
Team Polish Orchestrator
Coordinates a team of AI specialists to profile, optimize and harden a game feature or area
- Repository
- Donchitos/Claude-Code-Game-Studios
- Stars
- 32,965
- Licence
- MIT
- Files
- 1 file
What it does
When invoked with a feature or area name, this skill makes the agent run a structured multi-phase workflow in which specialist roles (performance analyst, technical artist, sound designer, QA tester, and sometimes engine and tools programmers) each examine the area, propose and apply polish and optimization work, and then report results. It asks the user to approve choices at phase transitions, and it reads a configuration file to decide how many specialists are active and how much review to apply. Without an argument it only prints usage guidance.
Great for
- Game developers polishing a specific feature such as combat, a menu or a level before release
- Teams using the Claude Code Game Studios template who want a repeatable polish pass
- Developers who want profiling, optimization and edge-case testing organised into one process
Skip it if
- Non-game software, since the roles and vocabulary are built around game studio workflows
- Users who want a quick one-off fix, because the pipeline spawns several subagents and is token-heavy
- Running unattended on a live project without review, because it can edit files and may apply changes automatically in autonomous mode
Try asking
- “/team-polish combat --review lean”
- “/team-polish main menu”
- “/team-polish inventory system --review full”
- “/team-polish level-1”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
team-polish.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip team-polish.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip team-polish.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Pass a specific feature or area name so the team has a clear scope; the usage message appears if you omit it
- Use --review lean or solo for smaller runs to limit token use and the number of spawned agents
- Read the 'Active set' announcement before Phase 1 to see which agents will actually run at your team.size setting
What's inside
- SKILL.md
Redistributed unmodified from Donchitos/Claude-Code-Game-Studios under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.