io.github.Arun-kc/schemabrain

Official

by Arun-kc · Python

SchemaBrain

The trust and intelligence layer between AI agents and your database.

Trust and intelligence layer enabling AI agents to safely interact with databases.

Topics

ai-agentsclaudellmmcpmodel-context-protocolpythonagentic-aiai-securitydatabase-securityagent-toolsaudit-logmcp-serverpii-protectionpostgresqlsemantic-layersemantic-firewallagent-trustknowledge-graphschema-intelligence

Install

Configuration

Environment variables

DATABASE_URLrequiredsecret

Postgres connection URL (postgresql+psycopg://USER:PASSWORD@HOST:PORT/DBNAME). The agent never sees this; the URL stays in the env so credentials never appear in argv or logs.

Arguments

subcommanddefault serve

--url-envdefault DATABASE_URL

--store-pathdefault {store_path}

claude_desktop_config.json

{
  "mcpServers": {
    "schemabrain": {
      "command": "uvx",
      "args": [
        "schemabrain",
        "serve",
        "--url-env",
        "DATABASE_URL",
        "--store-path",
        "{store_path}"
      ],
      "env": {
        "DATABASE_URL": "<YOUR_DATABASE_URL>"
      }
    }
  }
}