent who is unfamiliar with this codebase understand how it works.
Audience:
- A musician, a composer, who happens to also like writing code.
- A software developer who happens to also like writing music.
Things to remember:
- Do not leak particulars of a user session, or your LLM context window, into code comments or repository writing. Code comments are not an extension of the conversation you had with the user.
- Code comments should not churn every time the code changes. Do not list each item that a function touches in its documentation such that you have to add to the list every time you edit the function's behavior. Write more generally so that the code comment can outlive future tweaks to the function.
Avoid implementation details in the headers of src/mx/api. Those are particularly musician-facing
code comments. The more maintainer facing comments go in the deeper layers of the system.
Please
Use plain grammar and simple sentence structure. Use common vocabulary that a human composer or software engineer would use in conversation. When the first half of a sentence is enough, don't add a continuation of the sentence that says the same thing in counterpoint.
Pretend you are humble engineer explaining how something works, or why it is the way that it is, without trying to look clever. Go out of your way to write with humility and to be easily understood. You are not here to show how smart you are. Do not increase the cognitive load of those trying to read your work.
Banned Words and Structures
- "knob": you invented this and you overuse it. stop.
- Don't say "writes no attribute". That's a weird non-human style. Instead say "does not write an attribute". More generally do not say "{{verbs}} no {{noun}}". Instead say "does not {{verb}} a {{noun}}"
- "load bearing": it's weird, you invented it, and you overuse it.
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.
- 2d ago First seen · 48 lines · 22 tokens per session scan A 1b186fcebd79
mx-comments is a skill published in the GitHub repository webern/mx (93 stars, last pushed 3d ago), licensed MIT. It adds 22 tokens to every session and 479 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
musicbox
Use when the user wants to play/pause/skip music, control volume, seek, search songs/playlists, query NetEase Music data, or operate NetEase MusicBox. Drives MusicBox through the musicbox CLI + daemon; never simulates terminal keypresses to the curses TUI.
spotatui-dj
Be the DJ for spotatui, the terminal music player, by driving its MCP server. Use whenever the user asks for music, asks you to DJ, wants tracks queued, played, skipped, or searched, asks what they have been listening to, or mentions spotatui.
videoagent-audio-studio
Tired of juggling multiple audio APIs? This skill gives you one-command access to TTS, music generation, sound effects, and voice cloning. Use when you want to generate any audio without managing multiple API keys.
agents
Build voice AI agents with ElevenLabs. Use when creating voice assistants, customer service bots, interactive voice characters, or any real-time voice conversation experience, and when configuring an agent's tools, workflows, or procedures, including creating, editing, compiling, and publishing procedure drafts on an…
update-skills-from-changelog
Update ElevenLabs agent skills from a merged weekly changelog in elevenlabs-dx, then open a pull request in elevenlabs/skills. Trigger after a changelog merges to main on elevenlabs-dx, or when asked to update skills from changelog YYYY-MM-DD.
dubbing
Dub audio and video into other languages using the ElevenLabs Dubbing API (dubbingv2), preserving the original speakers' voices. Use when translating videos, podcasts, or recordings into other languages, localizing media content, reviewing or correcting dubbing transcripts and translations, or regenerating a dub after…