4/ Curious: do you have a playbook like this in your team/company? And do you feed it to your AI tools?

3/ Also: we migrated from one giant Markdown file → Astro Starlight docs + added llms.txt + llms-full.txt so AI tools can consume it cleanly.

2/ Why open source it:

- we work in hybrid teams → one public URL is just easier to share
- in the era of agentic coding, it’s a cheat code: give your agents the playbook as context

1/ What changed for us:

- faster onboarding (expectations are explicit)
- fewer “opinion vs opinion” review threads
- consistent foundation (we also have an internal starter template around it)

I've been thinking about going the other direction: creating an ESLint preset that enforces immutable patterns instead. Might be a nice lightweight alternative, but I haven't built it yet - for now, we just encourage these patterns in our internal style guide.

This is super cool! I honestly never thought about overriding the built-ins like that. It’s a powerful approach, though it might feel a bit "too much" for some teams, and polishing the last missing bits for seamless production use might be tricky. Still, love the idea.

Thanks for the shout-out! Really appreciate it.

Yeah, using string literals as types is pretty powerful