Database Schema Scanner

The Database Schema scanner extracts structural facts — table names, column names and types, relation and key flags — from your project's schema definitions. It's one of a handful of extended scanners (alongside Code Quality and Breaking Change Exposure, plus the code graph) that opens schema and ORM source files rather than manifests alone. Read exactly what it reads and never reads below — see scanners-overview for how this compares across scanners.

What It Reads

For Drizzle and TypeORM, this means the scanner opens plain TypeScript source files — not just config or manifests — looking specifically for ORM table/entity definitions.

What It Never Reads

Configuration

Enabled by default, alongside every other extended scanner:

// vibgrate.config.ts
const config: VibgrateConfig = {
  scanners: {
    databaseSchema: { enabled: true },
  },
};

Disable it like any other scanner:

const config: VibgrateConfig = {
  scanners: {
    databaseSchema: { enabled: false },
  },
};

A large monorepo's schema is capped before upload (300 models / 100 fields per model / 500 scanned files, by default) so it can't dominate the payload. Raise or lower those caps per repo:

const config: VibgrateConfig = {
  scanners: {
    databaseSchema: {
      enabled: true,
      maxModels: 500,
      maxFieldsPerModel: 200,
    },
  },
};

Vibgrate Cloud enforces its own fixed upper ceiling regardless of what a repo configures, so raising these only ever narrows the gap to that ceiling.

Availability

The scan itself is local and free, like every extended scanner — running vg or vg scan collects this data on your own machine at no cost. It only leaves your machine when you run vg push or scan with a DSN configured. Viewing a repository's schema in the Vibgrate Cloud dashboard's Database Schema tab requires a Vibgrate Team plan or higher — see vibgrate.com/pricing.

Example Output

```json { "databaseSchema": { "providers": ["postgresql"], "models": [ { "name": "User", "source": "prisma", "files": ["prisma/schema.prisma"], "fields": [ { "name": "id", "type": "String", "isId": true, "isUnique": false, "isList": false, "isOptional": false, "isRelation": false }, { "name": "email", "type": "String", "isId": false, "isUnique": true, "isList": false, "isOptional": false, "isRelation": false }, { "name": "posts", "type": "Post", "isId": false, "isUnique": false, "isList": true, "isOptional": false, "isRelation": true } ] } ], "enums": [ { "name": "Role", "values": ["ADMIN", "MEMBER"] } ], "filesScanned": ["prisma/schema.prisma"] } } ```

Use Cases

• Migration planning — see every table and column a service depends on before splitting a monolith or moving databases.
• Schema drift detection — catch a model or column that changed without a corresponding migration file.
• Onboarding — give a new engineer a structural map of the data model without pulling database credentials.