{% extends 'base.html' %} {% load static %} {% block content %}
Available on PyPI

Universal Database Intelligence layer
for AI Agents

Securely connect AI agents to your databases. Enrich your schema with descriptions, aliases, and privacy controls. Built for the Model Context Protocol with SQLShield protection.

Get Started View on GitHub

Your Data, Securely Bridged to AI

TernoDBI handles the complexity of connection, security, and protocol translation.

DATA SOURCES
INTELLIGENCE LAYER
TernoDBI
AI CONSUMERS
Claude Desktop
Custom Agents

Quick Start

1. Check Health

Verify the API is reachable.

curl http://localhost:8000/api/query/health/

2. Create Token

Generate a scoped API key.

python manage.py issue_token --name "Agent 1" --type query

3. Connect Agent

Use with Claude Desktop via MCP.

uvx --from terno-dbi dbi-mcp query

Capabilities

SQLShield

Automatic read-only enforcement and AST validation to prevent SQL injection.

Access Control

Scoped API tokens with expiration, role-based permissions, and secure authentication.

Multi-DB

Unified API for Postgres, MySQL, Snowflake, BigQuery, Databricks, Oracle.

Native MCP

First-class support for the Model Context Protocol (Admin & Query servers).

Schema Enrichment

Generate descriptions, rename tables/columns, and enhance metadata so LLMs understand your schema better.

Privacy Controls

Hide tables/columns from LLMs, use public aliases, and apply row-level data filtering.

Pagination

Cursor-based result pagination for handling large datasets efficiently without overwhelming agents.

Multi-Source

Connect multiple datasources simultaneously and query across them from a single API.

API Reference

Quick overview of available endpoints. View full API documentation →

Query Service (Read Operations)
Method Endpoint Description
GET /api/query/health/ Health check
GET /api/query/datasources/ List all datasources
GET /api/query/datasources/{id}/ Get datasource details
GET /api/query/datasources/{id}/schema/ Get full schema
GET /api/query/datasources/{id}/tables/ List tables
POST /api/query/datasources/{id}/query/ Execute SQL query
Admin Service (Write Operations)
Method Endpoint Description
POST /api/admin/datasources/ Create datasource
PUT /api/admin/datasources/{id}/ Update datasource
DELETE /api/admin/datasources/{id}/delete/ Delete datasource
POST /api/admin/datasources/{id}/sync/ Sync metadata
PUT /api/admin/tables/{id}/ Update table metadata
PUT /api/admin/columns/{id}/ Update column metadata
{% endblock %}