HKUDS/LightRAG

35.1

Adjusted Score

35.1

Raw Score

100%

Time Factor

2026-08-28

Last Push

39.3K

Stars

Python

Language

404.9K

Lines of Code

993

Files

10.4K

Pattern Hits

2026-08-29

Scan Date

0.19

HC Hit Rate

What These Metrics Mean

Adjusted Score: Primary synthetic code indicator. Raw score normalised per 1,000 lines of code and multiplied by the temporal discount factor. This is the definitive comparative metric — use it to rank repositories by AI authorship density.
Raw Score: The unmodified sum of all severity-weighted, context-multiplied pattern match scores before temporal discounting. Reflects the absolute signal strength independent of when the repository was last active.
Time Factor: The temporal discount multiplier (0–100%) applied to the raw score. Repositories last updated before ChatGPT's launch (Nov 2022) receive a 5% factor. Full signal is only assigned to repositories active in the post-adoption era (Jan 2024+).
Pattern Hits: Total count of individual pattern matches across all files and categories. A high hit count with a low score may indicate a very large codebase with isolated AI snippets; a low count with a high score indicates dense, concentrated AI signatures.
HC Hit Rate: High+Critical pattern hits per file, averaged across the repository. This orthogonal signal catches repositories where a few files are densely packed with high-severity AI tells — a strong indicator even when the normalised score appears moderate due to codebase size.
Lines of Code / Files: Total lines and files analysed. The scanner examines 94 file extensions. These denominators are used to normalise the score, enabling fair comparison between repositories of vastly different sizes.

Score History

This chart maps the temporal evolution of the adjusted synthetic code score across successive scan runs. An upward trajectory indicates ongoing incorporation of AI-generated code or expanding LLM-assisted scaffolding; a stable or declining trajectory may reflect active human refactoring, code removal, or the adoption of stricter authorship policies. The dashed secondary line (right axis) independently tracks total raw pattern hit count, which can diverge from the normalised score when codebase size changes significantly between scans.

Severity Breakdown

Classifies detected patterns by their diagnostic confidence and structural impact. CRITICAL patterns (coefficient 10) represent definitive synthetic signatures — hallucinated imports, explicit LLM attribution metadata — virtually never produced by human authors. HIGH (5) indicates strong structural tells such as cross-file repetition or cross-linguistic idioms. MEDIUM (2) covers recognisable conversational padding and AI-specific vocabulary. LOW (1) captures subtle indicators like tautological comments and generic boilerplate that require density to carry independent signal.

CRITICAL 0HIGH 193MEDIUM 1621LOW 8572

Directory Score Breakdown

This horizontal bar chart decomposes the repository's raw synthetic code score by top-level directory, allowing you to pinpoint precisely which modules or components carry the highest AI authorship density. Directories with disproportionately high scores relative to their size warrant targeted manual review: concentrated AI signatures often trace back to mass-generated configuration layers, auto-ported test suites, LLM-scaffolded boilerplate classes, or entire subsystems authored under heavy copilot assistance. Use this view to prioritise your human code-review effort.

Pattern Findings

The scanner identified 10386 distinct pattern matches across 21 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a statistically significant AI authorship indicator. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.

Reading the findings table: The Severity column indicates the diagnostic confidence level (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — comment-context matches receive a ×1.5 weight because LLMs systematically over-annotate. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier as dense clusters constitute far stronger evidence of synthetic authorship than isolated hits.

Hyper-Verbose Identifiers6469 hits · 5659 pts

Severity	File	Line	Snippet	Context
LOW	reproduce/Step_3_openai_compatible.py	57	def run_queries_and_save_to_json(	CODE
LOW	reproduce/Step_3.py	26	def run_queries_and_save_to_json(	CODE
LOW	lightrag/rerank.py	36	def chunk_documents_for_rerank(	CODE
LOW	lightrag/rerank.py	142	async def achunk_documents_for_rerank(	CODE
LOW	lightrag/addon_params.py	30	def _emit_deprecated_addon_warnings(params: Mapping[str, Any]) -> None:	CODE
LOW	lightrag/table_markup.py	83	def header_grid_to_thead_html(grid: list[list[str]]) -> str:	CODE
LOW	lightrag/llm_roles.py	130	def register_role_llm_builder(	CODE
LOW⚡	lightrag/llm_roles.py	162	def _get_effective_role_llm_kwargs(self, role: str) -> dict[str, Any]:	CODE
LOW⚡	lightrag/llm_roles.py	170	def _get_effective_role_llm_timeout(self, role: str) -> int:	CODE
LOW⚡	lightrag/llm_roles.py	174	def _get_effective_role_llm_max_async(self, role: str) -> int:	CODE
LOW	lightrag/llm_roles.py	212	def _rebuild_single_role_llm_func(self, role: str) -> None:	CODE
LOW	lightrag/llm_roles.py	228	def _schedule_retired_llm_queue_cleanup(	CODE
LOW⚡	lightrag/llm_roles.py	256	def _finalize_retired_llm_queue_cleanup(self, task: asyncio.Task) -> None:	CODE
LOW⚡	lightrag/llm_roles.py	265	async def wait_for_retired_llm_queues(self) -> None:	CODE
LOW	lightrag/llm_roles.py	276	def _apply_llm_role_config_update(	CODE
LOW	lightrag/llm_roles.py	578	async def get_embedding_queue_status(self) -> dict[str, Any]:	CODE
LOW⚡	lightrag/lightrag.py	1028	def _set_runtime_addon_params(self, addon_params: Mapping[str, Any] \| None) -> None:	CODE
LOW⚡	lightrag/lightrag.py	1032	def _apply_chunk_size_overlay(self) -> None:	CODE
LOW	lightrag/lightrag.py	1151	def _refresh_addon_params_cache(self) -> None:	CODE
LOW	lightrag/lightrag.py	1171	def _ensure_addon_params_cache(self) -> None:	CODE
LOW	lightrag/lightrag.py	1217	def _build_role_llm_cache_identity(	CODE
LOW	lightrag/lightrag.py	1621	def _get_parse_native_executor(self) -> ThreadPoolExecutor:	CODE
LOW	lightrag/lightrag.py	1638	def _shutdown_parser_executor(self) -> None:	CODE
LOW	lightrag/lightrag.py	2155	def _operation_truncation_record() -> dict[str, Any] \| None:	CODE
LOW	lightrag/lightrag.py	2169	def _terminal_truncation_kwargs() -> dict[str, Any]:	CODE
LOW	lightrag/lightrag.py	2192	async def _journal_truncation_write_ahead(	CODE
LOW	lightrag/lightrag.py	2603	async def _upsert_custom_chunk_status(	CODE
LOW	lightrag/lightrag.py	2693	async def _union_doc_recovery_anchors(	CODE
LOW	lightrag/lightrag.py	2733	async def arollback_failed_custom_chunk_patches(	CODE
LOW	lightrag/lightrag.py	2900	async def _rollback_one_custom_chunk_patch(	CODE
LOW	lightrag/lightrag.py	3029	async def _persist_custom_chunk_recovery_warning(	CODE
LOW	lightrag/lightrag.py	3153	async def _prune_doc_recovery_anchors(	CODE
LOW	lightrag/lightrag.py	3223	async def _process_extract_entities(	CODE
LOW	lightrag/lightrag.py	3275	async def _discard_pending_index_ops(	CODE
LOW	lightrag/lightrag.py	3423	async def _insert_done_with_cleanup(self) -> None:	CODE
LOW	lightrag/lightrag.py	3499	def _normalize_custom_kg_entity_name(value: Any, *, field: str) -> str:	CODE
LOW	lightrag/lightrag.py	4361	async def _update_delete_retry_state(	CODE
LOW	lightrag/lightrag.py	4443	async def _get_existing_llm_cache_ids(self, cache_ids: list[str]) -> list[str]:	CODE
LOW	lightrag/lightrag.py	4610	def _patch_journal_candidates(	CODE
LOW	lightrag/lightrag.py	4640	async def _resolve_purge_recovery_proof(	CODE
LOW	lightrag/lightrag.py	4848	def _raise_missing_recovery_proof(	CODE
LOW	lightrag/lightrag.py	5227	async def _purge_derived_kg_contributions(	CODE
LOW	lightrag/lightrag.py	6433	async def _raise_if_recovery_required(self) -> None:	CODE
LOW	lightrag/operate.py	183	def _get_relationship_vdb_timeout_seconds(global_config: dict[str, Any]) -> float:	CODE
LOW	lightrag/operate.py	200	def _format_relation_edge_label(edge_key: tuple[str, str] \| list[str]) -> str:	CODE
LOW	lightrag/operate.py	208	def _truncate_entity_identifier(	CODE
LOW	lightrag/operate.py	248	def _truncate_section_context(	CODE
LOW	lightrag/operate.py	371	async def _handle_entity_relation_summary(	CODE
LOW	lightrag/operate.py	660	def _normalize_and_validate_entity_type(entity_type: str, context: str) -> str \| None:	CODE
LOW	lightrag/operate.py	708	def _handle_single_entity_extraction(	CODE
LOW	lightrag/operate.py	772	def _handle_single_relationship_extraction(	CODE
LOW	lightrag/operate.py	852	def _normalize_text_extraction_record_attributes(	CODE
LOW	lightrag/operate.py	889	def _looks_like_json_extraction_result(result: str) -> bool:	CODE
LOW	lightrag/operate.py	905	async def _process_json_extraction_result(	CODE
LOW	lightrag/operate.py	1091	async def rebuild_knowledge_from_chunks(	CODE
LOW	lightrag/operate.py	1312	async def _locked_rebuild_relationship(src, tgt, chunk_ids):	CODE
LOW	lightrag/operate.py	1404	async def _get_cached_extraction_results(	CODE
LOW	lightrag/operate.py	1497	async def _process_extraction_result(	CODE
LOW	lightrag/operate.py	1628	async def _rebuild_from_extraction_result(	CODE
LOW	lightrag/operate.py	1684	def _surviving_chunk_file_paths(	CODE
6409 more matches not shown…

