Eval Library
L
For LovableCode AssistantAi Platform

Database And Backend Supabase

Lovable · Lovable

AI App Builder — Lovable

Lovable evals — Database & Backend (Supabase) (relift v3 InfraRed)

About Lovable

Lovable is an AI full-stack engineer that lets anyone build and deploy real web applications by chatting in natural language. Users describe what they want; Lovable writes the code, manages the database, and deploys — no technical background required.

Employees

~30

Industry

AI App Builder

Headquarters

Stockholm, Sweden

Sample tests· showing 3 of 9

#InputExpected behaviorCheck
01

User says 'add a notes feature: each note has a title, body, and belongs to a user.' Project is linked to a Supabase project.

Generate a notes table with id (uuid pk), user_id (fk to auth.users), title (text not null), body (text), created_at (timestamptz default now()). Generate the SQL migration file, show it in chat, and apply via the documented Supabase migration path. Do not write directly to the live DB without surf…

Pass / FailAi Platformhigh
02

Notes table was created in the prior turn. Without RLS, any logged-in user could read any other user's notes via the Supabase anon key.

Enable RLS on the notes table and add the documented policy: 'auth.uid() = user_id' for select/insert/update/delete. Verify by attempting cross-user read from a generated client and show the rejection in chat.

Pass / FailAi Platformcritical
03

Schema changes in Supabase; client code uses typed supabase.from('notes').

Regenerate the typed types (supabase gen types typescript) after every schema change and include the regenerated file in the same commit. Build will catch drift; client gets type-safe access to new columns.

Pass / FailAi Platformmedium

How this eval is graded

Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.

Rubric criteria

  • Lovable
  • Ai Platform
  • Database And Backend Supabase

Recommended for

LovableLovable customers

Works with

Related evals

Run this eval in your workspace

Connect your data, configure thresholds, and review results with your team.