Make repository AI Agent friendly

[apalchys]

Describe the feature
AI agents (VS Copilot, Cursor, Claude Code, Codex, etc.) are becoming an essential part of the development process. We need to adapt our repository for AI agents.

Expectation

A developer asks the AI agent to fix an issue from GitHub by providing a link, and the AI agent creates a merge request with the fix and follows our guidelines.

Input:

please fix https://github.com/rolling-scopes/rsschool-app/issues/<id>

Output:

<link to merge request with fix>

What to do

Here’s how we can achieve this: We need to treat AI agents like real remote coworkers living in a timezone with a 12-hour difference who have access only to our repository. All required context must be preserved inside the repository in some way.

I expect the following first steps:

• Create AGENTS.md. Links:
  • https://github.blog/ai-and-ml/github-copilot/how-to-write-a-great-agents-md-lessons-from-over-2500-repositories/
  • https://www.humanlayer.dev/blog/writing-a-good-claude-md
• Standardize tooling across the repository as much as possible: client, server, and NestJS should share the same commands, test configuration, formatters, lint output, etc. (it’s not bad now but needs review)
• Identify custom solutions that don’t add value and remove them
• Document the data model
• Review onboarding documents and make them clear (CONTRIBUTING.md, etc.)
• Use TypeScript types as much as possible (e.g., consider using branded types: https://egghead.io/blog/using-branded-types-in-typescript)
• Find GitHub Actions that can proactively enrich issue context based on the codebase

This issue is a good starting point for people who are interested in AI and new to the RS App repository. A fresh look may help identify missing information or context and enrich our repository with all the required information.