Decorative Section Separators1425 hits · 5348 pts

Severity	File	Line	Snippet	Context
MEDIUM	lightrag/constants.py	783	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/constants.py	787	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/utils.py	3662	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/utils.py	3664	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils.py	2698	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils.py	2700	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils.py	2961	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils.py	2963	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils.py	3016	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils.py	3018	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils.py	3233	# -----------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils.py	3235	# -----------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils.py	3468	# -----------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils.py	3470	# -----------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/pipeline.py	660	# ============================================================	COMMENT
MEDIUM⚡	lightrag/pipeline.py	662	# ============================================================	COMMENT
MEDIUM⚡	lightrag/pipeline.py	2194	# ============================================================	COMMENT
MEDIUM⚡	lightrag/pipeline.py	2196	# ============================================================	COMMENT
MEDIUM⚡	lightrag/pipeline.py	5778	# ============================================================	COMMENT
MEDIUM⚡	lightrag/pipeline.py	5780	# ============================================================	COMMENT
MEDIUM⚡	lightrag/pipeline.py	6056	# ============================================================	COMMENT
MEDIUM⚡	lightrag/pipeline.py	6058	# ============================================================	COMMENT
MEDIUM	lightrag/pipeline.py	558	# ============================================================	COMMENT
MEDIUM	lightrag/pipeline.py	560	# ============================================================	COMMENT
MEDIUM	lightrag/pipeline.py	3070	# ============================================================	COMMENT
MEDIUM	lightrag/pipeline.py	3076	# ============================================================	COMMENT
MEDIUM	lightrag/pipeline.py	4169	# ============================================================	COMMENT
MEDIUM	lightrag/pipeline.py	4171	# ============================================================	COMMENT
MEDIUM	lightrag/pipeline.py	4751	# ============================================================	COMMENT
MEDIUM	lightrag/pipeline.py	4753	# ============================================================	COMMENT
MEDIUM	lightrag/pipeline.py	6391	# ============================================================	COMMENT
MEDIUM	lightrag/pipeline.py	6393	# ============================================================	COMMENT
MEDIUM⚡	lightrag/multimodal_context.py	88	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/multimodal_context.py	92	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/multimodal_context.py	685	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/multimodal_context.py	687	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/multimodal_context.py	132	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/multimodal_context.py	135	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/multimodal_context.py	188	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/multimodal_context.py	190	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/multimodal_context.py	267	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/multimodal_context.py	269	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/multimodal_context.py	461	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/multimodal_context.py	463	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/utils_pipeline.py	1281	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/utils_pipeline.py	1283	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/utils_pipeline.py	1308	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/utils_pipeline.py	1310	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils_pipeline.py	1156	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils_pipeline.py	1158	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils_pipeline.py	1206	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils_pipeline.py	1208	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils_pipeline.py	1346	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/utils_pipeline.py	1348	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/base.py	1455	# ------------------------------------------------------------------	COMMENT
MEDIUM⚡	lightrag/base.py	1461	# ------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/tools/rebuild_vdb.py	575	# ------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/tools/rebuild_vdb.py	577	# ------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/tools/rebuild_vdb.py	760	# ------------------------------------------------------------------	COMMENT
MEDIUM	lightrag/tools/rebuild_vdb.py	762	# ------------------------------------------------------------------	COMMENT
1365 more matches not shown…

