VoltSP Pipeline Engineering

Use this skill as a compact execution guide, then load only the reference file needed for the current task.

Prefer official VoltSP documentation whenever details conflict with memory.

Choose the track

1. Choose pipeline definition style:

• Java DSL: read references/pipelines-java.md and references/maven-setup.md.
• YAML API: read references/pipelines-yaml.md.

2. Choose runtime/deployment context:

• Local CLI: read references/cli-bare-metal.md.
• Kubernetes/Helm: read references/kubernetes.md.

3. Configure values and secrets:

• Runtime configuration model + interpolation + secure config: read references/configuration.md.

4. Implement pipeline logic and tests:

• For list of plugins and their APIS: read references/plugins-catalog.md.
• Plugin deep dives and response format: read references/plugins.md.
• Test strategy (unit/in-process/Testcontainers): read references/testing.md.
• E2E-backed pipeline implementation patterns: read references/pipeline-kinds.md.

Deterministic resources

Use deterministic helpers for repetitive or fragile setup work:

• Create starter files from templates with: python3 scripts/scaffold_voltsp_pipeline.py --out-dir <dir> --pipeline-name <name> --track <java|yaml|both>.
• Validate separation of pipeline definition vs runtime config/helm values with: python3 scripts/check_voltsp_yaml_layout.py --definition <pipeline.yaml> [--config <config.yaml>] [--helm-values <values.yaml>].

Template files used by the scaffold script:

• assets/templates/java/Pipeline.java.tmpl
• assets/templates/maven/pom.xml
• assets/templates/yaml/pipeline-definition.yaml
• assets/templates/yaml/runtime-config.yaml
• assets/templates/helm/values.yaml

Output rules

• Return copy-pasteable commands and files.
• Keep environment-specific values in config, not hardcoded in pipeline logic.
• Use secure paths for secrets (--configSecure, Helm configurationSecure, or secret interpolation) instead of plain values.