What it does

Flags WHERE clauses that filter on columns that do not have an index defined in the database schema.

Why it matters

Filtering on unindexed columns causes full-table scans, degrading query performance as the table grows. Adding an appropriate index can dramatically speed up queries.

How it works

The rule loads the Prisma schema, finds the table referenced in the query, extracts the column names used in the WHERE clause, and checks each column against the schema indexes (@id, @unique, @@index).

When to disable

Disable this rule for small tables where the cost of an index outweighs the query performance benefit, or for tables that are only queried occasionally.

Examples

// Assume `users.name` has no index in the schema
const user = await db
  .select({ id: users.id })
  .from(users)
  .where(eq(users.name, "test")) 
  .limit(1);

// Assume `users.email` is marked @unique in the schema
const user = await db
  .select({ id: users.id })
  .from(users)
  .where(eq(users.email, "test@test.com")) 
  .limit(1);