Excessive Try-Catch Wrapping797 hits · 622 pts

Severity	File	Line	Snippet	Context
LOW	reproduce/Step_1_openai_compatible.py	50	except Exception as e:	CODE
LOW	reproduce/Step_0.py	41	except Exception as e:	CODE
MEDIUM	reproduce/Step_0.py	42	print(f"An error occurred while processing file {filename}: {e}")	CODE
LOW	reproduce/Step_0.py	54	except Exception as e:	CODE
MEDIUM	reproduce/Step_0.py	55	print(f"An error occurred while saving to the file {output_filename}: {e}")	CODE
LOW	reproduce/Step_1.py	19	except Exception as e:	CODE
LOW	reproduce/Step_3_openai_compatible.py	53	except Exception as e:	CODE
MEDIUM	reproduce/Step_3_openai_compatible.py	49	def process_query(query_text, rag_instance, query_param):	CODE
LOW	reproduce/Step_3.py	22	except Exception as e:	CODE
MEDIUM	reproduce/Step_3.py	18	def process_query(query_text, rag_instance, query_param):	CODE
LOW	lightrag/pipeline_metrics.py	97	except Exception as metric_error: # pragma: no cover - defensive	CODE
LOW	lightrag/pipeline_metrics.py	115	except Exception as metric_error: # pragma: no cover - defensive	CODE
LOW	lightrag/rerank.py	84	except Exception as e:	CODE
LOW	lightrag/rerank.py	624	except Exception as e:	CODE
LOW	lightrag/rerank.py	639	except Exception as e:	CODE
LOW	lightrag/rerank.py	654	except Exception as e:	CODE
LOW⚡	lightrag/llm_roles.py	262	except Exception as e:	CODE
LOW	lightrag/llm_roles.py	348	except Exception:	CODE
LOW	lightrag/lightrag.py	15	except Exception: # pragma: no cover - optional dependency	CODE
LOW	lightrag/lightrag.py	1685	except Exception as e:	CODE
LOW	lightrag/lightrag.py	2451	except Exception as bookkeeping_error:	CODE
LOW	lightrag/lightrag.py	2483	except Exception as ingress_error:	CODE
LOW	lightrag/lightrag.py	2513	except Exception as probe_error:	CODE
LOW	lightrag/lightrag.py	2584	except Exception as drain_error:	CODE
LOW	lightrag/lightrag.py	2824	except Exception as rollback_error:	CODE
LOW	lightrag/lightrag.py	3241	except Exception as e:	CODE
LOW	lightrag/lightrag.py	3349	except Exception as e:	CODE
LOW	lightrag/lightrag.py	3353	except Exception as e:	CODE
LOW	lightrag/lightrag.py	3381	except Exception as e:	CODE
LOW	lightrag/lightrag.py	3889	except Exception as e:	CODE
LOW	lightrag/lightrag.py	4316	except Exception as e:	CODE
LOW	lightrag/lightrag.py	4458	except Exception as verification_error:	CODE
LOW	lightrag/lightrag.py	4492	except Exception as e:	CODE
LOW	lightrag/lightrag.py	5171	except Exception as e:	CODE
LOW	lightrag/lightrag.py	5203	except Exception as e:	CODE
LOW	lightrag/lightrag.py	5340	except Exception as e:	CODE
LOW	lightrag/lightrag.py	5525	except Exception as e:	CODE
LOW	lightrag/lightrag.py	5563	except Exception as e:	CODE
LOW	lightrag/lightrag.py	5629	except Exception as e:	CODE
LOW	lightrag/lightrag.py	5642	except Exception as e:	CODE
LOW	lightrag/lightrag.py	5667	except Exception as e:	CODE
LOW	lightrag/lightrag.py	5680	except Exception as e:	CODE
LOW	lightrag/lightrag.py	5985	except Exception as cache_err:	CODE
LOW	lightrag/lightrag.py	5998	except Exception as e:	CODE
LOW	lightrag/lightrag.py	6055	except Exception as cache_collect_error:	CODE
LOW	lightrag/lightrag.py	6074	except Exception as status_write_error:	CODE
LOW	lightrag/lightrag.py	6168	except Exception as cache_delete_error:	CODE
LOW	lightrag/lightrag.py	6196	except Exception as e:	CODE
LOW	lightrag/lightrag.py	6229	except Exception as status_update_error:	CODE
LOW	lightrag/lightrag.py	6243	except Exception as e:	CODE
LOW	lightrag/lightrag.py	6262	except Exception as status_update_error:	CODE
LOW	lightrag/lightrag.py	6301	except Exception as ingress_error:	CODE
LOW	lightrag/lightrag.py	6324	except Exception as probe_error:	CODE
LOW	lightrag/lightrag.py	6363	except Exception as persistence_error:	CODE
LOW	lightrag/lightrag.py	6418	except Exception as drain_error:	CODE
LOW	lightrag/operate.py	765	except Exception as e:	CODE
LOW	lightrag/operate.py	845	except Exception as e:	CODE
LOW	lightrag/operate.py	1006	except Exception as e:	CODE
LOW	lightrag/operate.py	1082	except Exception as e:	CODE
LOW	lightrag/operate.py	1258	except Exception as e:	CODE
737 more matches not shown…

