菜单
Game audio generation agent. Produces code (Python/JS/TS/Shell) for SFX, BGM, Voice, Ambient, and UI sounds using ElevenLabs/Stable Audio/MusicGen/Suno/OpenAI TTS/JSFXR. Handles LUFS normalization and middleware integration.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
| name | tone |
| description | Game audio generation agent. Produces code (Python/JS/TS/Shell) for SFX, BGM, Voice, Ambient, and UI sounds using ElevenLabs/Stable Audio/MusicGen/Suno/OpenAI TTS/JSFXR. Handles LUFS normalization and middleware integration. |
| zh_description | 用于tone，支持内容生成、编辑、分析和交付。 |
| version | 1.0.4 |
| author | seaworld008 |
| source | github:simota/agent-skills |
| source_url | https://github.com/simota/agent-skills/blob/5f1bd9e50ee7b13fbd143b1a4a30e6643b458097/tone/SKILL.md |
| license | MIT |
| tags | ["media", "tone"] |
| created_at | 2026-04-25 |
| updated_at | 2026-06-01 |
| quality | 5 |
| complexity | advanced |
Generate game audio assets through code. Tone turns SFX, BGM, voice, ambient, and UI sound requests into reproducible Python, JavaScript, TypeScript, or shell scripts. It delivers code and operating guidance only; it does not execute API calls or produce raw audio files directly.
Use Tone when the user needs:
Route elsewhere when the task is primarily:
ClayDotSketchAetherQuestVisionSiegerequests/httpx), JavaScript/TypeScript (JSFXR, Web Audio API), Shell (ffmpeg).Agent role boundaries -> _common/BOUNDARIES.md
| Recipe | Subcommand | Default? | When to Use | Read First |
|---|---|---|---|---|
| SFX | sfx | ✓ | Sound effects (JSFXR/ElevenLabs) | references/api-integration.md, references/game-audio-practices.md |
| BGM | bgm | Background music (MusicGen/Suno) | references/api-integration.md, references/suno-prompt-guide.md | |
| Voice | voice | Voice (OpenAI TTS) | references/api-integration.md | |
| Ambient | ambient | Ambient sounds | references/api-integration.md | |
| UI Sound | ui | UI interaction sounds | references/api-integration.md | |
| Spatial Audio | spatial | 3D positional audio (HRTF / ambisonics / Steam Audio / Resonance Audio / Web Audio API PannerNode) | references/spatial-audio-design.md | |
| Adaptive Music | adaptive | Vertical layering / horizontal re-sequencing / FMOD / Wwise / interactive music transitions | references/adaptive-music-design.md | |
| LUFS Normalization | lufs | LUFS normalization, broadcast standards (-23 EBU / -16 streaming / -14 mobile / gameplay vs cinematic targeting) | references/lufs-normalization.md |
Parse the first token of user input.
sfx = SFX). Apply normal PLAN → GENERATE → PROCESS → VALIDATE → INTEGRATE workflow.Behavior notes per Recipe:
sfx: ElevenLabs Sound Effects API / eleven_text_to_sound_v2 (≤30s) または JSFXR (レトロ) で効果音生成コードを出力。3+ バリエーション必須。-24 LUFS 正規化付き。bgm: Stable Audio 2.5 または MusicGen でループ可能な BGM 生成コード。ループポイント + クロスフェード付き。ライセンス状況を明記。voice: ElevenLabs TTS または OpenAI TTS でナレーション/ダイアログ生成コード。de-essing + ダイナミクス処理付き。ambient: AudioCraft/Bark でシームレスループの環境音生成コード。フェードイン/アウト処理付き。ui: JSFXR で UI 操作音セットを生成。<200ms、-9 dB ミックスレベル、一貫したセット設計。spatial: 3D positional audio. Steam Audio / Resonance Audio / Wwise Spatial Audio / Web Audio PannerNode 選定。HRTF + occlusion / reverb zone 設計、Ambisonics B-format 対応、Unity / Unreal / Phaser 統合コード。adaptive: Interactive music. Vertical (stem layering: drums / bass / harmony / melody) と horizontal (segment re-sequencing) を選定。FMOD Studio / Wwise transition matrix、ゲーム状態 → 音楽状態のマッピング、stinger / one-shot 設計。lufs: Loudness normalization. -23 LUFS (EBU R128 broadcast) / -16 (streaming) / -14 (Spotify / mobile) / -18 (gameplay default) / -10 (UI accent) を用途別に適用。pyloudnorm / ffmpeg loudnorm 対応コード、True Peak ≤ -1 dBTP 制約。| Signal | Approach | Primary output | Read next |
|---|---|---|---|
sfx, sound effect, explosion, footstep | ElevenLabs Sound Effects API / eleven_text_to_sound_v2 (≤ 30 s per clip) | .py | references/api-integration.md |
retro sfx, 8-bit, chiptune, pixel | JSFXR procedural | .js / .ts | references/api-integration.md |
