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.
git clone --depth 1 https://github.com/punt-labs/voxWrote 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/commands/punt-labs/vox/music)<a href="https://agentmods.dev/commands/punt-labs/vox/music"><img src="https://agentmods.dev/badge/commands/punt-labs/vox/music.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00004 | $0.02555 |
| Opus 5 | $0.00002 | $0.01277 |
| Sonnet 5 | $0.00001 | $0.00511 |
| Haiku 4.5 | $0.00000 | $0.00255 |
Grade A, and why
music 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/music command
Control vibe-driven background music generation. /music on plays the first
track as soon as it is ready and then, with no further commands, generates the
rest of the (vibe, style) pool (up to 12 distinct tracks) in the background
and auto-advances to a different track as each one ends. Once the pool has
12, generation stops and playback rotates the pool (shuffled, never the
just-played track) at zero credits.
While music is playing, a vibe or style change re-pools the music: the current
song finishes, then playback switches to that mood's pool — rotating it in for
free if those tracks already exist, or generating them if the pool is new. When
music is off, a vibe change updates the session mood only; it does not start
or generate music. There is no confirmation and no credit prompt — the re-pool
is the intended effect of the mood change. (The /vibe skill drives this: a
vibe change while music plays returns a music_hint, and the agent authors the
new pool and calls this tool.)
You author the prompts. vox is a pipe to ElevenLabs; it never decides what a
genre sounds like. When you turn music on (and on any style/vibe change) YOU --
the agent, using your own genre knowledge -- write a base_prompt plus exactly
12 literal, genre-accurate variations (one per pool slot) and pass them to the
music tool. voxd generates track i from base_prompt + variations[i], so
the 12 tracks are 12 distinct tracks within the genre. If you pass neither,
voxd falls back to a bare "<style> music, <mood>. instrumental, loopable."
prompt -- functional, but flavorless.
Usage
/music on-- start music with current vibe/music on style techno-- start music with a style modifier/music on --title "Focus Beats"-- start music and name the album "Focus Beats"/music stop-- stop music/music next-- optional manual skip (playback auto-advances on its own): jump to the next track now — rotate the pool if it holds 12+ (zero credits), else generate a fresh one/music prev-- step back to the previous part of the now-playing album/music pause-- suspend the current album in place/music resume-- resume the suspended album in place/music play <name>-- replay a saved album by name/music play-- with no argument, replay the last-played album (errors and lists the catalog if nothing has played yet)/music list-- show saved albums with metadata/music status-- show current music state (same as/musicwith no argument)/music-- show current music state
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.
- 7d ago First seen · 196 lines · 4 tokens per session scan A 7c1d998995b3
music is a command published in the GitHub repository punt-labs/vox (3 stars, last pushed 4d ago), licensed MIT. It adds 4 tokens to every session and 2,555 once invoked, about $0.0000 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 2026-08-31.
Other commands, from other repositories
gen-video
Generate video output in one of three modes: Remotion (programmatic React-based video), Gen-AI (shot prompt pack for Runway/Kling/Sora/Pika/Luma), or Remotion+Gen-AI (programmatic scaffold with AI footage embedded).
brand-generate
Generate an on-brand document from a saved Brand Profile.
video
WATCH a video and return a structured VISUAL breakdown (not just a transcript) with Antigravity (agy / Gemini 3.x, natively multimodal in video). Scene-by-scene segments with timestamps, on-screen text/OCR (slides, charts, UI, captions), key visual events, and a summary. For screencasts, tutorials, presentations…
partition-dev
Derive test cases from Z specification using TTF testing tactics.
review
Peer review a PR/FAQ document for quality and decision readiness.
b-animate-dev
Animate and model-check a B machine with probcli.