Open source

The database toolkit built for PostgreSQL.

One binary. Pure SQL. Migrations, seeds, models, snapshots, and deep introspection. No framework required.

$ cargo install pgcrate

Rust not installed? brew install pgcrate coming soon.

Everything you need

One tool, complete workflow

Migrations

Pure SQL with up/down sections. Timestamp-based ordering. Baseline support for brownfield adoption.

$ pgcrate up

Seeds

Load reference data from CSV or SQL. Type inference. Diff against database state.

$ pgcrate seed run

Models

SQL transformations with DAG ordering. Views, tables, incrementals. Built-in data tests.

$ pgcrate model run

Introspection

Deep table inspection with constraints, indexes, RLS policies. Role and grant inspection.

$ pgcrate describe users

Snapshots

Save and restore database state. Profile-based selective dumps. Great for testing.

$ pgcrate snapshot save dev

Anonymization

Export production data safely with TOML-based rules. Mask emails, fake names, redact PII.

$ pgcrate anonymize dump

No framework required

Pure SQL migrations

No DSL. No magic. Just SQL you already know. Write your schema changes in plain SQL with up/down sections.

Timestamp-based ordering
Baseline for brownfield adoption
Generate from existing database

20250101_create_users.sql

-- up
CREATE TABLE users (
    id SERIAL PRIMARY KEY,
    email TEXT UNIQUE NOT NULL,
    created_at TIMESTAMPTZ DEFAULT now()
);

-- down
DROP TABLE users;

models/marts/user_stats.sql

-- materialized: table
-- deps: staging.stg_users, staging.stg_orders
-- tests: not_null(user_id), unique(user_id)

SELECT
    u.id AS user_id,
    u.email,
    COUNT(o.id) AS order_count,
    SUM(o.total) AS lifetime_value
FROM staging.stg_users u
LEFT JOIN staging.stg_orders o ON u.id = o.user_id
GROUP BY u.id, u.email

Transform your data

SQL models with DAG ordering

Define SQL transformations with dependency tracking. pgcrate runs them in the right order, every time.

Views, tables, and incrementals
Built-in data tests
Auto-generated documentation

Understand your database

Deep introspection

See everything about your tables, roles, and permissions. Compare schemas across environments. Debug RLS policies.

Table structure, indexes, constraints
RLS policies and grants
Schema diff between databases

pgcrate describe users

Table: public.users
Columns:
  id         serial PRIMARY KEY
  email      text NOT NULL UNIQUE
  created_at timestamptz DEFAULT now()

Indexes:
  users_pkey        PRIMARY KEY (id)
  users_email_key   UNIQUE (email)

RLS: enabled
  tenant_isolation  USING (tenant_id = current_tenant())

Production ready

Built for CI/CD

JSON output on every command. Meaningful exit codes. Production safeguards that prevent accidents.

--json flag on all commands
Exit codes: 0=ok, 1=action needed, 2=error
--yes required for destructive ops

ci.yml

# Check migration status
pgcrate status --json

# Run health checks
pgcrate doctor --json

# Apply migrations
pgcrate up --json