Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.
Python agentagentic-aiagentic-frameworkagentic-workflowaiautomationclaudedeepseekgenaigptllmlow-codemcpnextjsno-codeopenaipythonskillsworkflow
test: use SQLite sessions in core app apps (#39101)
Asuka Minato3f1036e55af6
diff --git a/api/tests/unit_tests/core/app/apps/agent_app/test_app_runner.py b/api/tests/unit_tests/core/app/apps/agent_app/test_app_runner.pyindex 95505a9949f..63c198b2c43 100644--- a/api/tests/unit_tests/core/app/apps/agent_app/test_app_runner.py+++ b/api/tests/unit_tests/core/app/apps/agent_app/test_app_runner.py@@ -7,7 +7,6 @@ from __future__ import annotationsfrom collections.abc import Callable, Iteratorfrom datetime import UTC, datetimefrom decimal import Decimal-from types import SimpleNamespacefrom typing import Any, overridefrom unittest.mock import MagicMock@@ -33,6 +32,8 @@ from pydantic_ai.messages import (ToolCallPart,ToolReturnPart,)+from sqlalchemy import event, select+from sqlalchemy.orm import Sessionfrom clients.agent_backend import (AgentBackendRunEventAdapter,@@ -47,7 +48,7 @@ from core.app.apps.agent_app.app_runner import AgentAppRunnerfrom core.app.apps.agent_app.runtime_request_builder import AgentAppRuntimeRequestBuilderfrom core.app.apps.agent_app.session_store import AgentAppSessionScope, StoredAgentAppSessionfrom core.app.apps.exc import GenerateTaskStoppedError-from core.app.entities.app_invoke_entities import InvokeFrom, UserFrom+from core.app.entities.app_invoke_entities import DifyRunContext, InvokeFrom, UserFromfrom core.app.entities.queue_entities import (QueueAgentMessageEvent,QueueAgentThoughtEvent,@@ -55,20 +56,33 @@ from core.app.entities.queue_entities import (QueueMessageEndEvent,)from core.workflow.nodes.agent_v2.ask_human_resume import AskHumanResumeOutcome+from core.workflow.nodes.agent_v2.dify_tools_builder import WorkflowAgentToolLayers+from graphon.model_runtime.entities.llm_entities import LLMResultfrom graphon.model_runtime.errors.invoke import InvokeRateLimitErrorfrom models.agent_config_entities import AgentSoulConfigfrom models.model import MessageAgentThought[email protected](autouse=True)+def bind_agent_db(monkeypatch: pytest.MonkeyPatch, sqlite_session: Session) -> None:+ """Bind the runner's ORM writes to the shared SQLite session."""+ monkeypatch.setattr(app_runner_module.db, "session", sqlite_session)+++def _thought_rows(session: Session) -> list[MessageAgentThought]:+ session.expire_all()+ return list(session.scalars(select(MessageAgentThought).order_by(MessageAgentThought.position)).all())++class _FakeCredentialsProvider:def fetch(self, provider_name: str, model_name: str) -> dict[str, Any]:return {"openai_api_key": "sk-test"}class _NoToolsBuilder:- def build_layers(self, **kwargs):+ def build_layers(self, **kwargs: Any) -> WorkflowAgentToolLayers:del kwargs- return SimpleNamespace(plugin_tools=None, core_tools=None, exposed_tool_names=lambda: [])+ return WorkflowAgentToolLayers()class _FakeQueueManager:@@ -354,27 +368,6 @@ class _ProcessStreamingFakeAgentBackendRunClient(FakeAgentBackendRunClient):)-class _FakeDbSession:- def __init__(self) -> None:- self.rows: dict[str, MessageAgentThought] = {}- self.rollback_count = 0-- def add(self, row: MessageAgentThought) -> None:- self.rows[str(row.id)] = row-- def commit(self) -> None:- pass-- def get(self, _model: type[MessageAgentThought], row_id: str) -> MessageAgentThought | None:- return self.rows.get(row_id)-- def delete(self, row: MessageAgentThought) -> None:- self.rows.pop(str(row.id), None)-- def rollback(self) -> None:- self.rollback_count += 1--class _FakeSessionStore:def __init__(self,@@ -450,8 +443,8 @@ def _soul() -> AgentSoulConfig:)-def _dify_ctx() -> Any:- return SimpleNamespace(+def _dify_ctx() -> DifyRunContext:+ return DifyRunContext(tenant_id="tenant-1",app_id="app-1",user_id="user-1",@@ -497,9 +490,14 @@ def _message_end(qm: _FakeQueueManager) -> QueueMessageEndEvent:return next(e for e in qm.events if isinstance(e, QueueMessageEndEvent))-def _saved_user_query(qm: _FakeQueueManager) -> str:+def _llm_result(qm: _FakeQueueManager) -> LLMResult:llm_result = _message_end(qm).llm_resultassert llm_result is not None+ return llm_result+++def _saved_user_query(qm: _FakeQueueManager) -> str:+ llm_result = _llm_result(qm)prompt_messages = llm_result.prompt_messagesassert len(prompt_messages) == 1content = prompt_messages[0].content@@ -507,7 +505,7 @@ def _saved_user_query(qm: _FakeQueueManager) -> str:return content-def test_successful_turn_publishes_chunk_and_message_end_and_saves_session():+def test_successful_turn_publishes_chunk_and_message_end_and_saves_session() -> None:client = FakeAgentBackendRunClient() # SUCCESS: output {"text": "hello agent"}store = _FakeSessionStore()qm = _FakeQueueManager()@@ -522,8 +520,8 @@ def test_successful_turn_publishes_chunk_and_message_end_and_saves_session():assert len(chunk_events) == 1assert len(end_events) == 1assert chunk_events[0].chunk.delta.message.content == "hello agent"- assert end_events[0].llm_result.message.content == "hello agent"- assert end_events[0].llm_result.model == "gpt-4o-mini"+ assert _llm_result(qm).message.content == "hello agent"+ assert _llm_result(qm).model == "gpt-4o-mini"assert _saved_user_query(qm) == "hello"# The conversation session snapshot is persisted for multi-turn continuity.assert store.saved@@ -550,9 +548,9 @@ def test_turn_uses_resolved_backend_binding_before_backend_invocation() -> None:assert len(store.resolved_scopes) == 1-def test_successful_turn_routes_stream_text_to_agent_message_and_uses_terminal_output(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_successful_turn_routes_stream_text_to_agent_message_and_uses_terminal_output(+ sqlite_session: Session,+) -> None:client = _StreamingFakeAgentBackendRunClient()store = _FakeSessionStore()qm = _FakeQueueManager()@@ -565,22 +563,21 @@ def test_successful_turn_routes_stream_text_to_agent_message_and_uses_terminal_oassert [event.chunk.delta.message.content for event in chunk_events] == ["hello agent"]assert [event.chunk.delta.message.content for event in agent_message_events] == ["hello ", "agent"]assert len(end_events) == 1- assert end_events[0].llm_result.message.content == "hello agent"- assert end_events[0].llm_result.usage.prompt_tokens == 3- assert end_events[0].llm_result.usage.completion_tokens == 5- assert end_events[0].llm_result.usage.total_tokens == 8- assert end_events[0].llm_result.usage.prompt_price == Decimal("0.000015")- assert end_events[0].llm_result.usage.completion_price == Decimal("0.000150")- assert end_events[0].llm_result.usage.total_price == Decimal("0.000165")- assert end_events[0].llm_result.usage.currency == "USD"- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ llm_result = _llm_result(qm)+ assert llm_result.message.content == "hello agent"+ assert llm_result.usage.prompt_tokens == 3+ assert llm_result.usage.completion_tokens == 5+ assert llm_result.usage.total_tokens == 8+ assert llm_result.usage.prompt_price == Decimal("0.000015")+ assert llm_result.usage.completion_price == Decimal("0.000150")+ assert llm_result.usage.total_price == Decimal("0.000165")+ assert llm_result.usage.currency == "USD"+ rows = _thought_rows(sqlite_session)assert rows == []assert store.saved-def test_successful_turn_routes_single_agent_message_delta(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_successful_turn_routes_single_agent_message_delta(sqlite_session: Session) -> None:client = _StreamingSingleAgentMessageDeltaFakeAgentBackendRunClient()store = _FakeSessionStore()qm = _FakeQueueManager()@@ -593,12 +590,42 @@ def test_successful_turn_routes_single_agent_message_delta(monkeypatch):assert [event.chunk.delta.message.content for event in chunk_events] == ["hello agent"]assert [event.chunk.delta.message.content for event in agent_message_events] == ["hello"]assert len(end_events) == 1- assert end_events[0].llm_result.message.content == "hello agent"- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ assert _llm_result(qm).message.content == "hello agent"+ rows = _thought_rows(sqlite_session)assert rows == []-def test_successful_turn_with_null_terminal_output_publishes_empty_answer_not_literal_null():+def test_thought_commit_failure_rolls_back_and_turn_continues(sqlite_session: Session) -> None:+ rollback_events: list[Session] = []+ should_fail = True++ def fail_first_commit(_session: Session) -> None:+ nonlocal should_fail+ if should_fail:+ should_fail = False+ raise RuntimeError("forced thought commit failure")++ def record_rollback(session: Session) -> None:+ rollback_events.append(session)++ event.listen(sqlite_session, "before_commit", fail_first_commit)+ event.listen(sqlite_session, "after_rollback", record_rollback)+ try:+ client = _StreamingSingleAgentMessageDeltaFakeAgentBackendRunClient()+ store = _FakeSessionStore()+ qm = _FakeQueueManager()++ _run(_runner(client, store), qm)+ finally:+ event.remove(sqlite_session, "before_commit", fail_first_commit)+ event.remove(sqlite_session, "after_rollback", record_rollback)++ assert rollback_events == [sqlite_session]+ assert _thought_rows(sqlite_session) == []+ assert _llm_result(qm).message.content == "hello agent"+++def test_successful_turn_with_null_terminal_output_publishes_empty_answer_not_literal_null() -> None:client = _NullOutputFakeAgentBackendRunClient()store = _FakeSessionStore()qm = _FakeQueueManager()@@ -611,12 +638,12 @@ def test_successful_turn_with_null_terminal_output_publishes_empty_answer_not_liassert chunk_events == []assert agent_message_events == []assert len(end_events) == 1- assert end_events[0].llm_result.message.content == ""+ assert _llm_result(qm).message.content == ""-def test_successful_turn_with_stream_text_and_null_terminal_output_keeps_empty_message(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_successful_turn_with_stream_text_and_null_terminal_output_keeps_empty_message(+ sqlite_session: Session,+) -> None:client = _StreamingTextNullOutputFakeAgentBackendRunClient()store = _FakeSessionStore()qm = _FakeQueueManager()@@ -629,15 +656,13 @@ def test_successful_turn_with_stream_text_and_null_terminal_output_keeps_empty_massert chunk_events == []assert [event.chunk.delta.message.content for event in agent_message_events] == ["streamed answer"]assert len(end_events) == 1- assert end_events[0].llm_result.message.content == ""- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ assert _llm_result(qm).message.content == ""+ rows = _thought_rows(sqlite_session)assert len(rows) == 1assert rows[0].answer == "streamed answer"-def test_successful_turn_routes_agent_answer_to_agent_message(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_successful_turn_routes_agent_answer_to_agent_message(sqlite_session: Session) -> None:client = _AgentAnswerStreamingFakeAgentBackendRunClient()store = _FakeSessionStore()qm = _FakeQueueManager()@@ -650,21 +675,21 @@ def test_successful_turn_routes_agent_answer_to_agent_message(monkeypatch):assert [event.chunk.delta.message.content for event in agent_message_events] == ["hello ", "agent"]end_events = [e for e in qm.events if isinstance(e, QueueMessageEndEvent)]assert len(end_events) == 1- assert end_events[0].llm_result.message.content == "final answer"+ assert _llm_result(qm).message.content == "final answer"thought_events = [e for e in qm.events