Interpolation Variables

Greyhound resolves ${variable} placeholders inside both .greyhound/config.yaml and .greyhound/applications.yaml before schemas are validated. These variables originate from the deployment context and are injected by createAppFromYaml / createAppImportsFromYaml via deepStringInterpolate.

Use ${variable} anywhere a string value is accepted (service env vars, image tags, repository references, etc.). Interpolation runs before TypeBox validation, so the resolved values must satisfy the underlying schema.

Static Variables

Variable	Description
${cluster.account.accountId}	AWS account ID for the cluster handling the deployment.
${cluster.context}	Kubernetes context/identifier for the target cluster.
${cluster.dnsDomain}	Default DNS domain configured on the cluster (empty string if undefined).
${commit.sha}	Full SHA of the commit used for this deployment (may be per-service).
${commit.branch}	Branch name for the commit (may be null when deploying by tag).
${commit.tag}	Tag name for the commit (may be null when deploying by branch).
${commit.tag_or_branch}	Tag when available, otherwise branch -- handy for generic labels.
${env.name}	greyhound environment name receiving the deployment.

Database Cluster Variables

For each Aurora cluster attached to the environment, the following variables are exposed:

${database.<poolName|index>.writer_endpoint}
${database.<poolName|index>.host}
${database.<poolName|index>.port}
${database.<poolName|index>.reader_endpoint}

• <poolName> resolves to the Aurora pool name when the cluster was checked out from a pool.
• If no pool name is available, a zero-based <index> is used instead.
• ${...}.host returns the same string as ${...}.writer_endpoint; the alias remains for compatibility, but prefer writer_endpoint when you want to highlight write access.

These entries are ideal for constructing database URLs, connection strings, or injecting host/port data into app configs.

Where Interpolation Applies

File	Notes
.greyhound/config.yaml	App schema fields (services, jobs, builds, configmaps, etc.) can reference interpolation variables.
.greyhound/applications.yaml	Application import fields (repositories, config sources, optional app names, environment overrides, etc.) also support interpolation.

Example Usage

# .greyhound/config.yaml
services:
  - name: api
    image: 123456789012.dkr.ecr.us-east-1.amazonaws.com/api:${commit.sha}
    env:
      - name: CLUSTER_CONTEXT
        value: ${cluster.context}
      - name: FEATURE_FLAG
        value: ${commit.tag_or_branch}
      - name: DB_HOST
        value: ${database.prod-write.writer_endpoint}
      - name: DB_PORT
        value: ${database.prod-write.port}
      - name: DB_USER
        valueFrom:
          secretKeyRef:
            name: db-credentials
            key: username
      - name: DB_PASSWORD
        valueFrom:
          secretKeyRef:
            name: db-credentials
            key: password

      # Kubernetes evaluates $(VAR_NAME) substitutions after interpolation
      - name: DATABASE_URL
        value: postgres://$(DB_USER):$(DB_PASSWORD)@$(DB_HOST):$(DB_PORT)/app