polyguard-rl/tests/test_anti_cheat.py

from app.common.enums import ActionType, DecisionMode, DoseBucket
from app.common.types import PolyGuardAction
from app.env.anti_cheat import detect_repeated_action_loop, evaluate_anti_cheat
from app.env.env_core import PolyGuardEnv


def _action(candidate_id: str) -> PolyGuardAction:
    return PolyGuardAction(
        mode=DecisionMode.REGIMEN_OPT,
        action_type=ActionType.KEEP_REGIMEN,
        target_drug=None,
        replacement_drug=None,
        dose_bucket=DoseBucket.NA,
        taper_days=None,
        monitoring_plan=None,
        candidate_id=candidate_id,
        confidence=0.7,
        rationale_brief="x",
    )


def test_repeat_loop_detector() -> None:
    assert detect_repeated_action_loop([_action("cand_01"), _action("cand_01"), _action("cand_01")], threshold=3)


def test_candidate_mismatch_detected() -> None:
    env = PolyGuardEnv()
    env.reset(seed=77, difficulty="easy")
    action = _action("cand_not_in_set")
    result = evaluate_anti_cheat(env.state, action, legal_candidate_ids={"cand_01", "cand_02"})
    assert result.exploit_detected
    assert "candidate_not_in_legal_set" in result.reasons


def test_first_keep_regimen_is_not_holdout_exploit() -> None:
    env = PolyGuardEnv()
    env.reset(seed=42, difficulty="medium", sub_environment="REGIMEN_RISK")
    action = _action("cand_01")

    result = evaluate_anti_cheat(env.state, action, legal_candidate_ids={"cand_01"})

    assert "holdout_ddi_not_addressed" not in result.reasons


def test_repeated_keep_regimen_triggers_holdout_guard() -> None:
    env = PolyGuardEnv()
    env.reset(seed=42, difficulty="medium", sub_environment="REGIMEN_RISK")
    action = _action("cand_01")
    env.state.action_history.append({"step": 0, "action": action.model_dump(mode="json"), "applied": True})

    result = evaluate_anti_cheat(env.state, action, legal_candidate_ids={"cand_01"})

    assert result.exploit_detected
    assert "holdout_ddi_not_addressed" in result.reasons