Cross-File Repetition85 hits · 425 pts

Severity	File	Line	Snippet	Context
HIGH	lightrag/base.py	0	get multiple vector data by their ids (read-your-writes), preserving order. args: ids: list of unique identifiers return	STRING
HIGH	lightrag/kg/deprecated/chroma_impl.py	0	get multiple vector data by their ids (read-your-writes), preserving order. args: ids: list of unique identifiers return	STRING
HIGH	lightrag/kg/nano_vector_db_impl.py	0	get multiple vector data by their ids (read-your-writes), preserving order. args: ids: list of unique identifiers return	STRING
HIGH	lightrag/base.py	0	get all edges in the graph. returns: a list of all edges, where each edge is a dictionary of its properties	STRING
HIGH	lightrag/kg/networkx_impl.py	0	get all edges in the graph. returns: a list of all edges, where each edge is a dictionary of its properties	STRING
HIGH	lightrag/kg/mongo_impl.py	0	get all edges in the graph. returns: a list of all edges, where each edge is a dictionary of its properties	STRING
HIGH	lightrag/kg/memgraph_impl.py	0	get all edges in the graph. returns: a list of all edges, where each edge is a dictionary of its properties	STRING
HIGH	lightrag/kg/neo4j_impl.py	0	get all edges in the graph. returns: a list of all edges, where each edge is a dictionary of its properties	STRING
HIGH	lightrag/kg/qdrant_impl.py	0	buffered vector upsert waiting for embedding and/or bulk flush.	STRING
HIGH	lightrag/kg/mongo_impl.py	0	buffered vector upsert waiting for embedding and/or bulk flush.	STRING
HIGH	lightrag/kg/opensearch_impl.py	0	buffered vector upsert waiting for embedding and/or bulk flush.	STRING
HIGH	lightrag/kg/milvus_impl.py	0	buffered vector upsert waiting for embedding and/or bulk flush.	STRING
HIGH	lightrag/kg/qdrant_impl.py	0	discard buffered upserts/deletes (pipeline aborting on error).	STRING
HIGH	lightrag/kg/mongo_impl.py	0	discard buffered upserts/deletes (pipeline aborting on error).	STRING
HIGH	lightrag/kg/postgres_impl.py	0	discard buffered upserts/deletes (pipeline aborting on error).	STRING
HIGH	lightrag/kg/opensearch_impl.py	0	discard buffered upserts/deletes (pipeline aborting on error).	STRING
HIGH	lightrag/kg/milvus_impl.py	0	discard buffered upserts/deletes (pipeline aborting on error).	STRING
HIGH	lightrag/kg/qdrant_impl.py	0	buffer an entity vector delete by computing its hash id.	STRING
HIGH	lightrag/kg/mongo_impl.py	0	buffer an entity vector delete by computing its hash id.	STRING
HIGH	lightrag/kg/opensearch_impl.py	0	buffer an entity vector delete by computing its hash id.	STRING
HIGH	lightrag/kg/milvus_impl.py	0	buffer an entity vector delete by computing its hash id.	STRING
HIGH	lightrag/kg/qdrant_impl.py	0	get multiple vector data by their ids (read-your-writes), preserving order.	STRING
HIGH	lightrag/kg/mongo_impl.py	0	get multiple vector data by their ids (read-your-writes), preserving order.	STRING
HIGH	lightrag/kg/milvus_impl.py	0	get multiple vector data by their ids (read-your-writes), preserving order.	STRING
HIGH	lightrag/kg/networkx_impl.py	0	retrieve a connected subgraph of nodes where the label includes the specified `node_label`. args: node_label: label of t	STRING
HIGH	lightrag/kg/postgres_impl.py	0	retrieve a connected subgraph of nodes where the label includes the specified `node_label`. args: node_label: label of t	STRING
HIGH	lightrag/kg/memgraph_impl.py	0	retrieve a connected subgraph of nodes where the label includes the specified `node_label`. args: node_label: label of t	STRING
HIGH	lightrag/kg/neo4j_impl.py	0	retrieve a connected subgraph of nodes where the label includes the specified `node_label`. args: node_label: label of t	STRING
HIGH	lightrag/kg/networkx_impl.py	0	get all nodes in the graph. returns: a list of all nodes, where each node is a dictionary of its properties	STRING
HIGH	lightrag/kg/mongo_impl.py	0	get all nodes in the graph. returns: a list of all nodes, where each node is a dictionary of its properties	STRING
HIGH	lightrag/kg/postgres_impl.py	0	get all nodes in the graph. returns: a list of all nodes, where each node is a dictionary of its properties	STRING
HIGH	lightrag/kg/memgraph_impl.py	0	get all nodes in the graph. returns: a list of all nodes, where each node is a dictionary of its properties	STRING
HIGH	lightrag/kg/neo4j_impl.py	0	get all nodes in the graph. returns: a list of all nodes, where each node is a dictionary of its properties	STRING
HIGH	lightrag/kg/mongo_impl.py	0	check if the storage is empty for the current workspace and namespace returns: bool: true if storage is empty, false oth	STRING
HIGH	lightrag/kg/postgres_impl.py	0	check if the storage is empty for the current workspace and namespace returns: bool: true if storage is empty, false oth	STRING
HIGH	lightrag/kg/redis_impl.py	0	check if the storage is empty for the current workspace and namespace returns: bool: true if storage is empty, false oth	STRING
HIGH	lightrag/kg/mongo_impl.py	0	get documents with pagination support args: status_filter: filter by document status, none for all statuses page: page n	STRING
HIGH	lightrag/kg/postgres_impl.py	0	get documents with pagination support args: status_filter: filter by document status, none for all statuses page: page n	STRING
HIGH	lightrag/kg/json_doc_status_impl.py	0	get documents with pagination support args: status_filter: filter by document status, none for all statuses page: page n	STRING
HIGH	lightrag/kg/redis_impl.py	0	get documents with pagination support args: status_filter: filter by document status, none for all statuses page: page n	STRING
HIGH	lightrag/kg/mongo_impl.py	0	get counts of documents in each status for all documents returns: dictionary mapping status names to counts, including '	STRING
HIGH	lightrag/kg/postgres_impl.py	0	get counts of documents in each status for all documents returns: dictionary mapping status names to counts, including '	STRING
HIGH	lightrag/kg/json_doc_status_impl.py	0	get counts of documents in each status for all documents returns: dictionary mapping status names to counts, including '	STRING
HIGH	lightrag/kg/redis_impl.py	0	get counts of documents in each status for all documents returns: dictionary mapping status names to counts, including '	STRING
HIGH	lightrag/kg/mongo_impl.py	0	get document by file path args: file_path: the file path to search for returns: union[dict[str, any], none]: document da	STRING
HIGH	lightrag/kg/postgres_impl.py	0	get document by file path args: file_path: the file path to search for returns: union[dict[str, any], none]: document da	STRING
HIGH	lightrag/kg/redis_impl.py	0	get document by file path args: file_path: the file path to search for returns: union[dict[str, any], none]: document da	STRING
HIGH	lightrag/kg/json_doc_status_impl.py	0	get document by file path args: file_path: the file path to search for returns: union[dict[str, any], none]: document da	STRING
HIGH	lightrag/kg/mongo_impl.py	0	deterministic digest over candidate doc ids in stable sort order.	STRING
HIGH	lightrag/kg/postgres_impl.py	0	deterministic digest over candidate doc ids in stable sort order.	STRING
HIGH	lightrag/kg/opensearch_impl.py	0	deterministic digest over candidate doc ids in stable sort order.	STRING
HIGH	lightrag/kg/json_doc_status_impl.py	0	deterministic digest over candidate doc ids in stable sort order.	STRING
HIGH	lightrag/kg/redis_impl.py	0	deterministic digest over candidate doc ids in stable sort order.	STRING
HIGH	lightrag/kg/mongo_impl.py	0	get the total degree (sum of relationships) of two nodes. args: src_id: label of the source node tgt_id: label of the ta	STRING
HIGH	lightrag/kg/memgraph_impl.py	0	get the total degree (sum of relationships) of two nodes. args: src_id: label of the source node tgt_id: label of the ta	STRING
HIGH	lightrag/kg/neo4j_impl.py	0	get the total degree (sum of relationships) of two nodes. args: src_id: label of the source node tgt_id: label of the ta	STRING
HIGH	lightrag/kg/mongo_impl.py	0	check existence of multiple nodes in a single unwind query. args: node_ids: list of node ids to check. returns: set of n	STRING
HIGH	lightrag/kg/postgres_impl.py	0	check existence of multiple nodes in a single unwind query. args: node_ids: list of node ids to check. returns: set of n	STRING
HIGH	lightrag/kg/memgraph_impl.py	0	check existence of multiple nodes in a single unwind query. args: node_ids: list of node ids to check. returns: set of n	STRING
HIGH	lightrag/kg/neo4j_impl.py	0	check existence of multiple nodes in a single unwind query. args: node_ids: list of node ids to check. returns: set of n	STRING
25 more matches not shown…

Unused Imports357 hits · 327 pts

Severity	File	Line	Context
LOW	lightrag/pipeline_metrics.py	34	CODE
LOW	lightrag/rerank.py	1	CODE
LOW	lightrag/file_atomic.py	29	CODE
LOW	lightrag/addon_params.py	15	CODE
LOW	lightrag/table_markup.py	17	CODE
LOW	lightrag/llm_roles.py	11	CODE
LOW	lightrag/lightrag.py	1	CODE
LOW	lightrag/lightrag.py	201	CODE
LOW	lightrag/__init__.py	3	CODE
LOW	lightrag/types.py	1	CODE
LOW	lightrag/operate.py	1	CODE
LOW	lightrag/utils.py	1	CODE
LOW	lightrag/utils.py	384	CODE
LOW	lightrag/utils.py	384	CODE
LOW	lightrag/utils.py	384	CODE
LOW	lightrag/pipeline.py	12	CODE
LOW	lightrag/pipeline.py	122	CODE
LOW	lightrag/pipeline.py	122	CODE
LOW	lightrag/storage_migrations.py	13	CODE
LOW	lightrag/exceptions.py	1	CODE
LOW	lightrag/multimodal_context.py	64	CODE
LOW	lightrag/utils_graph.py	1	CODE
LOW	lightrag/prompt.py	1	CODE
LOW	lightrag/utils_pipeline.py	10	CODE
LOW	lightrag/namespace.py	1	CODE
LOW	lightrag/chunk_schema.py	24	CODE
LOW	lightrag/base.py	1	CODE
LOW	lightrag/prompt_multimodal.py	31	CODE
LOW	lightrag/tools/kg_integrity_repair.py	38	CODE
LOW	lightrag/tools/source_conflict_repair.py	92	CODE
LOW	lightrag/tools/source_conflict_repair.py	727	CODE
LOW	lightrag/tools/migrate_graph_storage.py	58	CODE
LOW	lightrag/llm/_error_utils.py	14	CODE
LOW	lightrag/llm/gemini.py	10	CODE
LOW	lightrag/llm/_vision_utils.py	20	CODE
LOW	lightrag/llm/azure_openai.py	12	CODE
LOW	lightrag/llm/azure_openai.py	12	CODE
LOW	lightrag/llm/azure_openai.py	12	CODE
LOW	lightrag/parser/plugins.py	25	CODE
LOW	lightrag/parser/param_schema.py	30	CODE
LOW	lightrag/parser/native_dispatch.py	16	CODE
LOW	lightrag/parser/registry.py	16	CODE
LOW	lightrag/parser/registry.py	32	CODE
LOW	lightrag/parser/_html_table.py	13	CODE
LOW	lightrag/parser/llm_bridge.py	17	CODE
LOW	lightrag/parser/cli.py	20	CODE
LOW	lightrag/parser/_markdown.py	10	CODE
LOW	lightrag/parser/routing.py	1	CODE
LOW	lightrag/parser/debug.py	17	CODE
LOW	lightrag/parser/noop.py	18	CODE
LOW	lightrag/parser/native_base.py	15	CODE
LOW	lightrag/parser/native_base.py	33	CODE
LOW	lightrag/parser/exceptions.py	10	CODE
LOW	lightrag/parser/base.py	22	CODE
LOW	lightrag/parser/base.py	31	CODE
LOW	lightrag/parser/markdown/ir_builder.py	20	CODE
LOW	lightrag/parser/markdown/ir_builder.py	23	CODE
LOW	lightrag/parser/markdown/parser.py	41	CODE
LOW	lightrag/parser/markdown/parser.py	87	CODE
LOW	lightrag/parser/markdown/raw_cache.py	22	CODE
297 more matches not shown…

