A real PostgreSQL server lives in this app
PostgreSQL 18.2 embedded in the app — offline, no root. Plus eleven more SQL and NoSQL engines, an AI SQL console, a visual ER designer and safe mode.
Postgres, zero config
A real database on the phone, ready in seconds.
- Real ARM postgres binary, no root, no network
- Instant start from a pre-built data directory
- Health-checked over a Unix socket
One client, every engine
Local, cloud and NoSQL without switching UIs.
- SQLite, PostgreSQL, Neon, Supabase, PlanetScale, D1
- MongoDB, Turso, Upstash and Firestore
- SSH tunnel, five SSL modes, connection test
SQL in plain language
Describe it; get SQL with the trade-offs.
- Natural language to SQL with a confidence score
- Warnings and alternative queries
- Matched tables and columns, named parameters
Safe mode and masking
You can't accidentally wipe production.
- Intercepts DELETE without WHERE, DROP, TRUNCATE
- Data masking with six types, auto-detected columns
- Slow-query log, active-query kill, biometric lock
// spec
The spec sheet
plan.free
- ·Embedded Postgres, one database
- ·All eleven engines to connect
- ·SQL console and data viewer
- ·Safe mode and SQL/CSV backup
plan.premium
- +Unlimited databases and queries
- +ER designer and schema diff
- +Mock data, masking, monitoring
- +Scheduled backups and dashboards
// works with
From the blog
// faq
Frequently asked questions
Can you run PostgreSQL on an Android phone?
Yes. PocketCode runs a real embedded PostgreSQL 18.2 server directly on the device (64-bit ARM), so you can create, query, and manage a full Postgres database with no external server and no internet connection.
Which database engines are supported?
12 engines in total. Besides embedded PostgreSQL and local SQLite, you can connect to and manage ten SQL and NoSQL databases in the cloud — including Supabase, Neon, PlanetScale, Turso, MongoDB Atlas and Cloudflare D1 — browsing tables, running queries, and editing rows from one interface.
Can it generate ER diagrams?
Yes. The database manager reverse-engineers your schema into entity-relationship diagrams and exports them in 7 formats, so you can visualize and document how tables relate straight from your phone.
Can you use AI to write SQL?
Yes. The built-in AI assistant turns plain-language requests into SQL and can explain a query's execution plan, helping you write and optimize queries without memorizing exact syntax for every engine.
Does it protect sensitive data?
Yes. It offers data masking with 6 masking types plus scheduled backups (4 kinds to 3 destinations), so production values can be hidden while you work and your data stays recoverable if something goes wrong.