Database Schema

Stops Claude from hallucinating column names mid-session by enforcing a simple rule: read the schema file before touching any database code. Works with Drizzle, Prisma, Supabase, SQLAlchemy, whatever. You get a checklist to verify columns exist, types match, and relationships are correct before writing queries. The real win is combining this with generated types so TypeScript or mypy catches schema mismatches at compile time instead of your tests catching them at runtime. Includes a schema reference template for quick lookups and stack-specific type generation commands. Medium effort because you need to actually maintain that schema file, but it pays off when you're not debugging "column does not exist" errors.

skills.sh installs: 200
GitHub stars: 685
last updated: Jun 3, 2026

View on GitHub

Install

$ npx -y skills add alinaqi/claude-bootstrap --skill database-schema --agent claude-code

Installs into .claude/skills of the current project.

Files

Database Schema Awareness Skill

Problem: Claude forgets schema details mid-session - wrong column names, missing fields, incorrect types. TDD catches this at runtime, but we can prevent it earlier.

Core Rule: Read Schema Before Writing Database Code

MANDATORY: Before writing ANY code that touches the database:

┌─────────────────────────────────────────────────────────────┐
│  1. READ the schema file (see locations below)              │
│  2. VERIFY columns/types you're about to use exist          │
│  3. REFERENCE schema in your response when writing queries  │
│  4. TYPE-CHECK using generated types (Drizzle/Prisma/etc)   │
└─────────────────────────────────────────────────────────────┘

If schema file doesn't exist → CREATE IT before proceeding.