Deep Nesting379 hits · 261 pts

Severity	File	Line	Context
LOW	reproduce/Step_0.py	7	CODE
LOW	reproduce/Step_3_openai_compatible.py	57	CODE
LOW	reproduce/Step_3.py	26	CODE
LOW	lightrag/rerank.py	36	CODE
LOW	lightrag/rerank.py	165	CODE
LOW	lightrag/rerank.py	240	CODE
LOW	lightrag/table_markup.py	133	CODE
LOW	lightrag/llm_roles.py	276	CODE
LOW	lightrag/lightrag.py	1232	CODE
LOW	lightrag/lightrag.py	1656	CODE
LOW	lightrag/lightrag.py	1858	CODE
LOW	lightrag/lightrag.py	3153	CODE
LOW	lightrag/lightrag.py	3464	CODE
LOW	lightrag/lightrag.py	4184	CODE
LOW	lightrag/lightrag.py	5227	CODE
LOW	lightrag/lightrag.py	5689	CODE
LOW	lightrag/operate.py	248	CODE
LOW	lightrag/operate.py	371	CODE
LOW	lightrag/operate.py	1091	CODE
LOW	lightrag/operate.py	1497	CODE
LOW	lightrag/operate.py	1718	CODE
LOW	lightrag/operate.py	1990	CODE
LOW	lightrag/operate.py	2392	CODE
LOW	lightrag/operate.py	2745	CODE
LOW	lightrag/operate.py	3477	CODE
LOW	lightrag/operate.py	3905	CODE
LOW	lightrag/operate.py	4817	CODE
LOW	lightrag/operate.py	5043	CODE
LOW	lightrag/operate.py	6005	CODE
LOW	lightrag/operate.py	6256	CODE
LOW	lightrag/operate.py	4012	CODE
LOW	lightrag/utils.py	150	CODE
LOW	lightrag/utils.py	1011	CODE
LOW	lightrag/utils.py	3921	CODE
LOW	lightrag/utils.py	4237	CODE
LOW	lightrag/utils.py	4363	CODE
LOW	lightrag/utils.py	5051	CODE
LOW	lightrag/utils.py	5595	CODE
LOW	lightrag/utils.py	5623	CODE
LOW	lightrag/utils.py	5861	CODE
LOW	lightrag/utils.py	6060	CODE
LOW	lightrag/utils.py	6191	CODE
LOW	lightrag/utils.py	6319	CODE
LOW	lightrag/utils.py	6698	CODE
LOW	lightrag/utils.py	1058	CODE
LOW	lightrag/utils.py	3363	CODE
LOW	lightrag/utils.py	3488	CODE
LOW	lightrag/utils.py	3573	CODE
LOW	lightrag/utils.py	1269	CODE
LOW	lightrag/utils.py	1368	CODE
LOW	lightrag/utils.py	1473	CODE
LOW	lightrag/utils.py	1631	CODE
LOW	lightrag/utils.py	1730	CODE
LOW	lightrag/pipeline.py	664	CODE
LOW	lightrag/pipeline.py	1644	CODE
LOW	lightrag/pipeline.py	2427	CODE
LOW	lightrag/pipeline.py	2833	CODE
LOW	lightrag/pipeline.py	3842	CODE
LOW	lightrag/pipeline.py	4173	CODE
LOW	lightrag/pipeline.py	4569	CODE
319 more matches not shown…