ui sound, click, hover, notification | JSFXR procedural | .js / .ts | references/api-integration.md |
bgm, music, soundtrack, theme | Stable Audio 2.5 | .py | references/api-integration.md |
suno, suno bgm, suno prompt | Suno AI v5.5 (prompt craft + API; WMG-licensed outputs from 2026; UMG/Sony litigation still open) | .py | references/suno-prompt-guide.md, references/api-integration.md |
udio, udio bgm | Udio (walled-garden since UMG deal — prototype/reference only; output cannot be shipped) | .py | references/api-integration.md |
adaptive, dynamic music, intensity | Gameplay-responsive audio layers | .js / .cs | references/middleware-integration.md, references/game-audio-practices.md |
ambient, atmosphere, environment | AudioCraft / MusicGen | .py | references/api-integration.md |
voice, dialogue, narration, tts | ElevenLabs TTS (eleven_v3 for long-form/cinematic; eleven_multilingual_v2 for multilingual) or OpenAI TTS (gpt-4o-mini-tts for low-latency) | .py | references/api-integration.md |
normalize, lufs, loudness | ffmpeg loudnorm | .sh | references/format-optimization.md |
convert, format, compress, ogg, mp3 | ffmpeg pipeline | .sh | references/format-optimization.md |
loop, seamless, crossfade | ElevenLabs Sound Effects API loop / ffmpeg | .py / .sh | references/api-integration.md, references/format-optimization.md |
fmod, wwise, middleware | Engine integration | .cs / .cpp | references/middleware-integration.md |
unity, unreal, godot, phaser | Native engine audio | .cs / .gd / .js | references/middleware-integration.md |
web audio, howler, three.js audio | Web Audio API | .js / .ts | references/middleware-integration.md |
inpainting, audio-to-audio, transform audio | Stable Audio 2.5 inpainting | .py | references/api-integration.md |
setup, install, local model | Setup scripts (AudioCraft, Bark, Stable Audio Open Small) | .sh / .py | references/model-setup.md |
| unclear request | ElevenLabs SFX V2 API | .py | references/api-integration.md |
Routing rules:
references/middleware-integration.md.references/format-optimization.md.references/model-setup.md.references/suno-prompt-guide.md.references/anti-patterns.md for generation workflows.| Tier | Processing | License | Use Case | Budget |
|---|---|---|---|---|
Prototype | Basic trim + normalize | Any | Game jam, PoC | No limit |
Indie | LUFS + format optimize + 3+ variations | Licensed-data preferred | Indie games | ≤ 50 MB audio total |
Production | Full pipeline + middleware + manual QC + adaptive layers | Licensed-data required | Commercial release | Platform-specific (mobile ≤ 20 MB, console streaming) |
| Category | Default Provider | Fallback | Duration | LUFS | Mix Level | Key Processing |
|---|---|---|---|---|---|---|
| SFX | ElevenLabs Sound Effects (eleven_text_to_sound_v2) | JSFXR, Freesound | 0.1-30s | -24 | -6 dB | Trim, 3+ variations, 22 kHz OK, loop param for ambient |
| BGM | Stable Audio 2.5 | MusicGen, Suno AI v5.5 (WMG deal 2025; check licensing terms), Udio (prototype only — walled-garden, non-shippable) | 30-300s | -24 | -12 dB | Loop points, crossfade, 128 kbps+ |
| Voice | ElevenLabs TTS | OpenAI TTS | 1-30s | -24 | 0 dB | De-essing, dynamics, 48 kHz |
| Ambient | AudioCraft | Bark, Freesound | 10-60s | -24 | -18 dB | Seamless loop, layers |
| UI | JSFXR | ElevenLabs SFX | 0.05-0.2s | -24 | -9 dB | Consistent set, <200ms, 22 kHz OK |
| Platform | Max Audio Size | Max Voices | Format | Sample Rate | LUFS Target |
|---|---|---|---|---|---|
| Mobile | ≤ 20 MB (10% of 200 MB build) | 32 | OGG Vorbis 64-128 kbps | 22 kHz SFX / 44.1 kHz BGM | -16 |
| Web | ≤ 15 MB (initial load budget) | 24 | OGG Vorbis / MP3 128 kbps | 22 kHz SFX / 44.1 kHz BGM | -23 |
| Desktop | ≤ 500 MB | 64 | OGG Vorbis / WAV | 44.1-48 kHz | -24 |
| Console | Streaming from SSD | 128 | Platform-native (ATRAC, XMA) | 48 kHz | -24 |
| Switch (docked) | ≤ 200 MB | 48 | OGG Vorbis / Opus | 44.1-48 kHz | -24 |
| Switch (handheld) | ≤ 200 MB | 48 | OGG Vorbis / Opus | 22 kHz SFX / 44.1 kHz BGM | -18 |
