Deleted, skipped, focused (.only), exited (sys.exit(0)) or force-passed tests
Tests and verification: one of the ways AI coding agents bloat software, from lessloc's pattern catalog. lessloc catches it as weakened-tests.
What makes deleted, skipped or force-passed tests a bad practice?
The standards that call it a bad practice: Clean Code T4 (an ignored test is a question); GitHub's AGENTS.md guidance: never remove a failing test.
What does lessloc do about deleted, skipped or force-passed tests?
lessloc catches it as weakened-tests. Skips that depend on the environment are allowed.
weakened-tests(tier 0): Test assertions or tests removed, or tests newly skipped, focused or forced to pass. Tier 0 in a session;check, where a person may remove a test on purpose, reports it at tier 1.
Examples
Caught as weakened-tests:
python
import pytest
def test_pay():
pytest.skip("flaky")
assert pay(2) == 2Written well:
python
def test_pay():
assert pay(2) == 2Related patterns
The other patterns in Tests and verification:
- Behavior added with no test changed in the same change: lessloc gives every session the rule: new behavior gets one focused test.
- Tests that assert nothing, or assert a constant: lessloc catches it as
empty-test. - Test-only branches in production code: lessloc catches it as
test-logic. - Tests that mock the code they test: lessloc catches it as
empty-test. - Production code that branches on a value from the tests: lessloc catches it as
test-logic. - Stopping short: a requested function never written: lessloc catches it as
unaddressed: names a request asks to exist ("add a--statusflag", "implementexport_csv()") are looked for in the final repository; one that appears nowhere holds the turn. - Stopping short: part of the problem left failing: lessloc catches it as
unfinished: in a turn that changed code, the agent's last reply is read sentence by sentence, with headings and lead-ins speaking for what is under them. - Saying "done" without a passing run on the final state: lessloc catches it as
tests-failedandunverified: the agent's own test runs, as its host reported them, are the evidence for the exact tree; a run that fails on the final state, or a change to how the tests run, holds the turn once, and a final state no run verified is written to the record;lessloc test,verifyandcirecord evidence too. - A refactor that changes existing tests: lessloc catches it as
refactor-tests.