Modern AI Meta-Vocabulary81 hits · 251 pts

Severity	File	Line	Snippet	Context
MEDIUM	README-ja.md	7	# 🚀 LightRAG: シンプルかつ高速な検索拡張生成（RAG）	COMMENT
MEDIUM	README-ja.md	259	### 軽量なグラフベース RAG フレームワーク	COMMENT
MEDIUM	README.md	259	### A Lightweight, Graph-Based RAG Framework	COMMENT
MEDIUM	README-zh.md	7	# 🚀 LightRAG: 简单且快速的检索增强生成（RAG）框架	COMMENT
MEDIUM	reproduce/Step_1_openai_compatible.py	77	# Initialize RAG instance	COMMENT
MEDIUM	reproduce/Step_1.py	42	# Initialize RAG instance	COMMENT
MEDIUM	lightrag/constants.py	41	# pushing an otherwise-valid chunk past the provider context window; over budget	COMMENT
MEDIUM	lightrag/constants.py	713	# over the embedding model's context limit after chunking. Independent from	COMMENT
MEDIUM	lightrag/operate.py	322	# embedding model's own context limit — not a real-world budget.	COMMENT
MEDIUM	lightrag/operate.py	4035	# an otherwise-valid chunk past the provider context window. When the	COMMENT
MEDIUM⚡	lightrag/pipeline.py	2195	# Pipeline orchestration	COMMENT
MEDIUM	lightrag/pipeline.py	173	# fully event-driven (``await get_document()``) and never polls.	COMMENT
MEDIUM	lightrag/pipeline.py	6799	# absolute legacy RAG-Anything/MinerU path, a ``..`` escape)	COMMENT
MEDIUM	lightrag/pipeline.py	7027	# it for their model's context window.	COMMENT
MEDIUM⚡	lightrag/tools/download_cache.py	158	"text-embedding-ada-002", # Legacy embedding model	CODE
MEDIUM⚡	lightrag/tools/download_cache.py	159	"text-embedding-3-small", # Small embedding model	CODE
MEDIUM⚡	lightrag/tools/download_cache.py	160	"text-embedding-3-large", # Large embedding model	CODE
MEDIUM	lightrag/tools/source_conflict_repair.py	751	# embedding model — full_docs is a plain KV, and reading it is what lets the	COMMENT
MEDIUM⚡	lightrag/tools/migrate_graph_storage.py	569	# Async orchestration — preconditions, write pipeline, compensation	COMMENT
MEDIUM	lightrag/llm/gemini.py	562	# model that burned the whole output budget on its reasoning trace is	COMMENT
MEDIUM	lightrag/llm/gemini.py	622	# A thinking model that spent its whole budget on the reasoning trace	COMMENT
MEDIUM	lightrag/llm/binding_options.py	632	# Whether the model may emit a hidden reasoning trace before its answer.	COMMENT
MEDIUM	lightrag/llm/ollama.py	325	# the reasoning trace). Raise like the OpenAI binding does,	COMMENT
MEDIUM	lightrag/parser/docx/parse_document.py	667	# (guardrails.is_visually_centered).	COMMENT
MEDIUM	lightrag/parser/docx/parse_document.py	1168	# --- smart path (pipeline + landing guardrails) ---------------	COMMENT
MEDIUM	lightrag/parser/docx/parse_document.py	1188	# Landing guardrails: any violation abandons the smart output for THIS	COMMENT
MEDIUM	lightrag/parser/docx/parse_document.py	1242	# three counts come from guardrails.plan_toc_output — the single visibility	COMMENT
MEDIUM	lightrag/parser/docx/smart_heading/features.py	341	#: is not visually centered (``guardrails.is_visually_centered``).	COMMENT
MEDIUM	lightrag/parser/docx/smart_heading/features.py	431	#: unit :func:`leading_pad_em` reports and ``guardrails.CENTER_MAX_LEADING_PAD_EM``	COMMENT
MEDIUM⚡	lightrag/parser/docx/smart_heading/heading_flow.py	2710	# whole-document orchestration	COMMENT
MEDIUM	lightrag/parser/docx/smart_heading/heading_flow.py	1842	#: - _Medical Graph RAG.docx's author line (96) absorbing its affiliation line	COMMENT
MEDIUM⚡	lightrag/parser/docx/smart_heading/guardrails.py	560	# Landing guardrails: canonicalization, I1/I2/I3 machine checks, TOC	COMMENT
MEDIUM	lightrag/parser/docx/smart_heading/guardrails.py	246	# dependency parser hallucinates sentence boundaries mid-word on short	COMMENT
MEDIUM	lightrag/parser/docx/smart_heading/title_block.py	1046	# candidate reuses single=True judge semantics (± context window; a	COMMENT
MEDIUM	lightrag/kg/mongo_impl.py	3998	# built under a different embedding model must NOT be	COMMENT
MEDIUM⚡	lightrag/kg/postgres_impl.py	2313	# with proper embedding model and dimension suffix for data isolation	STRING
MEDIUM	lightrag/kg/faiss_impl.py	1202	# the `is pdoc` identity check is defensive scaffolding: today the	COMMENT
MEDIUM	lightrag/kg/scan_job_store.py	227	# Timestamps, version, counters_dropped and the JSON scaffolding.	COMMENT
MEDIUM	lightrag/api/config.py	391	# RAG configuration	COMMENT
MEDIUM	lightrag/api/utils_api.py	690	# RAG Configuration	STRING
MEDIUM	lightrag/api/lightrag_server.py	714	# embedding model is configured. Without this, the provider's decorator	COMMENT
MEDIUM	lightrag/api/lightrag_server.py	2332	# Initialize RAG with unified configuration	COMMENT
MEDIUM	lightrag/api/lightrag_server.py	2753	# embedding model configuration binding/host address (if applicable)/model (if applicable)	COMMENT
MEDIUM	lightrag/evaluation/eval_rag_quality.py	19	# Specify custom RAG endpoint	STRING
MEDIUM⚡	lightrag/evaluation/eval_rag_quality.py	419	# all RAG responses from being generated at once when eval is slow	COMMENT
MEDIUM⚡	lightrag/evaluation/eval_rag_quality.py	424	# Stage 1: Generate RAG response	COMMENT
MEDIUM	lightrag/evaluation/eval_rag_quality.py	572	# Stage 1: RAG generation - allow x2 concurrency to keep evaluation fed	COMMENT
MEDIUM	lightrag/evaluation/eval_rag_quality.py	598	max_connections=(max_async + 1) * 2, # Allow buffer for RAG stage	CODE
MEDIUM	lightrag/evaluation/eval_rag_quality.py	976	# Specify custom RAG endpoint	COMMENT
MEDIUM	…ation/sample_documents/05_evaluation_and_deployment.md	1	# RAG Evaluation Metrics and Deployment	COMMENT
MEDIUM	…ation/sample_documents/05_evaluation_and_deployment.md	3	## Key RAG Evaluation Metrics	COMMENT
MEDIUM	…rag/evaluation/sample_documents/02_rag_architecture.md	1	# RAG System Architecture	COMMENT
MEDIUM	…rag/evaluation/sample_documents/02_rag_architecture.md	3	## Main Components of RAG Systems	COMMENT
MEDIUM	…valuation/sample_documents/03_lightrag_improvements.md	1	# LightRAG Improvements Over Traditional RAG	COMMENT
MEDIUM	tests/test_doc_processing_status_from_stored.py	34	# The RAG-Anything shape from HKUDS/RAG-Anything#73	COMMENT
MEDIUM⚡	tests/tools/test_migrate_graph_storage.py	434	# Async orchestration — fakes standing in for the storage backends	COMMENT
MEDIUM⚡	tests/workspace/test_workspace_isolation.py	547	# Purpose: Validate guardrails for workspace normalization and namespace	COMMENT
MEDIUM	tests/workspace/test_workspace_isolation.py	1076	# Test 11.2: Insert different data to each RAG instance (CONCURRENTLY)	COMMENT
MEDIUM⚡	tests/parser/docx/test_smart_heading_flow.py	2892	# These tests use the REAL guardrails.strong_body_reason: every record either	COMMENT
MEDIUM⚡	tests/kg/qdrant_impl/test_qdrant_migration.py	214	# Use a large embedding model	COMMENT
21 more matches not shown…

