Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This 73-skill collection translates a professional game production pipeline into a structured, AI-assisted workflow. It guides indie developers, designers, producers, and technical leads through every phase: brainstorming and prototyping, system design and architecture, sprint planning and implementation, testing and QA, localization and security auditing, and finally launch readiness and live operations. Each skill is designed so a small team can cover every role from creative director to release manager without the full studio staff.
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Follow the documented setup, then try a first task.
Clone the complete game-studio template
Installs: Complete Claude Code Game Studios template bundle with 49 agents, 73 skills, hooks, rules, documentation, and project files. skillPaths is a bounded selection prioritizing the starter tasks; other documented skills are also included. · Claude Code
Before you start
Git
Claude Code
Optional for additional hook validation: jq and Python 3
In a terminal, clone the template and enter its directory. Replace `my-game` in both lines with your desired project-directory name, or leave it unchanged to use `my-game`.
git clone https://github.com/Donchitos/Claude-Code-Game-Studios.git my-game
cd my-game
From the cloned project directory in the terminal, start Claude Code.
claude
Inside the Claude Code session, enter `/start`. This is an agent-session command, not a terminal command; it asks about the project's current state and guides you to the appropriate workflow.
Use the asset-audit skill to audit the art assets under `assets/art/`. Read the project's CLAUDE.md and relevant art or asset standards first, then report naming, size, format, orphaned-reference, and missing-reference issues with prioritized recommendations. Keep this read-only and do not modify or delete any assets.
A specific task description replacing `[replace with a specific development task]`
CLAUDE.md project context
Relevant design documents or code, if available
Past sprint data, if available
Use the estimate skill for this task: `[replace with a specific development task]`. Inspect the relevant project context, design documents, affected code, tests, dependencies, and available sprint history. Return optimistic, expected, and pessimistic estimates, a confidence level, the largest risk, and a suggested task breakdown. Keep the analysis read-only.
Use the playtest-report skill in `new` mode to produce a blank playtest report template for an upcoming session. Present the template in the conversation only; do not save it to a file.
Claude