| name | use-this-repo |
| description | Help a user understand this repository, decide which AI structure pieces they need, and adapt the template to their own project. |
Use this repo
Use this skill when the user wants to:
- understand what this repository is for
- learn how the Claude, GitHub, and shared skill layers fit together
- decide which files and folders to copy into their own project
- adapt this repository structure to a real project
- create a small adoption plan from this repository
What this repository is
This repository is a practical learning repo for AI-enabled project structure.
It demonstrates:
- shared agent instructions in
AGENTS.md
- Claude-specific project guidance in
CLAUDE.md
- internal Claude rules, agents, and skills in
.claude/
- internal GitHub/Copilot instructions, agents, and skills in
.github/
- external/shared user-facing skills in
skills/
Workflow
- Read
references/repo-map.md.
- Run
scripts/quickstart-check.sh to confirm the expected starter files are present.
- Identify the user's goal:
- learn the structure
- copy the structure
- simplify the structure
- adapt the structure to an existing repo
- Explain the repository in this order:
- root files
.claude/
.github/
skills/
- Use
assets/adoption-plan-template.md to produce a practical recommendation.
Output requirements
Always:
- explain this repository using the real folders and files that exist
- separate internal maintainer AI from external/shared user-facing AI
- recommend the smallest useful structure first
- keep the explanation concrete and specific to this repository
When suggesting adoption:
- tell the user what to copy first
- tell the user what can wait until later
- explain why each piece exists
- avoid suggesting extra folders unless they solve a real need
Default recommendation order
For most users, recommend this adoption order:
README.md
AGENTS.md
CLAUDE.md
.claude/
.github/
skills/
Important distinction
.claude/ = internal Claude setup for maintainers
.github/ = internal GitHub/Copilot setup for maintainers
skills/ = external/shared skills for users and outside agents
Do not mix those layers together.