Interactive wizard to design sound effects, producing SFX specifications in conversation context for immediate synthesis.
Guides interactive sound effect design, producing detailed synthesis specifications for immediate audio generation.
/plugin marketplace add nethercore-systems/nethercore-ai-plugins/plugin install sound-design@nethercore-ai-plugins[sound type, e.g., 'sword swing', 'jump', 'explosion']Guide the user through designing a sound effect, producing a detailed specification with layering and implementation guidance.
Check for existing sonic identity (.studio/sonic-identity.md)
Parse the argument to identify sound category:
Clarify exactly what sound is needed:
Ask: "Let me understand the sound you need. You mentioned '[argument]'. Can you describe:"
Categorize into archetype:
Understand where the sound exists:
Ask: "Where does this sound occur in your game?"
Spatial context:
Frequency context:
Priority: Based on sonic identity if exists, or ask: "How critical is this sound for gameplay?"
Define how the sound should feel.
Based on sonic identity style, or ask:
"How should this sound feel?"
For weapons/impacts:
For movement:
For UI:
"What material quality?" (if applicable)
Break down into component layers.
Consult sfx-design:references/layering-recipes.md for category-specific layer templates (impacts, movement, UI, weapons, creatures, ambient).
Present layer template based on sound category. Standard approach uses 3-4 layers:
Ask user to confirm or modify layer approach.
For each layer, define creation method:
Method options:
Consult sfx-design:references/material-signatures.md for material-specific frequency/envelope parameters.
For synthesis layers, specify: Waveform, Envelope (ADSR), Filter (type/cutoff/Q), Effects.
Define randomization for repeated sounds:
Ask: "This sound may play [frequency]. Should it have variations?"
Consult sfx-design:references/variation-strategies.md for variation types (pre-made, pitch, volume, layer, parameter) with implementation patterns.
For frequent sounds (footsteps, hits), recommend at least pitch + volume randomization (±2 semitones, ±2 dB).
Present the SFX specification in the conversation. Do NOT write to a file - the specification stays in context for immediate use by sfx-architect.
Present this specification:
# SFX Specification: [Sound Name]
## Overview
- **Category:** [Impact/Movement/UI/etc.]
- **Trigger:** [What causes this sound]
- **Context:** [Where it plays]
- **Priority:** [1-5]
- **Duration:** [Approximate length]
## Character
- **Feel:** [Description]
- **Weight:** [Light/Medium/Heavy]
- **Material:** [If applicable]
- **Reference:** [Similar sounds in other games]
## Layers
### Layer 1: [Name]
| Property | Value |
|----------|-------|
| Purpose | [What this provides] |
| Method | [Synthesis/Sample/Hybrid] |
| Duration | [Length] |
| Frequency range | [Low/Mid/High] |
**Synthesis details (if applicable):**
- Waveform: [Type]
- Envelope: A:[X]ms D:[X]ms S:[X] R:[X]ms
- Filter: [Type] @ [Frequency], Q=[X]
- Effects: [List]
**Sample details (if applicable):**
- Source: [Description of what to record]
- Processing: [Any processing needed]
### Layer 2: [Name]
[Same structure...]
### Layer 3: [Name]
[Same structure...]
## Mix
### Layering
| Layer | Volume | Pan | Notes |
|-------|--------|-----|-------|
| 1 | 0 dB | C | Primary transient |
| 2 | -3 dB | C | Body layer |
| 3 | -6 dB | C | Tail/ambience |
(Fill in actual values based on the designed layers)
### Master Processing
- Compression: [Settings if any]
- EQ: [Adjustments if any]
- Limiting: [Peak target]
## Variations
### Random Parameters
| Parameter | Range | Notes |
|-----------|-------|-------|
| Pitch | ±2 semitones | Natural variation |
| Volume | ±1 dB | Subtle dynamics |
(Adjust ranges based on user preferences from Phase 6)
### Pre-made Variations (if applicable)
- Variation A: [Description]
- Variation B: [Description]
- Variation C: [Description]
## Implementation
### Spatial
- **Mode:** [2D/3D/Both]
- **Attenuation:** Min [X]m, Max [X]m
- **Curve:** [Linear/Log/Custom]
### Playback
- **Polyphony:** [Max instances]
- **Priority:** [Steal behavior]
- **Bus:** [Which audio bus]
### Memory
- **Estimated size:** [KB]
- **Load type:** [Resident/Streaming/OnDemand]
## Technical Notes
- [Any special considerations]
- [Integration specifics]
- [Testing notes]
---
*Generated by sound-design plugin*
After presenting, summarize and offer next steps:
"SFX Design Complete!
Summary:
Ready to synthesize?
The sfx-architect agent can create a .spec.py file with synthesis code from this design.
Would you like me to:
If user wants synthesis code: Use Task tool with subagent_type: "sound-design:sfx-architect" Pass the full specification from this conversation as context.