Problem Statement
We need at least one automated test that proves ObservaKit boots, connects, runs a check, and returns a result successfully. This is the baseline "is the product alive?" test.
Proposed Solution
- stack up → source connected → check executed → result returned.
- Create a tests/smoke/ directory with a test_smoke.py file
- Create a docker-compose.test.yml or a test environment config that spins up only the services needed for the smoke test
- Write a smoke test that: boots required services → hits the health endpoint → submits a minimal test payload → asserts the expected response/result
- Use pytest + httpx or FastAPI TestClient for API calls — do not use manual curl scripts
- Document the smoke test in SETUP.md — add a "Running the Smoke Test" section
- If the smoke test reveals a bug, log it in a BUGS.md file with: description, steps to reproduce, expected vs actual behavior, severity
- Confirm smoke test runs successfully in a clean environment (not just your local machine)
Acceptance Criteria:
make test-smoke runs without manual intervention and either passes or clearly reports a known failure with a logged bug. Anyone on the team can run this in under 5 minutes from a fresh clone.
Problem Statement
We need at least one automated test that proves ObservaKit boots, connects, runs a check, and returns a result successfully. This is the baseline "is the product alive?" test.
Proposed Solution
Acceptance Criteria:
make test-smoke runs without manual intervention and either passes or clearly reports a known failure with a logged bug. Anyone on the team can run this in under 5 minutes from a fresh clone.