team-audio

Invoke the skill with a feature or area argument such as combat or boss encounter. It spawns audio-director, sound-designer, technical-artist, and gameplay-programmer subagents in sequence, asking for your approval at each decision point before producing a final audio design document.

It compiles all subagent outputs into an audio design document saved at design/audio/audio-[feature].md. The summary includes audio event count, estimated asset count, implementation tasks, and open questions between team members.

Yes. The technical-artist subagent designs middleware integration for Wwise, FMOD, or native audio, and an engine specialist validates the approach against the configured engine, such as Unity Audio Mixer, Godot AudioStreamPlayer, or Unreal MetaSounds.

An accessibility-specialist subagent runs in parallel with sound design to identify audio events carrying critical gameplay information, specify subtitle and caption requirements, and ensure no gameplay state is communicated by audio alone.

The skill surfaces the blocker immediately, checks whether later phases depend on it, and offers options to skip, retry with narrower scope, or stop. It always produces a partial report rather than discarding completed work.

Yes. Pass --review with a mode or set production/review-mode.txt. The lean mode keeps only phase-gate reviews, while solo mode skips all director gates and runs the pipeline without agent gates.