~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google, ...). 10+ search engines - arXiv, PubMed, your private documents. Everything Local & Encrypted.
Python academiaanthropicarxivbravedeep-researchencryptionhome-automationhomeserverlocallocal-deep-researchlocal-llmmistralollamaopenaipubmedresearchresearch-toolretrieval-augmented-generationsearxngself-hosted
chore(research): stop creating empty research_outputs/research_<id> dirs (#5152)
LearningCircuitef7dd733ca2b
diff --git a/changelog.d/+no-empty-research-output-dirs.misc.md b/changelog.d/+no-empty-research-output-dirs.misc.mdnew file mode 100644index 0000000000..06277d88c8--- /dev/null+++ b/changelog.d/+no-empty-research-output-dirs.misc.md@@ -0,0 +1 @@+Running a research no longer creates an empty `research_outputs/research_<id>` directory on disk. That directory was vestigial leftover from an older on-disk report flow; reports are stored in the encrypted per-user database via `storage.save_report`, and the file path was never written to, so the directory just accumulated one empty folder per research run.diff --git a/src/local_deep_research/web/services/research_service.py b/src/local_deep_research/web/services/research_service.pyindex cfce0314e9..51e40363d6 100644--- a/src/local_deep_research/web/services/research_service.py+++ b/src/local_deep_research/web/services/research_service.py@@ -925,10 +925,6 @@ def run_research_process(research_id, query, mode, **kwargs):f"custom_endpoint={custom_endpoint}, strategy={strategy}")- # Set up the AI Context Manager- output_dir = OUTPUT_DIR / f"research_{research_id}"- output_dir.mkdir(parents=True, exist_ok=True)-# Create a settings context that uses snapshot - no database access in threadssettings_context = SnapshotSettingsContext(settings_snapshot, username=usernamediff --git a/tests/web/services/test_research_service_coverage.py b/tests/web/services/test_research_service_coverage.pyindex c62d5bdf35..c794ffd219 100644--- a/tests/web/services/test_research_service_coverage.py+++ b/tests/web/services/test_research_service_coverage.py@@ -974,6 +974,100 @@ class TestRunResearchProcessQuickMode:assert research.status == "completed"+class TestRunResearchProcessNoOutputDir:+ """Regression: run_research_process must not create an empty+ research_outputs/research_<id> directory on disk.++ Reports are persisted via storage.save_report into the encrypted+ per-user database; the on-disk output directory is not used by the+ encrypted-database version, so run_research_process must not create it.+ """++ def test_no_research_output_dir_created(self, tmp_path):+ mock_session = MagicMock()+ research = _make_research_mock()+ mock_session.query.return_value.filter_by.return_value.first.return_value = research++ mock_system = MagicMock()+ mock_system.analyze_topic.return_value = {+ "findings": [],+ "formatted_findings": "# Summary",+ "iterations": 1,+ }++ mock_formatter = MagicMock()+ mock_formatter.format_document.return_value = "formatted content"++ mock_storage = MagicMock()+ mock_storage.save_report.return_value = True++ mock_sources_service = MagicMock()+ mock_sources_service.save_research_sources.return_value = 0++ with (+ # Point the module's output directory at an empty tmp_path so we+ # can assert nothing gets written under it.+ patch(f"{RS}.OUTPUT_DIR", tmp_path),+ patch(+ "local_deep_research.web.routes.globals.is_termination_requested",+ return_value=False,+ ),+ patch(+ "local_deep_research.web.routes.globals.is_research_active",+ return_value=True,+ ),+ patch(+ "local_deep_research.web.routes.globals.update_progress_and_check_active",+ return_value=(50, True),+ ),+ patch(f"{RS}.get_llm", return_value=MagicMock()),+ patch(f"{RS}.get_search", return_value=MagicMock()),+ patch(f"{RS}.AdvancedSearchSystem", return_value=mock_system),+ patch(f"{RS}.get_citation_formatter", return_value=mock_formatter),+ patch(+ f"{RS}.get_user_db_session",+ side_effect=_fake_session_ctx(mock_session),+ ),+ patch(f"{RS}.cleanup_research_resources"),+ patch(f"{RS}.SocketIOService"),+ patch(f"{RS}.set_search_context"),+ patch(f"{RS}.calculate_duration", return_value=10.0),+ patch(+ "local_deep_research.storage.get_report_storage",+ return_value=mock_storage,+ ),+ patch(+ f"{RS}.extract_links_from_search_results",+ return_value=[],+ ),+ patch(+ "local_deep_research.web.services.research_sources_service.ResearchSourcesService",+ return_value=mock_sources_service,+ ),+ patch(+ "local_deep_research.web.queue.processor_v2.queue_processor",+ MagicMock(),+ ),+ patch("local_deep_research.settings.logger.log_settings"),+ patch(+ "local_deep_research.config.thread_settings.set_settings_context"+ ),+ ):+ raw_fn = _get_raw_run_research_process()+ raw_fn(+ "r_no_dir",+ "test query",+ "quick",+ username="alice",+ settings_snapshot={"search.tool": "searxng"},+ )++ assert research.status == "completed"+ # The per-research output directory must never be created.+ assert not (tmp_path / "research_r_no_dir").exists()+ assert list(tmp_path.iterdir()) == []++class TestRunResearchProcessDetailedMode:"""Detailed/full report mode."""