JITMIND
JITMind is a just-in-time memory system for AI agents that combines self-editing bi-temporal memory, temporal graph reasoning, and an iterative Plan → Search → Integrate → Reflect loop for evidence- grounded answers. It fuses BM25 + dense + graph retrieval with robust ranking and provenance so long-running agents stay accurate as facts change.
File Explorer
- ci.yml
- 01_system_overview.svg
- 02_request_lifecycle.svg
- 03_memory_lifecycle.svg
- 04_bitemporal_model.svg
- 05_retrieval_fusion.svg
- 06_graph_memory_3tier.svg
- 07_ingestion_pipeline.svg
- 08_repo_capabilities_map.svg
- eval_runner_flow.svg
- examples_quickstart_flow.svg
- scripts_ops_flow.svg
- tests_quality_flow.svg
- 01_package_architecture_hero.png
- 02_memory_write_lifecycle.png
- 03_research_runtime_loop.png
- 04_retrieval_scoring_math.png
- 05_graph_semantics_crud.png
- 06_production_readiness_dashboard.png
- 07_package_architecture_alt.png
- 08_memory_write_lifecycle_alt.png
- 09_research_runtime_loop_alt.png
- 10_retrieval_scoring_alt.png
- 11_graph_semantics_crud_alt.png
- 12_production_readiness_dashboard_alt.png
- 01_package_architecture.svg
- 02_memory_write_sequence.svg
- 03_research_runtime_sequence.svg
- 04_schema_relations.svg
- 05_retrieval_scoring_engine.svg
- 06_graph_semantics_crud.svg
- 01_jitmind_graph_memory_overview.png
- 02_jitmind_temporal_validity_panel.png
- 03_jitmind_retrieval_fusion_panel.png
- 04_jitmind_self_editing_lifecycle.png
- 05_jitmind_hierarchical_memory_tiers.png
- 06_jitmind_production_readiness_dashboard.png
- sample_pattern_completion.svg
- sample_provenance_semantics.svg
- sample_temporal_validity.svg
- sample_light_pattern_completion.svg
- sample_light_provenance_semantics.svg
- sample_light_temporal_validity.svg
- README.md
- benchmark_memory_context.py
- bounded-memory-context.md
- temporal-queries.md
- download_narrativeqa.py
- download_ruler.py
- README.md
- hotpotqa_test.py
- locomo_test.py
- narrativeqa_test.py
- ragas_eval.py
- README.md
- ruler_test.py
- run.py
- basic_usage.py
- model_usage.py
- README.md
- ttl_usage.py
- README.md
- __init__.py
- memory_agent.py
- research_agent.py
- __init__.py
- generator.py
- retriever.py
- __init__.py
- cli.py
- ragas_eval.py
- __init__.py
- base.py
- openai_generator.py
- vllm_generator.py
- __init__.py
- graph_store.py
- ontology.py
- utils.py
- __init__.py
- chunking.py
- documents.py
- loaders.py
- pipeline.py
- __init__.py
- replay_buffer.py
- __init__.py
- consolidation.py
- __init__.py
- profile_agent.py
- profile_store.py
- __init__.py
- memory_prompts.py
- profile_prompts.py
- reflection_prompts.py
- research_prompts.py
- summarization_prompts.py
- __init__.py
- base.py
- bm25.py
- cohere_dense.py
- cohere_rerank.py
- dense_retriever.py
- graph_retriever.py
- index_retriever.py
- __init__.py
- advanced_memory.py
- memory.py
- memory_ops.py
- page.py
- result.py
- search.py
- self_rag.py
- tools.py
- ttl_memory.py
- ttl_page.py
- __init__.py
- raptor.py
- __init__.py
- atomic_io.py
- checkpoint.py
- file_lock.py
- retry.py
- __init__.py
- memory_context.py
- README.md
- download_data.sh
- e2e_live_test.py
- e2e_stress_live_test.py
- eval_hotpotqa.sh
- eval_locomo.sh
- eval_narrativeqa.sh
- eval_ruler.sh
- neo4j_local_down.sh
- neo4j_local_up.sh
- README.md
- run_maintenance.py
- test_all.sh
- README.md
- run_ttl_tests.py
- test_ingestion_dedup.py
- test_memory_context.py
- test_package_contract.py
- test_retry_call.py
- test_temporal_queries.py
- test_threaded_file_stores.py
- test_ttl_before_after.py
- test_ttl_memory.py
- test_ttl_page.py
- test_ttl_standalone.py
- .env.example
- .gitignore
- docker-compose.neo4j.yml
- LICENSE
- pyproject.toml
- README.md
- requirements.txt
- setup.py
# Use via CDN
jsDelivrjsDelivr serves any public GitHub repository as a CDN with zero setup. Pick a version and a file to get a ready-to-paste link and snippet.
Command Glossary
Commands referenced in this DOCs, explained below.
eval
View Details ▼
eval
Execute arguments as a single command in the current shell and return its result.
eval "{{echo foo}}"
Call `echo` with the "foo" argument:
eval "{{foo=bar}}"
Set a variable in the current shell:
pip install
View Details ▼
pip install
Install Python packages.
pip install {{package1 package2 ...}}
Install one or more packages:
pip install {{package1 package2 ...}} {{[-U|--upgrade]}}
Upgrade all specified packages to the latest version, installing any that are not already present:
pip install {{package}}=={{version}}
Install a specific version of a package:
python3
View Details ▼
python3
This command is an alias of `python`.
tldr python
View documentation for the original command: