Supabase Core Skill Core concepts, CLI workflow, and patterns common to all Supabase projects. Sources: [Supabase Docs](https://supabase.com/docs) [Supabase CLI](https://supabase.com/docs/guides/local development/cli/getting started) Core Principle Local first, migrations in version control, never touch production directly. Develop locally with the Supabase CLI, capture all changes as migrations, and deploy through CI/CD. Supabase Stack Service Purpose Database PostgreSQL with extensions Auth User authentication, OAuth providers Storage File storage with RLS Edge Functions Serverless Deno functions Realtime WebSocket subscriptions Vector AI embeddings (pgvector) CLI Setup Install & Login Initialize Project Link to Remote Start Local Stack Migration Workflow Option 1: Dashboard + Diff (Quick Prototyping) Option 2: Write Migrations Directly (Recommended) Option 3: ORM Migrations (Best DX) Use Drizzle (TypeScript) or SQLAlchemy (Python) see framework specific skills. Deploy Migrations Database Patterns Enable RLS on All Tables Common RLS Policies Link to auth.users Seed Data supabase/seed.sql Environment Variables Required Variables Local vs Production Connection Pooling Edge Functions Create Function Basic Structure With Auth Context Deploy Storage Create Bucket (in migration) CLI Quick Reference CI/CD Template Anti Patterns Direct production changes Always use migrations Disabled RLS Enable on all user data tables Service key in client Never expose service role key No connection pooling Use pooler for serverless Committing .env Add to .gitignore Skipping migration review Always check generated SQL No seed data Use seed.sql for consistent local dev