test_current_shadow_plan_releases_legacy_high_risk_once (test_review_policy_integration.ProtocolPolicyIntegration.test_current_shadow_plan_releases_legacy_high_risk_once) ... ok test_gateway_shadow_verdict_is_recorded_and_sent_without_approval (test_review_policy_integration.ProtocolPolicyIntegration.test_gateway_shadow_verdict_is_recorded_and_sent_without_approval) ... ok test_gateway_strict_verdict_overrides_stale_shadow_cache (test_review_policy_integration.ProtocolPolicyIntegration.test_gateway_strict_verdict_overrides_stale_shadow_cache) ... ok test_score_only_high_risk_still_requires_approval (test_review_policy_integration.ProtocolPolicyIntegration.test_score_only_high_risk_still_requires_approval) ... ok test_disabling_keyword_rule_releases_existing_draft (test_review_policy_integration.VisualPolicyIntegration.test_disabling_keyword_rule_releases_existing_draft) ... [待回复恢复] 已从磁盘恢复 1 个未完成任务。 ok test_new_strict_verdict_overrides_old_shadow_config (test_review_policy_integration.VisualPolicyIntegration.test_new_strict_verdict_overrides_old_shadow_config) ... ok test_old_judge_callback_cannot_modify_new_turn (test_review_policy_integration.VisualPolicyIntegration.test_old_judge_callback_cannot_modify_new_turn) ... ok test_shadow_mode_survives_restart_and_releases_obsolete_review (test_review_policy_integration.VisualPolicyIntegration.test_shadow_mode_survives_restart_and_releases_obsolete_review) ... [待回复恢复] 已从磁盘恢复 1 个未完成任务。 ok test_shadow_never_clears_unknown_receipt (test_review_policy_integration.VisualPolicyIntegration.test_shadow_never_clears_unknown_receipt) ... ok test_old_gateway_or_local_result_cannot_overwrite_new_taskid_on_same_object (test_review_policy_regression.ProtocolGenerationIdentityTests.test_old_gateway_or_local_result_cannot_overwrite_new_taskid_on_same_object) ... ok test_configuration_failure_preserves_high_risk_hold (test_review_policy_regression.ReviewPolicyTests.test_configuration_failure_preserves_high_risk_hold) ... ok test_current_enforcing_plan_does_not_turn_old_shadow_observation_into_hold (test_review_policy_regression.ReviewPolicyTests.test_current_enforcing_plan_does_not_turn_old_shadow_observation_into_hold) ... ok test_disabled_gateway_cache_cannot_override_local_mode (test_review_policy_regression.ReviewPolicyTests.test_disabled_gateway_cache_cannot_override_local_mode) ... ok test_disabled_rules_cannot_hold_customer_or_reply_text (test_review_policy_regression.ReviewPolicyTests.test_disabled_rules_cannot_hold_customer_or_reply_text) ... ok test_explicit_current_shadow_rechecks_old_high_risk (test_review_policy_regression.ReviewPolicyTests.test_explicit_current_shadow_rechecks_old_high_risk) ... ok test_gateway_current_mode_uses_only_this_process_settings (test_review_policy_regression.ReviewPolicyTests.test_gateway_current_mode_uses_only_this_process_settings) ... ok test_gateway_shadow_is_effective_for_legacy_record (test_review_policy_regression.ReviewPolicyTests.test_gateway_shadow_is_effective_for_legacy_record) ... ok test_high_risk_shadow_observes_and_enforcing_modes_hold (test_review_policy_regression.ReviewPolicyTests.test_high_risk_shadow_observes_and_enforcing_modes_hold) ... ok test_local_judge_disabled_means_unknown_even_with_shadow_default (test_review_policy_regression.ReviewPolicyTests.test_local_judge_disabled_means_unknown_even_with_shadow_default) ... ok test_missing_mode_and_unknown_current_plan_remain_manual (test_review_policy_regression.ReviewPolicyTests.test_missing_mode_and_unknown_current_plan_remain_manual) ... ok test_missing_or_invalid_gateway_mode_is_unknown (test_review_policy_regression.ReviewPolicyTests.test_missing_or_invalid_gateway_mode_is_unknown) ... ok test_non_high_risks_do_not_trigger_judge_review (test_review_policy_regression.ReviewPolicyTests.test_non_high_risks_do_not_trigger_judge_review) ... ok test_plain_review_with_draft_can_be_rechecked (test_review_policy_regression.ReviewPolicyTests.test_plain_review_with_draft_can_be_rechecked) ... ok test_recheck_protects_receipt_and_manual_or_foreign_work (test_review_policy_regression.ReviewPolicyTests.test_recheck_protects_receipt_and_manual_or_foreign_work) ... ok