Self-Driving Database platform that proactively avoids incidents and optimizes your databases. It's plug-and-play with built-in analysis, but you can also provide your own scripts using plugins.
More screenshots, and details on inspector.azimutt.app.
Azimutt Inspector is made of three main parts:
- Collectors gather data from your database instances (connections, schemas, query stats, settings...)
- Analyzers detect issues and generate optimization suggestions and alerts, they can:
- use results from Collectors: several instances (clusters or environments) and over time (trends)
- connect to database instances to fetch more data (explain)
- use AI to analyze data and generate insights
- Notifiers alert your team via Slack, email, webhooks, or Alertmanager, to avoid alert fatigue, they can:
- filter alerts and suggestions (by severity, tags, status...)
- route them to the right channel (by instance, table...)
- deduplicate, throttle and group notifications
With them, Azimutt Inspector will autonomously provide you with guidance to improve and optimize your database with live production data.
The fastest way to get started with Azimutt Inspector is starting the Docker with a the QUICKSTART_DB environment variable to monitor.
Using the pre-built image:
# Run with your database URL (PostgreSQL example)
docker run --rm --pull always -p 3000:3000 -e QUICKSTART_DB=postgresql://user:password@host:5432/database --name inspector ghcr.io/azimuttapp/inspector:latestDocker networking:
- in Linux use
--network hostto allow the Docker to access your local network - in Mac/Windows use
host.docker.internalinstead oflocalhostin the connection URL
Building the image yourself:
# Build the Docker image locally
docker build -t azimutt/inspector .
# Run with your database URL (PostgreSQL example)
docker run --rm -p 3000:3000 -e QUICKSTART_DB=postgresql://user:password@host:5432/database --name inspector azimutt/inspectorOr launching with npm (requires Node.js 22+):
# Install dependencies
npm install
# Build the application
npm run build
# Start the application
QUICKSTART_DB=postgresql://user:password@host:5432/database npm startThen access the application at http://localhost:3000 (or http://localhost:5173 for the npm frontend).
After startup, collectors run immediately and analyzers follow. If you used QUICKSTART_DB, you'll see your first alerts and suggestions within a few seconds. If you configure database instances from the UI, collectors will run immediately, but analyzers will follow their schedule. For fast analysis you can manually trigger them from the database instance item.
Azimutt Inspector only requires read-only access to system tables on your managed databases (see install guide).
If you don't have a database to monitor, use the sample one in the demo folder with Docker Compose.
Azimutt Inspector is designed for on-premise deployment, your data never leaves your infrastructure. It connects to managed databases with read-only access using a connection pool with a single connection, and all stored secrets are encrypted at rest.
More details in the security documentation.
Azimutt Inspector Community edition is meant to bring huge value for personal use or small teams. Large organizations are encouraged to use the Enterprise Edition, it adds:
- Multi-database monitoring and auto-discovery
- User authentication (local, OAuth) and authorization (RBAC)
- Advanced collectors, analyzers, and notifiers (logs, AI-powered...)
- PostgreSQL storage
- MCP server
- Support
- And more...
More details in the Enterprise Edition documentation.
Azimutt Inspector is open-core software.
The core is open source under the MIT License — free to use, modify, and distribute without restriction.
The enterprise features (in ee/ directories) are licensed under the Elastic License 2.0 —
free to use internally, but you may not offer them as a managed service or circumvent license key mechanisms.
Please read CONTRIBUTING.md.