if isinstance(e, QueueAgentThoughtEvent)]assert len(thought_events) == 2- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert len(rows) == 1assert rows[0].answer == "hello agent"assert rows[0].thought == ""assert rows[0].tool == ""-def test_agent_message_deltas_are_debounced_to_agent_message(monkeypatch):+def test_agent_message_deltas_are_debounced_to_agent_message(+ monkeypatch: pytest.MonkeyPatch, sqlite_session: Session+) -> None:monkeypatch.setattr(app_runner_module.time, "monotonic", _MonotonicClock(0.0, 0.2))- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)client = _StreamingFakeAgentBackendRunClient()store = _FakeSessionStore()qm = _FakeQueueManager()@@ -675,13 +700,13 @@ def test_agent_message_deltas_are_debounced_to_agent_message(monkeypatch):agent_message_events = [e for e in qm.events if isinstance(e, QueueAgentMessageEvent)]assert [event.chunk.delta.message.content for event in chunk_events] == ["hello agent"]assert [event.chunk.delta.message.content for event in agent_message_events] == ["hello agent"]- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert rows == []-def test_successful_turn_persists_thinking_and_tool_process_events(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_successful_turn_persists_thinking_and_tool_process_events(+ sqlite_session: Session,+) -> None:client = _ProcessStreamingFakeAgentBackendRunClient()store = _FakeSessionStore()qm = _FakeQueueManager()@@ -695,7 +720,7 @@ def test_successful_turn_persists_thinking_and_tool_process_events(monkeypatch):thought_events = [e for e in qm.events if isinstance(e, QueueAgentThoughtEvent)]assert len(thought_events) >= 3- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert rows[0].thought == "I need to inspect the file."assert rows[0].tool == ""assert rows[1].tool == "bash"@@ -704,9 +729,9 @@ def test_successful_turn_persists_thinking_and_tool_process_events(monkeypatch):assert len(rows) == 2-def test_streaming_turn_cancels_after_persisting_seen_agent_answer(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_streaming_turn_cancels_after_persisting_seen_agent_answer(+ sqlite_session: Session,+) -> None:store = _FakeSessionStore()qm = _FakeQueueManager()client = _StreamingStopAfterFirstDeltaFakeAgentBackendRunClient(queue_manager=qm)@@ -718,15 +743,15 @@ def test_streaming_turn_cancels_after_persisting_seen_agent_answer(monkeypatch):agent_message_events = [e for e in qm.events if isinstance(e, QueueAgentMessageEvent)]assert chunk_events == []assert [event.chunk.delta.message.content for event in agent_message_events] == ["hello "]- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert len(rows) == 1assert rows[0].answer == "hello "assert client.cancelled_run_ids == ["fake-run-1"]-def test_tool_result_without_identity_does_not_attach_to_previous_tool(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_tool_result_without_identity_does_not_attach_to_previous_tool(+ sqlite_session: Session,+) -> None:qm = _FakeQueueManager()recorder = app_runner_module._AgentProcessRecorder(dify_context=_dify_ctx(),@@ -758,7 +783,7 @@ def test_tool_result_without_identity_does_not_attach_to_previous_tool(monkeypat))- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert len(rows) == 2assert rows[0].tool == "shell_run"assert rows[0].tool_input == '{"script": "npx skills find browser"}'@@ -768,9 +793,7 @@ def test_tool_result_without_identity_does_not_attach_to_previous_tool(monkeypatassert rows[1].observation == "Knowledge base search results: browser skill"-def test_answer_suffix_trim_keeps_non_terminal_prefix(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_answer_suffix_trim_keeps_non_terminal_prefix(sqlite_session: Session) -> None:qm = _FakeQueueManager()recorder = app_runner_module._AgentProcessRecorder(dify_context=_dify_ctx(),@@ -781,14 +804,12 @@ def test_answer_suffix_trim_keeps_non_terminal_prefix(monkeypatch):recorder.append_answer_text("intermediate final answer")recorder.trim_answer_suffix("final answer")- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert len(rows) == 1assert rows[0].answer == "intermediate "-def test_tool_call_part_binds_late_call_id_to_delta_row(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_tool_call_part_binds_late_call_id_to_delta_row(sqlite_session: Session) -> None:qm = _FakeQueueManager()recorder = app_runner_module._AgentProcessRecorder(dify_context=_dify_ctx(),@@ -840,16 +861,14 @@ def test_tool_call_part_binds_late_call_id_to_delta_row(monkeypatch):))- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert len(rows) == 1assert rows[0].tool == "knowledge_base_search"assert rows[0].tool_input == '{"query": "browser"}'assert rows[0].observation == "Knowledge base search results: browser skill"-def test_thinking_after_tool_starts_new_snapshot_row(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_thinking_after_tool_starts_new_snapshot_row(sqlite_session: Session) -> None:qm = _FakeQueueManager()recorder = app_runner_module._AgentProcessRecorder(dify_context=_dify_ctx(),@@ -898,16 +917,16 @@ def test_thinking_after_tool_starts_new_snapshot_row(monkeypatch):))- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert [row.thought for row in rows] == ["The first thought.", "", "The next thought."]assert rows[0].id != rows[2].idassert rows[1].tool == "shell_run"assert rows[1].tool_input == '{"cmd": "date"}'-def test_tool_result_without_call_id_matches_unique_open_tool_name(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_tool_result_without_call_id_matches_unique_open_tool_name(+ sqlite_session: Session,+) -> None:qm = _FakeQueueManager()recorder = app_runner_module._AgentProcessRecorder(dify_context=_dify_ctx(),@@ -942,16 +961,16 @@ def test_tool_result_without_call_id_matches_unique_open_tool_name(monkeypatch):))- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert len(rows) == 1assert rows[0].tool == "knowledge_base_search"assert rows[0].tool_input == '{"query": "browser"}'assert rows[0].observation == "Knowledge base search results: browser skill"-def test_repeated_tool_calls_without_call_id_or_index_create_distinct_rows(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_repeated_tool_calls_without_call_id_or_index_create_distinct_rows(+ sqlite_session: Session,+) -> None:qm = _FakeQueueManager()recorder = app_runner_module._AgentProcessRecorder(dify_context=_dify_ctx(),@@ -1012,7 +1031,7 @@ def test_repeated_tool_calls_without_call_id_or_index_create_distinct_rows(monke))- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert len(rows) == 2assert rows[0].tool == "shell_run"assert rows[0].tool_input == '{"script": "lookup find"}'@@ -1022,9 +1041,9 @@ def test_repeated_tool_calls_without_call_id_or_index_create_distinct_rows(monkeassert rows[1].observation == "out output"-def test_repeated_tool_calls_with_placeholder_call_id_and_reused_index_create_distinct_rows(monkeypatch):- fake_session = _FakeDbSession()- monkeypatch.setattr(app_runner_module.db, "session", fake_session)+def test_repeated_tool_calls_with_placeholder_call_id_and_reused_index_create_distinct_rows(+ sqlite_session: Session,+) -> None:qm = _FakeQueueManager()recorder = app_runner_module._AgentProcessRecorder(dify_context=_dify_ctx(),@@ -1063,7 +1082,7 @@ def test_repeated_tool_calls_with_placeholder_call_id_and_reused_index_create_di))- rows = sorted(fake_session.rows.values(), key=lambda row: row.position)+ rows = _thought_rows(sqlite_session)assert len(rows) == 2assert rows[0].tool == "shell_run"assert rows[0].tool_input == '{"script": "lookup find"}'@@ -1073,7 +1092,7 @@ def