Instruction file imported from gesh75/netlog-ai (.github/instructions/tests.instructions.md). Copyright stays with the author.

Tests

• pytest only. No unittest.TestCase classes in new code.
• Name tests test_<unit>_<scenario>_<expected> so failures read as sentences.
• Prefer fixtures over setup/teardown methods; prefer @pytest.mark.parametrize over loops.
• Every bug fix starts with a regression test that fails before the fix.
• Cover the error paths, not just the happy path: bad input, network failure, empty result, boundary values.
• Never hit the live network. Mock at the client boundary using monkeypatch or unittest.mock - both are already available. Do not add a new mocking dependency without raising it in the PR first.
• Tests must be order-independent. Do not assume execution order or shared mutable state.
• No secrets or real credentials in fixtures - use obviously fake values.