Overview

Execute large-scale changes across your codebase

Automations execute changes across your codebase - from routine maintenance to large-scale refactoring. They run on demand, trigger from pull request events, or execute on a schedule.

Each automation combines steps (prompts, shell scripts, PR creation) with your existing integrations. Executions run in isolated environments with your tools, dependencies, and guardrails.

How automations work

1. Define - Create an automation with steps
2. Trigger - Run manually, on PR events, or on a schedule
3. Execute - Run across repositories in parallel
4. Review - Inspect outputs, logs, and approve PRs before merging

Use cases

Manual: Documentation and metadata maintenance

Update README files and Backstage YAML across repositories on demand.

Example: Platform team triggers automation → 100 repositories analyzed → Pull requests created with updates

Pull request: Security code review

Analyze code changes for vulnerabilities when PRs are opened or updated.

Example: Developer opens PR → Automation analyzes changes → Security review comment posted

Scheduled: CVE remediation

Detect and fix security vulnerabilities on a recurring schedule - including dependency upgrades, API migrations, and test verification.

Example: Weekly scan detects CVE → Automation upgrades dependency and migrates breaking changes → PR created with complete fix

Access

Automations are available for Enterprise customers. Contact sales to enable automations for your organization.

Organization admins can share automations with groups, allowing team members to run pre-built automations without admin privileges.

Related: Tasks and Services

Automations run across repositories at scale. For per-environment automation (database seeding, server startup, test commands), see Tasks and Services. Tasks and Services are configured in automations.yaml and run within individual environments.

Next steps

Create your first automation

FAQ

Can I see automation environments in the sidebar?

No - automation environments are isolated execution contexts, not interactive environments.

How should I iterate on an automation?

Test on one or a few repositories first. Review results, adjust steps, then scale up gradually.

Projects vs repositories - when to use which?

Projects: Repetitive tasks on known repositories, pull request triggers

Repositories: Large-scale migrations across thousands of repos, fine-grained selection

Where should I install packages?

Always in your Dev Container configuration - packages are then available for all steps, pre-installed for faster runs.

How do I integrate with external systems?

Configure MCP integrations in Settings > Integrations. These are then available to your automations.