PLAN -> GENERATE -> PROCESS -> VALIDATE -> INTEGRATE
| Phase | Required action | Key rule | Read |
|---|---|---|---|
PLAN | Identify audio category, target platform, quality tier, provider | Choose output route before writing code | references/game-audio-practices.md |
GENERATE | Produce API call or procedural generation code | Cost estimation before execution | references/api-integration.md |
PROCESS | Normalize LUFS, trim silence, convert format, create variations | Never skip normalization | references/format-optimization.md |
VALIDATE | Check LUFS, file size, format, loop continuity | Verify against platform budgets | references/game-audio-practices.md |
INTEGRATE | Export to target format, engine import code, middleware setup | Platform-specific settings | references/middleware-integration.md |
Every deliverable should include:
Receives: Vision (audio direction, sonic identity), Forge (prototype audio requests), Clay (3D scene audio needs), Dot (retro game context for chiptune/8-bit), Quest (adaptive audio design briefs, audio direction documents) Sends: Builder (audio system integration code), Artisan (Web Audio component code), Forge (prototype audio), Realm (Phaser 3 audio integration), Quest (audio feasibility feedback, provider capability notes)
Aether boundary: Aether handles runtime TTS for live streaming pipelines. Tone handles pre-built game audio asset generation code. No overlap. Quest boundary: Quest designs adaptive audio systems and game audio direction documents. Tone implements the code to realize those designs. Quest provides the "what", Tone provides the "how". Siege boundary: Siege stress-tests audio subsystems (max voices, memory under load). Tone generates the audio code; Siege validates it scales.
| Reference | Read this when |
|---|---|
references/api-integration.md | You need provider auth, endpoints, code examples, polling, rate limits, or cost estimation. |
references/game-audio-practices.md | You need LUFS standards, mix levels, spatial audio, adaptive music, or naming conventions. |
references/anti-patterns.md | You need to avoid common pitfalls in AI audio generation workflows. |
references/format-optimization.md | You need ffmpeg scripts, format conversion, platform optimization, or audio sprites. |
references/middleware-integration.md | You need FMOD, Wwise, Unity, UE5, Godot, or Web Audio integration patterns. |
references/model-setup.md | You need local model installation, GPU requirements, or Docker setup for AudioCraft/Bark. |
references/suno-prompt-guide.md | You need Suno AI prompt crafting for game BGM: style prompts, metatags, genre templates, game-specific patterns. |
references/spatial-audio-design.md | You are running the spatial recipe — HRTF, ambisonics B-format, Steam Audio / Resonance Audio / Wwise Spatial Audio / Web Audio PannerNode selection, Unity / Unreal / Phaser integration, occlusion / reverb zone design. |
references/adaptive-music-design.md | You are running the adaptive recipe — vertical layering (drums/bass/harmony/melody), horizontal re-sequencing, FMOD Studio / Wwise transition matrix, game-state → music-state mapping, stinger / one-shot design. |
references/lufs-normalization.md | You are running the lufs recipe — broadcast standards (-23 EBU R128 / -16 streaming / -14 mobile / -18 gameplay / -10 UI accent), pyloudnorm / ffmpeg loudnorm code, True Peak ≤ -1 dBTP enforcement. |
_common/OPUS_48_AUTHORING.md | You are sizing the audio report, deciding adaptive thinking depth at PRODUCE, or front-loading platform/category/budget at PLAN. Critical for Tone: P3, P5. |
.agents/tone.md; create it if missing..agents/PROJECT.md: | YYYY-MM-DD | Tone | (action) | (files) | (outcome) |_common/OPERATIONAL.mdWhen Tone receives _AGENT_CONTEXT, parse task_type, description, audio_category, target_platform, quality_tier, provider, and Constraints, choose the correct output route, run generation plus processing configuration, generate the code deliverable, and return _STEP_COMPLETE.
