Examples
Compose-managed service volume reset without shell glue
A focused contract that uses action.kind: reset_compose_service_volume for governed destructive local service-data reset.
Recommended next
Typical signals
- the repo truthfully needs a fresh local database or other Compose-managed service data reset lane
- today that lane is shell
docker compose stop/rmplusdocker volume rmglue - the reset should stay explicit, reviewable, and separate from generic compose up/down tasks
Why it matters
- it turns a repo pattern into something humans, CI, and agents can read the same way
- it makes the next command obvious instead of implied by README drift
- it keeps readiness, setup, and execution explainable in receipts and diagnostics
Example contract
Use this when one destructive local recovery lane honestly owns stopping a Compose-managed service, removing one named volume, and restarting the service.
The example keeps Compose file root and project naming declarative under the action so Ota owns the full reset path instead of treating it as opaque shell.
version: 1project: name: action-reset-compose-service-volume description: Focused example for deterministic Compose-managed service volume resetexecution: default_context: host contexts: host: backend: native requirements: tools: docker: "*"tasks: postgres:reset: context: host action: kind: reset_compose_service_volume service: postgres volume: app_postgres-data compose: cwd: docker files: - docker/docker-compose.yml project_name: app requirements: tools: docker: "*" effects: external_state: - docker - postgres safe_for_agent: falseworkflows: default: local-reset local-reset: prepare: task: postgres:resetmetadata: ota: minimum_version: "1.6.22"Task notes
Use reset_compose_service_volume only when the truthful lane is one service-specific destructive local reset, not as a generic replacement for compose stop/down.
Keep action.compose.* explicit so Ota resets the right Compose project and file stack instead of guessing from shell cwd or flags.
Mark this lane outside routine safe execution boundaries when it destroys local service data.