Skip to main content
✨Managed Postgres & PostGIS • AI Agent Ready (MCP) • Cloud & Self-Hosted

The Postgres Backend for Modern Apps & AI Agents

Build fast on managed Centia Cloud or self-host with Docker. Get PostgreSQL with PostGIS, OAuth2, Realtime APIs, SDKs, and native MCP support out of the box.

Start in Centia CloudSelf-host with DockerRead the Docs →

Open source (AGPLv3)GitHubPostgreSQL 16 + PostGIS 3.4

import { PasswordFlow, Sql } from '@centia-io/sdk';

// Sign in once — the SDK stores and refreshes tokens for you
await new PasswordFlow({
  host: 'https://api.centia.io',
  clientId: 'your-client-id',
  username: 'your-username',
  password: process.env.CENTIA_PASSWORD,
  database: 'your-database',
}).signIn();

// Parameterized spatial SQL on PostgreSQL + PostGIS
const { data } = await new Sql().exec({
  q: `select name, st_asgeojson(geom) as location
      from places
      where st_dwithin(geom::geography,
        st_setsrid(st_makepoint(:lng, :lat), 4326)::geography, :radius)`,
  params: { lng: 12.5683, lat: 55.6761, radius: 2000 },
});

From Schema to Production in Minutes

Everything you need to ship full-stack backends without infrastructure complexity.

01

Deploy in Seconds

Launch in managed Centia Cloud with zero infrastructure setup, or spin up self-hosted with Docker Compose on your own servers.

Explore Docker setup →
02

Auto-Generated APIs

Instant REST, GraphQL, Spatial SQL, and Realtime WebSocket subscriptions over your PostgreSQL & PostGIS database tables.

View API docs →
03

Connect Apps & AI Agents

Query via TypeScript SDKs with built-in OAuth2, or connect LLMs and AI coding assistants directly via Model Context Protocol (MCP).

See SDK & MCP docs →

Instant Data APIs

Spatial SQL, GraphQL, and typed JSON-RPC methods over your PostgreSQL schema, plus realtime WebSocket subscriptions on data changes.

Explore the APIs →

Functions in Node.js & Python

Run your own sandboxed code on demand, on a schedule, or in response to events — with secure callbacks into your data.

See Functions docs →

Production Security by Default

OAuth2, rule-based access, and rate limiting are built in so you can ship faster without re-building backend security.

Built for Geospatial

PostGIS built in. Upload GeoJSON, Shapefile or GeoPackage and get spatial SQL, GraphQL and realtime APIs on your data.

From spatial data to API →
🤖 Native MCP Support

Built for the AI-Assisted Era

Give Cursor, Claude Desktop, and custom LLM agents direct context to inspect schemas, execute spatial SQL, and run backend workflows through Centia's official Model Context Protocol (MCP) server.

  • Automated Schema Context: AI agents understand your database tables, foreign keys, and GIS columns instantly.
  • Safe Execution Boundaries: Enforce strict OAuth2 permissions, SQL parameterization, and environment scoping.
  • Natural Language Workflows: Go from user prompt to live database queries and spatial analytics seamlessly.
Connect AI Agents via MCP →
centia-mcp-agent.log
[MCP Agent] Connected to Centia Backend (https://api.centia.io)
[MCP Schema] Loaded 14 tables (PostgreSQL 16 + PostGIS 3.4)
[MCP Command] Executing spatial query for prompt: "Find nearest parks"
[SQL Execution] SELECT name, ST_Distance(...) FROM parks LIMIT 5;
[Response] 5 rows returned in 12ms. Context injected.

Ready to build your next Postgres backend?

Centia Cloud is free while in beta — or spin up a self-hosted instance with Docker. See the FAQ for beta limits.

Start in Centia CloudRead the Docs →