Get Migratrix running fast

Install the Agent

The Agent authenticates using a workspace Agent token. You can provide it via command line or the MIGRATRIX_API_KEY environment variable.

Docker

docker run -d \
  --name migratrix-agent \
  --restart unless-stopped \
  -e MIGRATRIX_API_KEY="<your_agent_api_key_or_jwt>" \
  -e MIGRATRIX_DIR=/data/migratrix \
  -p 4002:4002 \
  -v migratrix-agent-data:/data/migratrix \
  ghcr.io/code-fighter-labs/migratrix-agent:latest

macOS

brew tap code-fighter-labs/tap
brew install migratrix-agent
migratrix-agent --apiKey "<your_agent_api_key_or_jwt>"

Linux

curl -sSL https://github.com/code-fighter-labs/homebrew-tap/releases/latest/download/install.sh | sh
migratrix-agent --apiKey "<your_agent_api_key_or_jwt>"

Windows

iwr https://github.com/code-fighter-labs/homebrew-tap/releases/latest/download/install.ps1 | iex
migratrix-agent --apiKey "<your_agent_api_key_or_jwt>"

Networking and HTTPS

Migratrix separates how the Agent listens from how clients reach it:

• BindAddress: where the Agent listens on the host, such as http://0.0.0.0:4002.
• PublicUrl: the externally reachable address used by the browser and web app, such as https://agent.company.com.

If the Migratrix web app is loaded over HTTPS, browsers will block connections to an HTTP agent. For production, set PublicUrl to an HTTPS address.

For localhost development, mkcert can be used to generate browser-trusted certificates for https://localhost:4002.

Connect Databases and Environments

1. Create or join a workspace.
2. Install an Agent in the network closest to the databases you want to manage.
3. Add connections for each environment such as DEV, QA, and PROD.
4. Assign environment-specific roles for admins, approvers, developers, and viewers.
5. Create one or more migration flows that reflect your promotion path.

Credentials remain on the Agent machine and are not stored on the Migratrix platform.

GitOps Workflow

Migratrix fits into a SQL-in-Git workflow while keeping execution and approvals close to the database delivery process.

1. Write a migration in Migratrix or sync one from GitHub or GitLab.
2. Review schema diff, impact analysis, and generated rollback details.
3. Open or update the pull request tied to the migration.
4. Merge the change and let the configured flow promote it through environments with approvals.
5. Track every step in the audit log and use rollback if recovery is needed.

Explorer

Explorer lets teams browse tables, preview rows, inspect schema objects, and run SQL with safeguards aligned to each environment.

• Read-only defaults in production.
• Environment-aware RBAC.
• Audited reads and writes.
• AI-assisted query generation from natural language prompts.

Migration Flows

Flows model the path from lower environments to production. Each stage can enforce its own approval policy, execution rules, and rollback behavior.

• Single-step or multi-stage flows.
• Approval gates for QA and production.
• Generated rollback scripts.
• Execution history, live logs, and audit metadata.

Compare and Sync

Compare and Sync is designed for schema drift detection and controlled alignment between environments.

• Schema diff across supported engines.
• Object-level visibility for tables, indexes, constraints, views, and more.
• Generated sync scripts to align target to source.
• Optional data comparison to inspect inserted, updated, and deleted rows.

Data Transfer

Migratrix supports same-type and cross-type transfers with batching, retries, and scheduling. This is useful for refreshes, sync jobs, and operational movement between environments.

Approvals, Audit, and Safety

• Every migration carries a traceable execution record.
• Approval decisions are tied to users, timestamps, and environments.
• Risk analysis flags destructive or suspicious SQL patterns before execution.
• Rollback readiness helps teams recover faster from bad deployments.

Need Help?

If you need help with onboarding, network topology, or enterprise rollout planning, use the contact form or email sales@migratrix.com.