Nicolò Boschi 5fddd9a79c feat: add reflect mode to LoComo benchmark and improve reflect agent (#428) ... * feat: add reflect mode to LoComo benchmark and improve reflect agent - Replace think mode with reflect mode in LoComo benchmark using reflect_async with Budget.HIGH - Add --question-index CLI flag to run a single question by its index - Track and display original question index in logs and visualizer - Update visualizer to show reflect mode results Reflect agent improvements: - tool_recall: always fetch chunks (max_chunk_tokens=1000 min, non-optional) - tool_search_observations: use include_source_facts=True instead of separate DB query - Use model_dump() throughout to avoid manual error-prone dict conversion - Enforce minimum 1000 tokens for max_tokens and max_chunk_tokens in _execute_tool - Fix NoneType error when LLM passes null for mental_model_ids/observation_ids arrays - Add non-conversational constraint to system prompt to prevent follow-up questions - Fix recall_fn Callable type hint to include max_chunk_tokens parameter - Fix main.py missing reranker_zeroentropy fields in HindsightConfig constructor * fix: update tests for reflect tool API changes - source_memory_ids -> source_fact_ids in test_search_observations (MemoryFact.model_dump() field name) - Remove proof_count check (not in MemoryFact, was ObservationResult-specific) - Remove max_results param from tool_recall call (no longer supported) - Fix recall_result["count"] -> len(recall_result["memories"])		2026-02-24 09:48:23 +01:00
..
fixtures	rename to hindsight (#2)	2025-11-25 19:28:26 +01:00
__init__.py	rename to hindsight (#2)	2025-11-25 19:28:26 +01:00
conftest.py	fix: improve claude code and codex for /reflect (#285)	2026-02-04 13:34:45 +01:00
test_admin_backup_restore.py	feat: run db migrations offline (optionally) (#114)	2026-01-07 15:49:51 +01:00
test_agents_api.py	feat: increase customization for reflect, retain and consolidation (#419)	2026-02-23 20:35:32 +01:00
test_async_batch_retain.py	fix: improve async batch retain with large payloads (#366)	2026-02-16 12:51:42 +01:00
test_async_retain_tags.py	fix: improve async batch retain with large payloads (#366)	2026-02-16 12:51:42 +01:00
test_base_path.py	feat: add reverse proxy support (#346)	2026-02-12 10:09:53 +01:00
test_batch_api.py	feat: support Batch API for retain (openai/groq) (#365)	2026-02-16 13:31:50 +01:00
test_batch_api_integration.py	feat: support Batch API for retain (openai/groq) (#365)	2026-02-16 13:31:50 +01:00
test_batch_api_validation.py	feat: support Batch API for retain (openai/groq) (#365)	2026-02-16 13:31:50 +01:00
test_batch_chunking.py	feat: extensions (#54)	2025-12-22 11:05:23 +01:00
test_causal_relations.py	feat: implement hierarchical configuration (system, tenant, bank) (#329)	2026-02-12 13:14:57 +01:00
test_causal_relationships.py	feat: implement hierarchical configuration (system, tenant, bank) (#329)	2026-02-12 13:14:57 +01:00
test_chunking.py	improve retain performances, caching and tests	2025-12-08 18:21:56 +01:00
test_combined_scoring.py	feat: extensions (#54)	2025-12-22 11:05:23 +01:00
test_config_validation.py	fix: hindsight-embed profiles are not loaded correctly (#316)	2026-02-06 17:13:54 +01:00
test_consolidation.py	feat: add reflect mode to LoComo benchmark and improve reflect agent (#428)	2026-02-24 09:48:23 +01:00
test_custom_embedding_dimension.py	feat: add ZeroEntropy reranker provider support (#420)	2026-02-23 10:12:32 +01:00
test_document_tracking.py	fix: document not tracked if has 0 extracted facts (#399)	2026-02-18 17:01:57 +01:00
test_extensions.py	misc: fix vertex/gemini errors and use it for ci tests (#414)	2026-02-20 22:35:38 +01:00
test_fact_extraction_analysis.py	feat: implement hierarchical configuration (system, tenant, bank) (#329)	2026-02-12 13:14:57 +01:00
test_fact_extraction_metadata.py	feat: include doc metadata in fact extraction (#424)	2026-02-23 11:31:02 +01:00
test_fact_extraction_output_ratio.py	feat: implement hierarchical configuration (system, tenant, bank) (#329)	2026-02-12 13:14:57 +01:00
test_fact_extraction_quality.py	feat: implement hierarchical configuration (system, tenant, bank) (#329)	2026-02-12 13:14:57 +01:00
test_fact_ordering.py	misc: fix vertex/gemini errors and use it for ci tests (#414)	2026-02-20 22:35:38 +01:00
test_file_retain.py	feat: accept pdf, images and office files (#390)	2026-02-17 18:15:03 +01:00
test_file_storage_s3.py	misc: fix vertex/gemini errors and use it for ci tests (#414)	2026-02-20 22:35:38 +01:00
test_hierarchical_config.py	feat: enable bank config API by default (#426)	2026-02-24 08:52:42 +01:00