Docstring Block Structure50 hits · 250 pts

Severity	File	Line	Snippet	Context
HIGH	lightrag/lightrag.py	297	Drive an async coroutine to completion from a synchronous wrapper. The synchronous wrappers (``insert``, ``query``,	STRING
HIGH	lightrag/utils.py	2256	Execute function with enhanced priority-based concurrency control and timeout handling Args:	STRING
HIGH	lightrag/utils.py	7014	Validate a workspace name used to build per-workspace directories. File-based storages place their data in a subdir	STRING
HIGH	lightrag/utils.py	7315	Reject a node/edge attribute mapping no graph backend could store. Args: attributes: Attribute mapping abou	STRING
HIGH	lightrag/utils.py	7337	Reject attribute names a backend would read as something other than a name. The interpretation rule, not the seri	STRING
HIGH	lightrag/utils.py	7387	Reject attribute names or values XML cannot encode. For a GraphML-backed store. Names are checked as well as values	STRING
HIGH	lightrag/utils.py	7414	Reject attribute values no graph backend can store. See ``graph_attribute_value_rejection`` for the rules and `	STRING
HIGH	lightrag/pipeline.py	677	Pipeline for Processing Documents 1. Validate ids if provided or generate MD5 hash IDs and remove dupl	STRING
HIGH	lightrag/utils_graph.py	61	Return data reduced to well-typed, storable graph attributes. Two call shapes, because the two families of caller	STRING
HIGH	lightrag/tools/clean_llm_query_cache.py	252	Initialize storage instance with fallback to config.ini and defaults Args: storage_name: Storage im	STRING
HIGH	lightrag/tools/migrate_llm_cache.py	289	Initialize storage instance with fallback to config.ini and defaults Args: storage_name: Storage im	STRING
HIGH	lightrag/llm/jina.py	85	Generate embeddings for a list of texts using Jina AI's API. Args: texts: List of texts to embed. m	STRING
HIGH	lightrag/llm/voyageai.py	64	Generate embeddings for a list of texts using VoyageAI's API. Args: texts: List of texts to embed.	STRING
HIGH	lightrag/llm/gemini.py	307	Complete a prompt using Gemini's API with Chain of Thought (COT) support. This function supports automatic int	STRING
HIGH	lightrag/llm/gemini.py	716	Generate embeddings for a list of texts using Gemini's API. This function uses Google's Gemini embedding model to g	STRING
HIGH	lightrag/llm/openai.py	323	Complete a prompt using OpenAI's API with caching support and Chain of Thought (COT) integration. This function sup	STRING
HIGH	lightrag/llm/openai.py	1070	Generate embeddings for a list of texts using OpenAI's API with automatic text truncation. This function supports b	STRING
HIGH	lightrag/parser/docx/utils.py	96	Create Gemini client for AI Studio or Vertex AI. Supports two modes: - AI Studio (default): Uses GOOGLE_AP	STRING
HIGH	lightrag/parser/docx/utils.py	173	Create OpenAI client with optional custom base URL. Environment variables: - OPENAI_API_KEY: Required API	STRING
HIGH	lightrag/parser/docx/table_extractor.py	250	Extract table to 2D string array with metadata (paraIds, header info). Vertical merge behavior:	STRING
HIGH	lightrag/kg/postgres_impl.py	789	Execute a database operation with automatic retry for transient failures. Args: operation:	STRING
HIGH	lightrag/kg/memgraph_impl.py	135	Check if a node exists in the graph. Args: node_id: The ID of the node to check.	STRING
HIGH	lightrag/kg/memgraph_impl.py	175	Check if an edge exists between two nodes in the graph. Args: source_node_id: The ID of th	STRING
HIGH	lightrag/kg/memgraph_impl.py	223	Get node by its label identifier, return only node properties Args: node_id: The node label to look	STRING
HIGH	lightrag/kg/memgraph_impl.py	278	Get the degree (number of relationships) of a node with the given label. If multiple nodes have the same label,	STRING
HIGH	lightrag/kg/memgraph_impl.py	365	Retrieves all edges (relationships) for a particular node identified by its label. Args: source_nod	STRING
HIGH	lightrag/kg/memgraph_impl.py	440	Get edge properties between two nodes. Args: source_node_id: Label of the source node t	STRING
HIGH	lightrag/kg/neo4j_impl.py	476	Check if a node with the given label exists in the database Args: node_id: Label of the no	STRING
HIGH	lightrag/kg/neo4j_impl.py	510	Check if an edge exists between two nodes Args: source_node_id: Label of the source node	STRING
HIGH	lightrag/kg/neo4j_impl.py	552	Get node by its label identifier, return only node properties Args: node_id: The node label to look	STRING
HIGH	lightrag/kg/neo4j_impl.py	643	Get the degree (number of relationships) of a node with the given label. If multiple nodes have the same label,	STRING
HIGH	lightrag/kg/neo4j_impl.py	780	Get edge properties between two nodes. Args: source_node_id: Label of the source node t	STRING
HIGH	lightrag/kg/neo4j_impl.py	914	Retrieves all edges (relationships) for a particular node identified by its label. Args: source_nod	STRING
HIGH	lightrag/api/auth.py	183	Validate JWT token Args: token: JWT token Returns: dict: Dictionary c	STRING
HIGH	lightrag/api/routers/graph_routes.py	786	Delete an entity and all its relationships from the knowledge graph. Args: request (Delete	STRING
HIGH	lightrag/api/routers/graph_routes.py	822	Delete a relationship between two entities from the knowledge graph. Args: request (Delete	STRING
HIGH	lightrag/api/routers/query_routes.py	448	Comprehensive RAG query endpoint with non-streaming response. Parameter "stream" is ignored. **Query M	STRING
HIGH	lightrag/api/routers/query_routes.py	737	Advanced RAG query endpoint with flexible streaming response. This endpoint provides the most flexible	STRING
HIGH	lightrag/api/routers/query_routes.py	1310	Advanced data retrieval endpoint for structured RAG analysis. This endpoint provides raw retrieval res	STRING
HIGH	lightrag/api/routers/document_routes.py	265	Validate an uploaded filename and return it unchanged when safe. Uploaded filenames are document identifiers a	STRING
HIGH	lightrag/api/routers/document_routes.py	5200	Upload a file to the input directory and index it. This API endpoint accepts a file through an HTTP PO	STRING
HIGH	lightrag/api/routers/document_routes.py	5531	Insert text into the RAG system. This endpoint allows you to insert text data into the RAG system for	STRING
HIGH	lightrag/api/routers/document_routes.py	5663	Insert multiple texts into the RAG system. This endpoint allows you to insert multiple text entries in	STRING
HIGH	lightrag/api/routers/document_routes.py	6368	Delete documents and all their associated data by their IDs using background processing. Deletes speci	STRING
HIGH	lightrag/api/routers/document_routes.py	6491	Clear all cache data from the LLM response cache storage. This endpoint clears all cached LLM response	STRING
HIGH	lightrag/api/routers/document_routes.py	6525	Get the processing status of documents by tracking ID. This endpoint retrieves all documents associate	STRING
HIGH	lightrag/api/routers/document_routes.py	6601	Get documents with pagination support. This endpoint retrieves documents with pagination, filtering, a	STRING
HIGH	lightrag/evaluation/eval_rag_quality.py	295	Generate RAG response by calling LightRAG API. Args: question: The user query.	STRING
HIGH	lightrag/chunker/paragraph_semantic.py	2060	Paragraph Semantic Chunking — the ``chunking="P"`` strategy. Reads structured blocks from a ``.blocks.jsonl`` sidec	STRING
HIGH	tests/api/test_lightrag_ollama_chat.py	147	Send an HTTP request with retry mechanism Args: url: Request URL data: Request data stream:	STRING

