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/guvense/hellyee/sound-checknpx skills add guvense/hellyee --skill sound-checkgit clone --depth 1 https://github.com/guvense/hellyeeWrote 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/guvense/hellyee/sound-check)<a href="https://agentmods.dev/skills/guvense/hellyee/sound-check"><img src="https://agentmods.dev/badge/skills/guvense/hellyee/sound-check.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 | $0.00144 | $0.01409 |
| Opus 5 | $0.00072 | $0.00705 |
| Sonnet 5 | $0.00029 | $0.00282 |
| Haiku 4.5 | $0.00014 | $0.00141 |
Grade A, and why
sound-check 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 4d 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 — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sound check
Everything in Live fails loudly except silence. A wrong note errors, a bad index errors, a missing file errors — but a track that makes no sound reports nothing at all. It writes clean, it places clean, it renders clean, and the only trace is a band that reads 0.0% in an analysis you might not run until the arrangement is finished.
So check before you build. Thirty seconds here saves rebuilding an arrangement.
When
- After loading instruments and writing the first session clips — before
place_in_arrangement. - After loading any drum rack, always.
- Whenever the user says something is missing, thin, or inaudible.
- Whenever a
record_masterband reads near zero where you expected content.
The check
1. Drum racks: read the pads, don't trust the map.
get_drum_map returns Live's standard mapping. get_drum_pads(track, device)
returns what the loaded rack actually fills. They are not the same — factory
kits routinely cover only 36–51, so the shaker at 70 is silence. Write only
notes the rack lists; if you want 16th shakers and there is no shaker pad, use
the closed hat at low velocity instead.
2. One measured pass over every track.
Fire nothing. Place a few bars, then:
measure_tracks(track_indices=[0,1,2,…], start_bar=<a bar where all of them play>, seconds=15)
One playback pass, ranked by peak, silent flagged under 0.02. Pass
start_bar — without it you measure wherever the playhead sits, and a playhead
parked outside the section reads 0.000 on every track, which looks exactly
like a broken set.
3. Read the ranking as a mix, not just a checklist.
For four-on-the-floor: kick on top, bass under it, lead below that, then percussion, then pads. If the ranking is upside down you have found a mix problem before writing 176 bars over it.
A track reads silent — the decision tree
Never raise the fader first. Work down this list:
| check | how | fix |
|---|---|---|
| Is there an instrument? | list_track_devices |
MIDI track with only effects makes no sound |
| Is it a drum note that exists? | get_drum_pads |
rewrite the pattern onto filled pads |
| Was a session clip ever fired? | — | back_to_arranger |
| Is the track muted, or another soloed? | get_song_status |
unmute |
| Does the section actually contain the clip? | get_arrangement_clips |
you measured the wrong bar |
| Is the filter shut? | list_device_parameters |
see the cutoff floors in the hellyee skill |
| Genuinely quiet, not silent? | solo + record_master 4 bars |
now a level problem — see below |
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.
- 4d ago First seen · 115 lines · 144 tokens per session scan A 487a5f029f7f
sound-check is a skill published in the GitHub repository guvense/hellyee (6 stars, last pushed 14d ago), licensed MIT. It adds 144 tokens to every session and 1,409 once invoked, about $0.0007 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 skills, from other repositories
mastering-chain
Master the current Ableton Live set — build a Glue Compressor → (EQ Eight) → Limiter chain on the Main track with conservative, streaming-safe settings. Use when the user asks to master the track, finish/polish the mix, "make it louder", add glue/punch to the whole mix, or prepare the song for export or streaming.
post-to-xhs
小红书内容发布技能。支持两种发布模式:(1) 上传图文模式 - 图片+短文;(2) 写长文模式 - 长篇文章+排版模板。 支持两种输入方式:用户提供完整内容和图片/图片URL,直接发布;或提供网页URL,自动提取内容和图片。 用户说"发长文"时使用长文模式,否则默认图文模式。.
AI Image & Video Generator — GPT Image 2, Seedance, ComfyUI
Generate images and videos from text with multi-provider routing — supports GPT Image 2.0 (near-perfect text rendering), Nanobanana 2, Seedream 5.0, Midjourney V8.1 (unified photorealistic + anime), Flux 2 Klein (cheap drafts), Seedance 2.0 / Veo 3.1 / Grok Video / Agnes Video, and local ComfyUI workflows. Includes…
sound-cues
Create and modify SoundCue assets — add/connect nodes (mixer, random, delay, attenuation, modulator, etc.) and set audio properties (SoundCueService). Use when the user asks to create a Sound Cue, build a SoundCue node graph, or wire audio playback logic.
vibeue
Unreal Engine 5 development using the VibeUE Python API. Use when working in Unreal Engine — blueprints, state trees, materials, actors, landscapes, animation, niagara, widgets, sound, foliage, gameplay tags, enhanced input, skeletons, PCG (procedural content generation), and more. VibeUE is an extension of Unreal's…
gemini-skill
通过 Gemini 官网(gemini.google.com)执行生图、对话等操作。用户提到"生图/画图/绘图/nano banana/nanobanana/生成图片"等关键词时触发。操作方式分三级优先级:首选 MCP 工具 → 次选 Skill 脚本 → 最次连接 Skill 浏览器手动操作(需用户授权)。禁止自行启动外部浏览器访问 Gemini。.