test_http_api_integration.py	fix: improve async batch retain with large payloads (#366)	2026-02-16 12:51:42 +01:00
test_iris_parser.py	feat: add iris as file parser (#395)	2026-02-18 14:09:56 +01:00
test_link_expansion_retrieval.py	feat: implement hierarchical configuration (system, tenant, bank) (#329)	2026-02-12 13:14:57 +01:00
test_link_utils.py	prepare for release	2025-12-03 11:52:25 +01:00
test_litellm_sdk_cross_encoder.py	feat: support litellm-sdk as reranker and embeddings (#357)	2026-02-12 23:53:12 +01:00
test_litellm_sdk_embeddings.py	feat: support litellm-sdk as reranker and embeddings (#357)	2026-02-12 23:53:12 +01:00
test_llm_provider.py	misc: fix vertex/gemini errors and use it for ci tests (#414)	2026-02-20 22:35:38 +01:00
test_llm_token_metrics.py	feat: support for codex and claude-code as llm (#276)	2026-02-02 12:54:44 +01:00
test_llm_tools.py	chore: internal renames (#204)	2026-01-27 09:53:28 +01:00
test_load_large_batch.py	feat: support for other text and vector search pg extensions (#355)	2026-02-12 14:13:04 +01:00
test_main_module.py	fix(python-client): async method parity and server keepalive timeout (#387)	2026-02-17 16:00:14 +01:00
test_mcp_endpoint_routing.py	Harden MCP server: fix routing, validation, and usage metering (#341)	2026-02-11 10:41:20 +01:00
test_mcp_extension.py	Harden MCP server: fix routing, validation, and usage metering (#341)	2026-02-11 10:41:20 +01:00
test_mcp_routing.py	Harden MCP server: fix routing, validation, and usage metering (#341)	2026-02-11 10:41:20 +01:00
test_mcp_tools.py	Harden MCP server: fix routing, validation, and usage metering (#341)	2026-02-11 10:41:20 +01:00
test_mental_model_hooks.py	Add extension hooks for mental model operations (#260)	2026-01-31 09:30:53 -05:00
test_mental_models.py	misc: fix vertex/gemini errors and use it for ci tests (#414)	2026-02-20 22:35:38 +01:00
test_metrics.py	chore: remove dead code (#245)	2026-01-30 09:16:32 +01:00
test_mpfp_retrieval.py	fix: improve mpfp retrieval (#146)	2026-01-12 18:58:05 +01:00
test_multilingual.py	misc: fix vertex/gemini errors and use it for ci tests (#414)	2026-02-20 22:35:38 +01:00
test_observations.py	feat: implement hierarchical configuration (system, tenant, bank) (#329)	2026-02-12 13:14:57 +01:00
test_per_operation_llm_config.py	feat: add more config options for llm retries (#234)	2026-01-29 17:50:43 +01:00
test_provider_default_models.py	feat: switch default model to gpt-4o-mini (#410)	2026-02-19 18:43:52 +01:00
test_query_analyzer.py	fix: ci and ui build (#9)	2025-12-03 15:47:53 +01:00
test_recall_chunks_independence.py	feat: allow chunks-only in recall (max_tokens=0) (#364)	2026-02-13 16:56:35 +01:00
test_reflect_agent.py	fix: graph endpoint not showing links for observations (#214)	2026-01-28 14:51:25 +01:00
test_reflect_empty_based_on.py	fix: resolve based_on schema/serialization issues in reflect API (#348)	2026-02-12 11:26:12 +01:00
test_reflect_tracing.py	feat: add otel traceability (#330)	2026-02-10 12:20:48 +01:00
test_reflections.py	feat: improve mental models ux on control plane (#297)	2026-02-04 15:49:03 +01:00
test_reranker_error_handling.py	fix: reranker crashes on provider error (#403)	2026-02-19 11:38:37 +01:00
test_retain.py	feat: increase customization for reflect, retain and consolidation (#419)	2026-02-23 20:35:32 +01:00
test_schema_isolation.py	fix: improve mental model consolidation (#197)	2026-01-26 09:54:25 +01:00
test_search_trace.py	feat: extensions (#54)	2025-12-22 11:05:23 +01:00
test_server_module.py	fix: improve mental model consolidation (#197)	2026-01-26 09:54:25 +01:00
test_sql_schema_safety.py	chore: internal renames (#204)	2026-01-27 09:53:28 +01:00
test_supabase_tenant.py	Add Supabase tenant extension as built-in (#267)	2026-02-09 10:16:47 +01:00
test_tags_visibility.py	chore: cleanup benchmarks runner with old flags (#212)	2026-01-28 13:22:48 +01:00
test_tei_cross_encoder.py	fix: add defensive error handling to PyTorch device detection (#221)	2026-01-28 18:14:54 +01:00
test_temporal_ranges.py	feat(mcp): add timestamp to retain (#190)	2026-01-23 16:00:43 +01:00
test_think.py	chore: remove dead code (#245)	2026-01-30 09:16:32 +01:00
test_tracing.py	misc: fix vertex/gemini errors and use it for ci tests (#414)	2026-02-20 22:35:38 +01:00
test_tracing_integration.py	feat: add otel traceability (#330)	2026-02-10 12:20:48 +01:00
test_tracing_spans_verification.py	feat: add otel traceability (#330)	2026-02-10 12:20:48 +01:00
test_vertexai_provider.py	misc: fix vertex/gemini errors and use it for ci tests (#414)	2026-02-20 22:35:38 +01:00
test_worker.py	feat: accept pdf, images and office files (#390)	2026-02-17 18:15:03 +01:00