Skip to content
Get packs

Guide · Agent skill

How to use the Test Setup Scaffolder agent skill (test-setup)

Updated · written with AI from the project's own documentation and checked against our catalogue data

The test-setup skill is an agent skill for Claude Code that scaffolds a testing foundation for a project: a tests/ directory layout, an engine test runner configuration, and a GitHub Actions workflow so tests run on every push. It is aimed at developers starting a game project in the Claude Code Game Studios setup, or any team that has not yet set up automated testing. After this guide you will know what the skill does, how to install it, how to trigger it, and what it needs before it runs.

Test Setup ScaffolderDonchitos/Claude-Code-Game-Studios · MIT

Quick start

  1. 1Download and unzip. Download the test-setup ZIP from this page and unzip it into ~/.claude/skills/ or your project's .claude/skills/ folder.
  2. 2Configure your engine. Run /setup-engine in your project so the engine is recorded in project.yaml or technical-preferences.md.
  3. 3Start the skill. Run /test-setup, or ask the agent in plain language to set up tests and CI before the first sprint.
  4. 4Review the plan. Read the Phase 2 plan the agent shows, which lists the directories and files it intends to create.
  5. 5Approve the file creation. Approve the plan so the agent creates only the missing files; existing test files are left alone.
  6. 6Add secrets if needed. For Unity, add the UNITY_LICENSE secret to your repository manually before CI can run.
  7. 7Check the summary. Read the final summary and confirm the tests/ layout, workflow file and smoke test are present.

What is an agent skill and what does test-setup do?

An agent skill is a folder containing a SKILL.md file and supporting reference files. When installed, the agent reads these instructions and follows them for a specific task instead of improvising. The test-setup skill is one of these, and its job is one-time setup of testing infrastructure.

When it runs, the agent detects the configured engine, proposes a plan, creates a tests/ directory layout, writes a GitHub Actions workflow at .github/workflows/tests.yml, and adds a small smoke test. The skill is designed to run once early in a project, before implementation begins, and it only creates files that are missing. It never overwrites existing test files.

  • Detects the configured engine and existing test state
  • Presents a plan and waits for your approval before creating files
  • Creates tests/ with unit, integration and smoke folders (Godot layout; Unity and Unreal use their engine test roots)
  • Writes .github/workflows/tests.yml so tests run on every push
  • Adds a smoke test seed and a summary of what was created

Install the test-setup skill

Download the ZIP for the test-setup skill from the item page on this directory. Then choose one of two routes depending on how you work.

For a personal setup available across projects, unzip the download into ~/.claude/skills/. For a single project, unzip it into that project's .claude/skills/ folder so it travels with the repository. If you use the Claude apps instead, you can upload the ZIP under Settings → Capabilities → Skills.

  • Personal: unzip into ~/.claude/skills/
  • Project only: unzip into your project's .claude/skills/
  • Claude apps: upload the ZIP under Settings → Capabilities → Skills

Before you run it: requirements

The skill needs a project with a configured engine. Set this up first with /setup-engine. If no engine is configured, the skill stops and tells you to run /setup-engine rather than guessing.

You also need Claude Code or a compatible SKILL.md agent, and a Git host with GitHub Actions for the CI workflow. Bash and grep must be available for the detection commands the skill is allowed to run.

  • Claude Code or a compatible SKILL.md agent
  • A project with a configured engine (via /setup-engine)
  • GitHub Actions for the CI workflow
  • For Unity projects: a UNITY_LICENSE secret added manually to the repository
  • Bash and grep available for the allowed detection commands

How do you trigger the skill with a natural request?

You can invoke the skill with the slash command /test-setup. The skill also accepts an optional argument, force. Using force skips the 'already exists' early exit, so the agent creates any missing files even when the test directory is already there. It still never overwrites existing files.

Natural requests should name the goal, for example asking the agent to set up tests and CI for the project before the first sprint. The agent will then detect the engine and show you its plan before it writes anything.

  • /test-setup
  • /test-setup force (creates missing files even if tests/ exists, never overwrites)
  • Natural request: set up the test framework and CI for this project before the first sprint
/test-setup

What happens during a run?

The skill works in phases. It first reads the engine from project.yaml, falling back to the Engine value in .claude/docs/technical-preferences.md. It checks for existing tests/ folders and a workflow file. Then it presents the plan, and only after you approve does it create the directory structure, the CI workflow, and the smoke test seed. A final summary lists what was done.

For Unity projects the unit and integration tests go under Assets/Tests/ (EditMode and PlayMode) instead of tests/. For Unreal they go under Source/<Module>/Private/Tests/. For Unity CI, you must add the UNITY_LICENSE secret yourself; the skill does not manage license secrets.

  • Phase 1: detect engine and existing state
  • Phase 2: present plan and wait for approval
  • Phase 3: create directory structure and tests/README.md
  • Phase 4: create the CI workflow
  • Phase 5: create the smoke test seed
  • Phase 6: post-setup summary

Safety notes: scripts and network

The skill runs scripts: it uses a small set of allowed Bash commands, such as a grep check for the Unity test framework in Packages/manifest.json and a helper script that resolves configuration keys. It does not use the network. The skill is restricted to the tools it lists, including Read, Glob, Grep and Write.

Because it writes files, review the plan before approving. The skill's collaborative rules are to ask before creating files and never overwrite existing test files, but checking the plan and the resulting diff is still good practice.

  • Runs scripts: yes (limited to the listed Bash commands)
  • Uses network: no
  • Review the Phase 2 plan before approving file creation
  • Do not add secrets on the agent's behalf; add UNITY_LICENSE yourself

What people use it for

  • Starting a new game project in the Claude Code Game Studios setup with tests and CI from day one
  • Giving a team that has no automated testing a consistent starting layout
  • Adding a GitHub Actions workflow so tests run on every push
  • Creating a smoke test list for a critical path check before a sprint gate
  • Preparing a Unity or Unreal project with the engine's own test folders

Frequently asked questions

How do I install the test-setup skill?

Download the ZIP from the item page and unzip it into ~/.claude/skills/ for personal use or into a project's .claude/skills/ folder for one project. In the Claude apps you can instead upload the ZIP under Settings → Capabilities → Skills.

What do I need to run test-setup?

You need Claude Code or a compatible SKILL.md agent, a project with a configured engine set up via /setup-engine, and GitHub Actions for the CI workflow. Bash and grep must also be available for the allowed detection commands.

Is the test-setup skill free?

The source describes the skill as licensed under MIT. Check the repository's licence file for the full terms before using it.

Will test-setup overwrite my existing tests?

No. The skill is designed to create only missing files and never overwrite existing test files. The force argument skips the already-exists early exit but still does not overwrite.

What if the skill says the engine is not configured?

Run /setup-engine first so the engine is recorded, then re-run /test-setup. The skill is designed to stop and redirect rather than guess the engine.

Should I use test-setup on a project that already has tests and CI?

The skill is meant for initial setup and could create conflicting files in a project that already has a working test suite and CI, so it is not recommended for that case.

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

Use test-setup if you are starting a game project in the Claude Code Game Studios setup, or a team that has no automated testing yet and wants a consistent layout and CI workflow from the start. Pick something else if your project already has a working test suite and CI, or if you cannot configure an engine first, since the skill is built to stop in that case.

Download the skill