Guide · Agent skill
How to use the Team Audio Orchestrator agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
Team Audio is an agent skill for Claude Code, from the Claude-Code-Game-Studios project, that runs a structured audio design process for one game feature, such as combat or a boss encounter. It is for game developers who already keep design documents in design/gdd/ and want a plan before they build the audio. After this guide you will know what the skill does, how to install it, how to trigger it with a natural request, and what it will and will not produce.
Team Audio OrchestratorDonchitos/Claude-Code-Game-Studios · MIT Download the skill Quick start
- 1Download the skill. Download the team-audio ZIP from the page and confirm it contains SKILL.md.
- 2Install it. Unzip into ~/.claude/skills/ for all projects, or into a project's .claude/skills/ folder, or upload it under Settings, Capabilities, Skills in the Claude apps.
- 3Prepare your design docs. Make sure your game design documents are in design/gdd/, and add design/audio/sound-bible.md if you have one.
- 4Run the skill with a feature. Invoke it with a feature or area such as combat or boss encounter, optionally adding --review full, lean or solo.
- 5Pick the automation mode. Choose collaborative, guided or autonomous behavior so you know how often the skill will stop and ask.
- 6Answer the decision points. Choose from the proposals the agents present at each step, so the chosen direction carries into the next step.
- 7Review the files. Open the working files under production/audio/ and use the specifications as a checklist for implementation.
What is an agent skill and what does team-audio do?
An agent skill is a folder containing a SKILL.md file with instructions that Claude follows when a matching request comes in. The team-audio skill tells the agent to behave like a small audio team. It reads your existing game design documents and any sound bible, then coordinates sub-agents that play four roles: audio-director, sound-designer, technical-artist and gameplay-programmer.
Together these roles move from direction through implementation. They propose a sonic identity, sound effect specifications, an audio bus structure and integration patterns for the feature you name. At decision points the skill pauses and asks you to pick from the proposals.
- Output is plans and specifications, not audio files or music.
- It reads design/gdd/, design/audio/sound-bible.md and assets/audio/ when they exist.
- It can run in different automation modes, described below.
Install the team-audio skill
Download the ZIP for the skill from the page you found it on. The skill's file is SKILL.md inside the team-audio folder. There are two ways to install it.
For Claude Code, unzip the download into ~/.claude/skills/ to make it available in all your projects, or into the .claude/skills/ folder of a single project to limit it to that project. For the Claude apps, open Settings, go to Capabilities, and upload the skill under Skills.
- Claude Code, all projects: unzip into ~/.claude/skills/
- Claude Code, one project: unzip into the project's .claude/skills/ folder
- Claude apps: Settings → Capabilities → Skills, then upload
Check what the skill needs before you run it
The skill is designed for Claude Code used with the Claude Code Game Studios agent set and its hooks folder. It also expects a bash shell, because its config step runs the yaml-helper.sh script. Without the hooks folder the config resolution will not work.
Your repository files are optional but improve the result. The skill reads design/gdd/ for the feature's design, design/audio/sound-bible.md for audio direction, and assets/audio/ for existing audio assets. If there is no sound bible, the skill says so and starts from the GDDs alone. If there is little design context, the output will start from less information.
- Claude Code with the Claude Code Game Studios agent set and hooks folder
- A bash shell for yaml-helper.sh
- Recommended: design documents in design/gdd/
- Optional: design/audio/sound-bible.md and assets/audio/
Trigger the skill with a feature or area
The skill takes an argument naming the feature or area to design audio for. The documented argument form is a feature or area followed by an optional review depth flag. Examples from the skill include combat, main menu, forest biome and boss encounter.
If you invoke the skill with no argument, it prints usage guidance and stops without starting any agents. The review flag accepts full, lean or solo and overrides the review setting in your config.
- Argument form: [feature or area] [--review full|lean|solo]
- Example areas: combat, main menu, forest biome, boss encounter
- No argument: the skill shows usage and exits
/team-audio boss encounter --review leanChoose the automation mode and team size
The skill's decision points work differently depending on the automation setting. In collaborative mode, you must approve each step before the pipeline moves on. In guided mode it advances automatically unless a step is blocked. In autonomous mode it runs end to end and records each step outcome. Decisions in categories listed as always-ask still prompt you in every mode.
The team.size setting controls how many agents are active. The default individual uses only the sound-designer, and the other roles are consulted through it. small adds the audio-director, technical-artist and gameplay-programmer. studio adds the accessibility-specialist and the primary engine specialist. Before it starts, the skill announces which agents it will run and which it will not.
- collaborative: you approve each step
- guided: advances automatically unless blocked
- autonomous: runs end to end and logs each step
- individual (default): sound-designer only
- small: adds audio-director, technical-artist, gameplay-programmer
- studio: adds accessibility-specialist and the engine specialist
Read the output and work with the results
The skill writes working artifacts under production/audio/ in a folder named after the feature. The agents return a short summary to the session, while their full analysis goes to those files. Expect a pipeline that can use substantial tokens, so it suits a deliberate planning session rather than a quick question.
Use the outputs as a checklist for implementation. Sound effect specifications, audio event lists and bus structure proposals give your team a concrete starting point. You still need to create or source the actual sounds.
- Working files are written under production/audio/[feature]/
- Each step ends in a choice you make from the proposals
- Treat the output as a plan to review, not a finished audio spec
What people use it for
- Planning the audio for a boss encounter, including the sonic identity and the sound effects it needs.
- Drafting a sound effect specification and audio event list for a combat system from existing design docs.
- Proposing an audio bus structure and integration patterns for a forest biome before implementation starts.
- Creating an audio design checklist for a small indie team before the audio programmer begins work.
Frequently asked questions
Is the team-audio skill free to use?
The source does not state pricing or licence terms beyond the licence listed for the repository, which is MIT. Check the repository's licence file for the terms that apply to you.
What do I need to run the team-audio skill?
You need Claude Code with the Claude Code Game Studios agent set and hooks folder, plus a bash shell for the yaml-helper.sh script. Design documents in design/gdd/ are recommended, and the sound bible and assets/audio/ are optional.
Does team-audio generate actual sound effects or music?
No. The skill produces plans and specifications, including sonic identity, sound effect specs, audio event lists and bus structure. You will need to create or source the audio files yourself.
How do I install team-audio in Claude Code?
Download the ZIP from the page, then unzip it into ~/.claude/skills/ to use it in all projects, or into a project's .claude/skills/ folder to use it in one project. In the Claude apps, upload it under Settings, Capabilities, Skills.
Why does team-audio say it needs design documents?
The skill is built to read the design documents in design/gdd/. Without them it will start from little context, so its proposals will be less specific to your game.
Does team-audio use the network or run scripts?
The skill runs scripts, specifically the yaml-helper.sh config resolution script through bash. The repository lists network use as no. It writes its working files to production/audio/ on your machine.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Team Audio suits game developers who already use the Claude Code Game Studios template and want a structured, checklist-style audio plan for a specific feature built from their design documents. Pick something else if you need finished audio, music, or a quick one-off answer, since the skill produces specifications and can use substantial tokens.
Download the skill