Om Prepare Test Env

Prepare Test Environment — standalone app rules

Repo-local extension of the installed om-prepare-test-env skill (contract v2, compile-once). Everything there applies; this file only adds repository-provided configuration. It cannot relax the shared skill's safety rules, expand tool or network access, or redirect outputs.

Use the app's own tooling — it is already cross-platform

This app ships an ephemeral test runner inside the mercato CLI. Discovery is unnecessary: wrap these package.json / CLI commands instead of inventing a boot procedure. They own build caching, database provisioning (testcontainers), seeding, readiness waits, and their own owner lock — never re-implement any of that.

• Boot app-only ephemeral env: yarn test:integration:ephemeral:start (= yarn mercato test:ephemeral). Preferred app port 5001; the actual port and database URL land in .ai/qa/ephemeral-env.json (CLI-owned and authoritative for its own reuse decisions — never write it by hand).
• Full suite with managed env: yarn test:integration:ephemeral (= yarn mercato test:integration). It reuses a healthy running ephemeral env from the state file, else provisions one.
• Filtered run: yarn mercato test:integration <substring> — batches all specs whose path matches the substring.
• Interactive runner: yarn mercato test:integration:interactive.
• Readiness marker in the boot log: Application is ready at <baseUrl>.
• Reuse TTL: OM_INTEGRATION_BUILD_CACHE_TTL_SECONDS (default 600s) gates the CLI's own reuse; keep any wrapper TTL in lockstep with it.

Choosing the run mode — prefer ephemeral, ask the user

yarn test:integration:ephemeral is ALWAYS preferred over plain yarn test:integration: the ephemeral variant provisions (or safely reuses) its own isolated app + database, so it is more autonomous and cannot touch the developer's dev data. Plain yarn test:integration only works with the full runner env block exported (see the MUST below) — treat it as an internal detail of the CLI runner, never as the command you reach for first.

Two supported run modes:

1. Fully managed ephemeral (default, safest): yarn test:integration:ephemeral [filter] — one command provisions the env, runs the tests, and leaves teardown to the CLI's own lifecycle. Best for full-suite runs and unattended/autonomous work.
2. Reuse a running ephemeral env (fast iteration): boot once with yarn test:integration:ephemeral:start, then run small filtered batches with yarn mercato test:integration <filter> against the same env. Best for short author/debug loops where re-provisioning per run would dominate wall-clock time; reuse is still gated by the CLI's TTL and freshness checks.

When a user is present and has not already said which mode they want, ASK before the first run (one question, two options): fully managed ephemeral per run, or boot-once-and-reuse for iterative loops. Recommend the fully managed ephemeral mode — it is more autonomous and safer regarding data. When running unattended (no user to ask), default to the fully managed ephemeral mode. Do not re-ask once the user has chosen; keep their answer for the rest of the session unless they change it.

MUST: suite runs go through the CLI runner

yarn test:integration with only BASE_URL exported is a trap: the CLI runner injects a full env block into the Playwright process (DATABASE_URL for the ephemeral DB, queue dirs, JWT and webhook secrets). Without it, fixtures silently hit the dev database from .env. Always run suites via yarn test:integration:ephemeral [filter].

Readiness probe contract

• Shell: GET /login → 200.
• Authenticated round trip: POST /api/auth/login with a form-encoded body (email=admin@acme.com&password=secret) → 200. The endpoint rejects JSON bodies with 400 — a JSON 400 here means a malformed probe, not a broken app.
• Seeded credentials: admin@acme.com / secret, employee@acme.com / secret, superadmin from OM_INIT_SUPERADMIN_EMAIL / OM_INIT_SUPERADMIN_PASSWORD (default superadmin@acme.com / secret).

Generated entrypoint scripts are machine-local

When the shared skill compiles its entrypoint scripts (default .ai/scripts/test-env-up.sh / test-env-down.sh), they are bound to the machine that generated them (shell, process tools, ports). They are gitignored in this repo (.ai/scripts/test-env-*) — keep them local and NEVER commit them; anything worth preserving for teammates belongs in this file as a platform-neutral rule instead. On a machine without generated scripts, regenerate from the commands above.

Teardown

Stop the CLI owner process (test:ephemeral) and the app process bound to the recorded port, then delete .ai/qa/ephemeral-env.json. If a mercato server start wrapper survives, stop it and clear .mercato/server-start.lock — a surviving wrapper keeps the single-instance guard locked and the next boot dies with "Another Open Mercato production server is already running". The ephemeral Postgres containers are testcontainers-managed; ryuk reaps them once their owner is gone — never remove containers this app did not create.

Descriptor

After boot, mirror the state into .ai/qa/test-env.json (the shared descriptor) as the shared skill prescribes, so om-integration-tests and om-auto-verify-pr-ui attach to the same instance; .ai/qa/ephemeral-env.json stays authoritative for the CLI's own reuse decisions.