How to use the AI Ready Repo Setup agent skill

What an agent skill is and what ai-ready does

An agent skill is a set of instructions that an AI agent loads when a request matches the skill's description. The skill is a SKILL.md file, and the agent reads it to learn how to handle a task.

This skill is a wrapper. Its SKILL.md tells the agent how to install the upstream ai-ready skill from johnpapa/ai-ready. The upstream skill is roughly 600 lines of detailed instructions that the author updates often. The wrapper keeps the skill discoverable here while the upstream repository stays the source of truth.

• Once the upstream skill is installed, asking the agent to make a repo 'AI-ready' lets it analyze your codebase.
• It can then generate agent instruction files (such as AGENTS.md and copilot-instructions.md), CI workflows, and issue templates.
• The description says the skill also mines PR review patterns and tailors files to your stack.

Install the ai-ready skill in GitHub Copilot CLI

The method documented in the skill file is to run the add command inside GitHub Copilot CLI. This downloads the latest version of the skill to your personal skills directory. Running the command again updates it to the latest version.

This page also describes a ZIP download. You can download the ZIP from this page and unzip it into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder to scope it to one repository. In the Claude apps, you can upload the skill under Settings, then Capabilities, then Skills.

Whichever route you use, check that the skill's files are in place before you rely on it. Your agent needs to be able to load skills, so confirm your tool supports the /skills command.

• Requirement: GitHub Copilot CLI with support for the /skills command.
• Requirement: network access to download the skill from johnpapa/ai-ready.
• Run the install yourself. The skill file says the user must run the install command, and the agent should not run it for you.

Review the skill before you load it

Read the downloaded skill before you use it. The ai-ready skill can generate workflows and configuration files that run in your project, so this step matters. The wrapper's own instructions say to look at the start of the file first.

This wrapper does not run scripts, but it does use the network to fetch the upstream skill. The upstream skill was not reviewed for this page, so read its contents yourself before you accept the files it generates.

• Runs scripts: no (for this wrapper).
• Uses network: yes, to download the skill from johnpapa/ai-ready.
• Inspect the file before loading it, and review any generated CI workflows and config before committing them.

Load the skill and trigger it with a request

After you have reviewed and installed the skill, reload skills so the agent sees it. Then make a plain request. The skill's description lists phrases that should trigger it, such as 'make this repo ai-ready', 'set up AI config', or 'prepare this repo for AI contributions'.

You do not need special syntax. Start the agent in the root of the repository you want to prepare, so it can analyze the codebase that is actually there.

• Good requests: 'make this repo ai-ready', 'set up AI config for this project', 'prepare this repo for AI contributions'.
• If the skill does not respond, reload skills and try the request again.
• Review what the agent proposes before you approve file changes.

Examples of what the agent can produce

Because the skill analyzes your codebase, the output depends on your stack and existing files. Expect the agent to explain what it found and what it plans to create. Read those files before you commit them.

The skill's description names these outputs, so treat them as the expected range rather than a guarantee for every repository.

• AGENTS.md and copilot-instructions.md, describing how AI agents should work in the repo.
• CI workflows that run in your project.
• Issue templates for AI-assisted contributions.
• Files customized to your language and framework choices.

Tips for getting good results

Run the skill on a clean working tree so you can see exactly which files were added or changed. Commit the generated files in a separate commit so they are easy to review or revert.

If you want an option that does not depend on this wrapper, look at the upstream repository johnpapa/ai-ready directly. This page does not compare it with other tools.

• Start with a clean git state so changes are easy to diff.
• Read generated workflows before merging, since they may run in your project.
• Keep the install and the generation as separate steps you control.