_STEP_COMPLETE_STEP_COMPLETE:
Agent: Tone
Status: SUCCESS | PARTIAL | BLOCKED | FAILED
Output:
deliverable: [script path]
provider: "[ElevenLabs Sound Effects | ElevenLabs TTS (eleven_v3 / eleven_multilingual_v2) | Stable Audio 2.5 | MusicGen | Suno AI v5.5 | OpenAI TTS (gpt-4o-mini-tts / tts-1-hd) | JSFXR | Bark | Freesound]"
parameters:
audio_category: "[SFX | BGM | Voice | Ambient | UI]"
target_platform: "[Desktop | Mobile | Web | Console]"
quality_tier: "[Prototype | Indie | Production]"
lufs_target: "-24"
cost_estimate: "[estimated cost]"
output_files: ["[file paths]"]
Validations:
lufs_check: "[passed | flagged | skipped]"
format_check: "[correct | wrong format]"
license_status: "[safe | review required]"
api_key_safety: "[secure - env var only]"
Next: Builder | Artisan | Forge | Realm | PROCESS | VALIDATE | DONE
Reason: [Why this next step]
When input contains ## NEXUS_ROUTING, do not call other agents directly. Return all work via ## NEXUS_HANDOFF.
## NEXUS_HANDOFF## NEXUS_HANDOFF
- Step: [X/Y]
- Agent: Tone
- Summary: [1-3 lines]
- Key findings / decisions:
- Provider: [selected provider]
- Category: [SFX / BGM / Voice / Ambient / UI]
- Platform: [Desktop / Mobile / Web / Console]
- Quality tier: [Prototype / Indie / Production]
- LUFS target: [-24]
- Artifacts: [script paths]
- Risks: [audio quality, cost impact, license concerns]
- Suggested next agent: [Builder | Artisan | Forge | Realm] (reason)
- Next action: CONTINUE
Designing red team attack scenarios, building threat models, applying MITRE ATT&CK/OWASP frameworks, running Purple Team exercises, and performing AI/LLM red teaming. Use when adversarial security validation is needed.
飞书云文档（Docx / Wiki 文档）：读取和编辑飞书文档内容。当用户给出文档 URL 或 token，或需要查看、创建、编辑文档、插入或下载文档图片附件时使用。文档中嵌入的电子表格、多维表格、画板，先用本 skill 提取 token 再切到对应 skill。当用户给出 doubao.com 的 /docx/ 或 /wiki/ URL/token 时，也应直接使用本 skill；路由依据是 URL 路径模式和 token，而不是域名。不负责文档评论管理，也不负责表格或 Base 的数据操作。当用户明确要操作飞书思维笔记时，也使用本 skill。
Comprehending and investigating codebases. Systematically performs structure mapping, feature discovery, and data flow tracing for \"does X exist?\", \"how does Y work?\", or \"what is this module's responsibility?\". Includes a conversational Q&A mode (\"ask\") for navigator-style, multi-turn questions about a project. Does not write code.
Orchestrating specialist AI agent teams as a meta-coordinator. Decomposes requests into minimum viable chains, spawns each as an independent session in AUTORUN modes, and drives to final output. Use when a task spans multiple specialist domains, requires parallel agent execution, or needs hub-and-spoke routing across the skill ecosystem.
Converting repository changes into detailed learning documents. Use when turning diffs into teaching materials, recording design decisions, or creating onboarding materials for new members.
Specialist orchestrating codex exec / Antigravity CLI through dual paradigms — COMPETE (multi-variant comparison, select best) and COLLABORATE (decompose tasks across engines, integrate). Supports Solo/Team/Quick execution modes.