test_repeated_tool_calls_with_placeholder_call_id_and_reused_index_create_diassert rows[1].observation == "out output"-def test_prior_session_snapshot_is_threaded_into_request():+def test_prior_session_snapshot_is_threaded_into_request() -> None:prior = CompositorSessionSnapshot(layers=[])client = FakeAgentBackendRunClient()store = _FakeSessionStore(loaded=prior)@@ -1085,7 +1104,7 @@ def test_prior_session_snapshot_is_threaded_into_request():assert client.request.session_snapshot is prior-def test_debug_session_scope_can_reuse_conversation_across_config_snapshots():+def test_debug_session_scope_can_reuse_conversation_across_config_snapshots() -> None:prior = CompositorSessionSnapshot(layers=[])client = FakeAgentBackendRunClient()store = _FakeSessionStore(loaded=prior)@@ -1111,7 +1130,7 @@ def test_debug_session_scope_can_reuse_conversation_across_config_snapshots():assert store.saved[0][0].agent_config_snapshot_id == "snap-new"-def test_failed_run_raises_agent_backend_error():+def test_failed_run_raises_agent_backend_error() -> None:client = FakeAgentBackendRunClient(scenario=FakeAgentBackendScenario.FAILED)store = _FakeSessionStore()qm = _FakeQueueManager()@@ -1123,7 +1142,7 @@ def test_failed_run_raises_agent_backend_error():assert store.saved == []-def test_agent_backend_failure_to_exception_maps_rate_limit_reason():+def test_agent_backend_failure_to_exception_maps_rate_limit_reason() -> None:err = app_runner_module._agent_backend_failure_to_exception(AgentBackendRunFailedInternalEvent(run_id="run-1",@@ -1136,7 +1155,7 @@ def test_agent_backend_failure_to_exception_maps_rate_limit_reason():assert str(err) == "quota exceeded"-def test_agent_backend_failure_to_exception_preserves_unknown_reason_context():+def test_agent_backend_failure_to_exception_preserves_unknown_reason_context() -> None:err = app_runner_module._agent_backend_failure_to_exception(AgentBackendRunFailedInternalEvent(run_id="run-1",@@ -1158,7 +1177,7 @@ def test_agent_backend_failure_to_exception_preserves_unknown_reason_context():assert str(err) == "Knowledge retrieval failed (agent_run_id=run-1)"-def test_stopped_task_cancels_agent_backend_run_and_skips_session_save():+def test_stopped_task_cancels_agent_backend_run_and_skips_session_save() -> None:client = _RecordingFakeAgentBackendRunClient()store = _FakeSessionStore()qm = _StoppedQueueManager()@@ -1170,14 +1189,14 @@ def test_stopped_task_cancels_agent_backend_run_and_skips_session_save():assert store.saved == []-def test_terminal_output_to_answer_handles_plain_string_and_dict():+def test_terminal_output_to_answer_handles_plain_string_and_dict() -> None:assert AgentAppRunner._terminal_output_to_answer(None) == ""assert AgentAppRunner._terminal_output_to_answer("plain text") == "plain text"assert AgentAppRunner._terminal_output_to_answer({"text": "hi"}) == "hi"assert AgentAppRunner._terminal_output_to_answer({"a": 1}) == '{"a": 1}'-def test_ask_human_pauses_turn_creates_form_and_persists_correlation():+def test_ask_human_pauses_turn_creates_form_and_persists_correlation() -> None:# ENG-635/637: the PAUSED scenario emits a dify.ask_human deferred call, so# the chat turn ends by creating a conversation-owned HITL form + saving the# pause correlation, instead of crashing. Stub the form repo (DB-free).@@ -1205,7 +1224,7 @@ def test_ask_human_pauses_turn_creates_form_and_persists_correlation():assert store.saved[0][4] == "fake-ask-human-1"-def test_submitted_form_resumes_turn_with_deferred_tool_results(monkeypatch):+def test_submitted_form_resumes_turn_with_deferred_tool_results(monkeypatch: pytest.MonkeyPatch) -> None:# ENG-638: a turn that runs while a pending form is answered threads the# human's reply into the request as deferred_tool_results.snapshot = CompositorSessionSnapshot(layers=[])