test_recheck_rejects_any_send_state (test_review_policy_regression.ReviewPolicyTests.test_recheck_rejects_any_send_state) ... ok test_recorded_enforcing_mode_cannot_be_overridden_by_current_shadow (test_review_policy_regression.ReviewPolicyTests.test_recorded_enforcing_mode_cannot_be_overridden_by_current_shadow) ... ok test_shadow_still_honors_enabled_keyword_rule (test_review_policy_regression.ReviewPolicyTests.test_shadow_still_honors_enabled_keyword_rule) ... ok test_valid_recorded_mode_never_reads_current_plan (test_review_policy_regression.ReviewPolicyTests.test_valid_recorded_mode_never_reads_current_plan) ... ok test_active_keyword_rule_remains_held (test_review_policy_regression.VisualReviewIntegrationTests.test_active_keyword_rule_remains_held) ... ok test_current_gateway_shadow_releases_known_old_model_hold (test_review_policy_regression.VisualReviewIntegrationTests.test_current_gateway_shadow_releases_known_old_model_hold) ... ok test_direct_model_round_clears_previous_judge_metadata (test_review_policy_regression.VisualReviewIntegrationTests.test_direct_model_round_clears_previous_judge_metadata) ... ok test_disabled_keyword_review_is_requeued_without_approval (test_review_policy_regression.VisualReviewIntegrationTests.test_disabled_keyword_review_is_requeued_without_approval) ... ok test_disabled_rules_do_not_release_protected_tasks (test_review_policy_regression.VisualReviewIntegrationTests.test_disabled_rules_do_not_release_protected_tasks) ... ok test_global_review_mode_does_not_clear_any_hold (test_review_policy_regression.VisualReviewIntegrationTests.test_global_review_mode_does_not_clear_any_hold) ... ok test_global_review_releases_in_auto_mode_when_unprotected (test_review_policy_regression.VisualReviewIntegrationTests.test_global_review_releases_in_auto_mode_when_unprotected) ... ok test_legacy_high_risk_without_valid_current_mode_stays_held (test_review_policy_regression.VisualReviewIntegrationTests.test_legacy_high_risk_without_valid_current_mode_stays_held) ... ok test_missing_rule_evidence_stays_conservative (test_review_policy_regression.VisualReviewIntegrationTests.test_missing_rule_evidence_stays_conservative) ... ok test_model_record_persists_actual_judge_mode (test_review_policy_regression.VisualReviewIntegrationTests.test_model_record_persists_actual_judge_mode) ... ok test_orchestration_binds_callback_to_generation_before_model_starts (test_review_policy_regression.VisualReviewIntegrationTests.test_orchestration_binds_callback_to_generation_before_model_starts) ... ok test_recorded_strict_visual_verdict_stays_held_with_cached_shadow (test_review_policy_regression.VisualReviewIntegrationTests.test_recorded_strict_visual_verdict_stays_held_with_cached_shadow) ... ok test_review_cleanup_rolls_back_when_save_raises (test_review_policy_regression.VisualReviewIntegrationTests.test_review_cleanup_rolls_back_when_save_raises) ... ok test_review_cleanup_rolls_back_when_save_returns_false (test_review_policy_regression.VisualReviewIntegrationTests.test_review_cleanup_rolls_back_when_save_returns_false) ... ok test_stale_callback_cannot_write_after_inplace_reset_or_new_id (test_review_policy_regression.VisualReviewIntegrationTests.test_stale_callback_cannot_write_after_inplace_reset_or_new_id) ... ok test_stale_callback_cannot_write_to_new_task_object (test_review_policy_regression.VisualReviewIntegrationTests.test_stale_callback_cannot_write_to_new_task_object) ... ok test_unknown_historical_reason_is_not_automatically_released (test_review_policy_regression.VisualReviewIntegrationTests.test_unknown_historical_reason_is_not_automatically_released) ... ok test_visual_enforcing_and_legacy_high_risk_still_hold (test_review_policy_regression.VisualReviewIntegrationTests.test_visual_enforcing_and_legacy_high_risk_still_hold) ... ok test_visual_shadow_high_risk_does_not_hold (test_review_policy_regression.VisualReviewIntegrationTests.test_visual_shadow_high_risk_does_not_hold) ... ok ---------------------------------------------------------------------- Ran 47 tests in 0.534s OK