soden46 infrastructure-devops
Use this skill for deployment, CI/CD, containers, scheduling, workers, runtime configuration, health checks, and operational automation.
Principles
Prefer reproducible and observable deployment processes.
Separate configuration from application source where appropriate.
Automated workflows should fail clearly and avoid silently publishing broken artifacts.
Consider:
- build reproducibility
- health checks
- restart behavior
- rollout strategy
- rollback strategy
- resource limits
- secret handling
- logs
- monitoring
Background and scheduled workloads should define:
- ownership
- retries
- duplicate execution behavior
- concurrency
- failure handling
Do not assume local development behavior represents production behavior.
Adaptation
Use project evidence to determine the actual language, framework, runtime, and existing conventions.
When stack-specific implementation guidance is needed, prefer project evidence, native framework or language mechanisms, and relevant user-installed specialist skills. Technology-specific guidance must not redefine or weaken the core engineering requirement.