audio-jingle

1# Audio Jingle Skill
2 
3Three sub-modes. The active project's `audioKind` decides which one
4runs:
5 
6| `audioKind` | Models we route to | Plan focus |
7|---|---|---|
8| `music` | Suno V5 (default), Udio, Lyria 2 | genre + tempo + instrumentation |
9| `speech` | MiniMax TTS (default), Fish, ElevenLabs V3 | script + voice + pacing |
10| `sfx` | ElevenLabs SFX (default), AudioCraft | texture + impact + duration |
11 
12## Resource map
13 
14```
15audio-jingle/
16├── SKILL.md
17└── example.html
18```
19 
20## Workflow
21 
22### Step 0 — Read the project metadata
23 
24`audioKind`, `audioModel`, `audioDuration` (seconds), and (for speech)
25`voice`. Branch by known values and use them verbatim. Missing metadata is not
26an instruction to ask: infer a safe default when possible, and emit a
27clarifying form only when the missing answer would materially change the
28requested output or prevent generation.
29 
30Important: `voice` is provider-specific. For `minimax-tts`, `--voice`
31must be a valid MiniMax `voice_id` (for example `male-qn-qingse`), not
32a natural-language description. If you only have a prose voice brief
33("warm female narrator", "neutral Mandarin"), keep that in your plan
34but omit `--voice` so the daemon's default voice id applies, or ask the
35user to choose a specific id.
36 
37### Step 1 — Plan
38 
39**Music**
40- Genre + reference artists (1-2)
41- Tempo (BPM) + key
42- Instrumentation (3-5 instruments max)
43- Vocals: yes / no / hummed / choir
44- Mood arc (intro → chorus → outro)
45 
46**Speech**
47- Script (final, not draft — TTS runs verbatim)
48…