Today I’m announcing the Open Beta release of BlueGreen, a Graph-Based Code and Workflow Automation platform.
BlueGreen executes declarative workflows defined in runbook.yml as directed graphs of steps. Each step runs in an isolated Ubuntu 24.04 container; logs stream live to the dashboard and are persisted for later inspection. The execution order is determined by step dependencies in the runbook.
Included runtime tooling
bashand standard Unix utilities (curl,wget,tar,git,sqlite3, etc.)- Node.js 24.11.1
- Python 3.14.1
- Go 1.25.4
- Ruby 3.4.7
- Docker and Docker Compose
Workflows are triggered by GitHub/GitLab webhooks or manually from the dashboard. Use BlueGreen for builds, tests, deployments, and general multi-step automation.
Getting started (first 5 minutes)
- Sign in with GitHub or GitLab at https://bluegreen.ci
- Import a project from your account to provision repository webhooks.
- Add a
runbook.ymlat the repository root (see Getting Started) if you don’t have one already. - Commit and push to trigger a run; runs are grouped by branch and triggered on new branches, pushes, or merge requests.
- Monitor execution and logs in the BlueGreen dashboard.
Open Beta infrastructure notes
This Beta runs on a single VM (6 vCPU, 12 GB RAM). Known limitations:
- Concurrent steps: up to 3 steps run concurrently; additional ready steps are queued as pending.
- Scheduled runs may remain pending during capacity constraints and will start when resources free up.
- Step maximum runtime: 1 hour (platform-enforced).
These limits are temporary for the Beta and will be relaxed as the platform scales.
Feedback and contribution
Your feedback is valuable. Report issues, request features, or share suggestions directly to me on LinkedIn.
Thank you for trying BlueGreen. Your testing helps shape the platform.