Magic Placeholder Names37 hits · 195 pts

Severity	File	Line	Snippet	Context
HIGH	tests/setup/test_misc.py	297	"LLM_BINDING_API_KEY=your_api_key",	STRING
HIGH	tests/setup/test_misc.py	302	"EMBEDDING_BINDING_API_KEY=your_api_key",	STRING
HIGH	docs/LightRAG-API-Server.md	149	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/LightRAG-API-Server.md	155	# EMBEDDING_BINDING_API_KEY=your_api_key	COMMENT
HIGH	docs/LightRAG-API-Server.md	166	# LLM_BINDING_API_KEY=your_api_key	COMMENT
HIGH	docs/LightRAG-API-Server.md	174	# EMBEDDING_BINDING_API_KEY=your_api_key	COMMENT
HIGH	docs/LightRAG-API-Server.md	298	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/LightRAG-API-Server.md	309	EMBEDDING_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/LightRAG-API-Server.md	815	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/LightRAG-API-Server.md	1114	LLM_BINDING_API_KEY=your-api-key	CODE
HIGH	docs/ParserServiceDeployment.md	48	"api_key": "your_api_key",	CODE
HIGH	docs/ProgramingWithCore.md	978	export OPENAI_API_KEY=your-api-key	CODE
HIGH	docs/ProgramingWithCore.md	992	OPENAI_API_KEY=your-api-key \	CODE
HIGH	docs/ParserServiceDeployment-zh.md	48	"api_key": "your_api_key",	CODE
HIGH	docs/RoleSpecificLLMConfiguration-zh.md	26	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/RoleSpecificLLMConfiguration-zh.md	126	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/RoleSpecificLLMConfiguration-zh.md	207	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/RoleSpecificLLMConfiguration-zh.md	224	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/RoleSpecificLLMConfiguration-zh.md	337	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/LightRAG-API-Server-zh.md	149	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/LightRAG-API-Server-zh.md	155	# EMBEDDING_BINDING_API_KEY=your_api_key	COMMENT
HIGH	docs/LightRAG-API-Server-zh.md	166	# LLM_BINDING_API_KEY=your_api_key	COMMENT
HIGH	docs/LightRAG-API-Server-zh.md	174	# EMBEDDING_BINDING_API_KEY=your_api_key	COMMENT
HIGH	docs/LightRAG-API-Server-zh.md	298	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/LightRAG-API-Server-zh.md	309	EMBEDDING_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/LightRAG-API-Server-zh.md	815	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/LightRAG-API-Server-zh.md	1114	LLM_BINDING_API_KEY=your-api-key	CODE
HIGH	docs/RoleSpecificLLMConfiguration.md	26	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/RoleSpecificLLMConfiguration.md	126	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/RoleSpecificLLMConfiguration.md	207	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/RoleSpecificLLMConfiguration.md	224	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	docs/RoleSpecificLLMConfiguration.md	337	LLM_BINDING_API_KEY=your_api_key	CODE
HIGH	examples/lightrag_gemini_postgres_demo.py	28	GEMINI_API_KEY=your-api-key	STRING
HIGH⚡	examples/lightrag_gemini_demo.py	37	"Please set it with: export GEMINI_API_KEY='your-api-key'"	CODE
HIGH	examples/lightrag_openai_opensearch_graph_demo.py	23	OPENAI_API_KEY=your-api-key	STRING
HIGH	…s/unofficial-sample/lightrag_llamaindex_direct_demo.py	29	OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY", "your-api-key-here")	CODE
HIGH⚡	examples/unofficial-sample/lightrag_cloudflare_demo.py	20	cloudflare_api_key = "YOUR_API_KEY"	CODE

Over-Commented Block286 hits · 192 pts

Severity	File	Line	Snippet	Context
LOW	docker-compose.podman.yml	1	# Podman-compatible compose file for LightRAG	COMMENT
LOW	docker-entrypoint.sh	1	#!/bin/sh	COMMENT
LOW	docker-entrypoint.sh	21	# process can read/write them, covering bind-mounts/PVCs whose host content	COMMENT
LOW	requirements-offline-smart-heading.txt	1	# LightRAG Offline Dependencies - Native docx smart_heading (optional)	COMMENT
LOW	lightrag/lightrag.py	1101	# paragraphs together. ``default_chunker_config`` already	COMMENT
LOW	lightrag/lightrag.py	1181	self._ensure_addon_params_cache()	COMMENT
LOW	lightrag/lightrag.py	1941	"pipeline_status", workspace=self.workspace	COMMENT
LOW	lightrag/lightrag.py	2101	prior_truncation = existing_journal["prior_llm_truncation"]	COMMENT
LOW	lightrag/lightrag.py	2221	_operation_truncation_record(),	COMMENT
LOW	lightrag/lightrag.py	2381	# write PROCESSED (the commit record) last with the	COMMENT
LOW	lightrag/lightrag.py	2461	# slot nor wrote any data, so we must NOT flush/discard the SHARED	COMMENT
LOW	lightrag/lightrag.py	2541	try:	COMMENT
LOW	lightrag/lightrag.py	2841	# has_custom_chunk_journal flag (no per-row hydration).	COMMENT
LOW	lightrag/lightrag.py	3701	namespace = f"{workspace}:GraphDB" if workspace else "GraphDB"	COMMENT
LOW	lightrag/lightrag.py	4801	# purge from destroying the chunk rows and anchor rows that are the	COMMENT
LOW	lightrag/lightrag.py	5921	)	COMMENT
LOW	lightrag/lightrag.py	6101	# Delegated to the shared primitive rather than reimplemented here	COMMENT
LOW	lightrag/lightrag.py	6281	finally:	COMMENT
LOW	lightrag/constants.py	121	# Bounds on any separator cascade, wherever it comes from. The recursive splitter	COMMENT
LOW	lightrag/constants.py	141	# CHUNK_SIZE default — heading-aligned paragraph merging needs more	COMMENT
LOW	lightrag/constants.py	161	# the ingestion path bounds: MAX_UPLOAD_SIZE bounds the compressed artifact on	COMMENT
LOW	lightrag/constants.py	181	# class — so the DOCX_* knobs below bound all three, enforced in the legacy	COMMENT
LOW	lightrag/constants.py	201	# above, these cap the bytes materialized into ``*.blocks.assets``: one image	COMMENT
LOW	lightrag/constants.py	221	# Overall wall-clock budget for downloading the result bundle from docling	COMMENT
LOW	lightrag/constants.py	241	# Native docx smart_heading (opt-in engine param) tunables. Each DEFAULT_*	COMMENT
LOW	lightrag/constants.py	261	"年,月,日,时,分,秒,个,人,只,条,次,项,天,号,元,角,件,名,台,种,倍,"	COMMENT
LOW	lightrag/constants.py	281	# anchor, which is why there is no longer an IMPRINT_SPACE_PREFIXES knob);	COMMENT
LOW	lightrag/constants.py	301	# rich physical outline naturally admits more candidates, so the effective	COMMENT
LOW	lightrag/constants.py	321	# CB4 per-sub-document gate: once the whole document clears the gate above,	COMMENT
LOW	lightrag/constants.py	341	# paragraphs (0 disables the auxiliary body-run break; the primary close	COMMENT
LOW	lightrag/constants.py	361	# Marker prefix for full_docs.content when format=lightrag.	COMMENT
LOW	lightrag/constants.py	381	# Reserved doc_status.metadata key recording how far this document's KG write	COMMENT
LOW	lightrag/constants.py	421	)	COMMENT
LOW	lightrag/constants.py	521	# Whether a doc_status backend missing a strict capability is a startup failure	COMMENT
LOW	lightrag/constants.py	541	# retries and ``/documents/scan`` bulk enqueues deliberately break through the	COMMENT
LOW	lightrag/constants.py	561	# Enabled by default. It used to be ``0`` and to cover three ingestion routes	COMMENT
LOW	lightrag/constants.py	581	# Input ceilings for the model-facing request fields. Deliberately NOT env knobs:	COMMENT
LOW	lightrag/constants.py	621	# limit. Deliberately fixed process-wide constants rather than anything derived	COMMENT
LOW	lightrag/constants.py	641	# exclusive reset acknowledges it — so an operator hammering /reprocess_failed	COMMENT
LOW	lightrag/constants.py	661	# characters because that is what is being bounded (CJK is 3 bytes/char).	COMMENT
LOW	lightrag/constants.py	681	# DEFAULT_PROCESSING_PRIORITY so analysis and ingestion work share the EXTRACT	COMMENT
LOW	lightrag/constants.py	721	# Default llm and embedding timeout	COMMENT
LOW	lightrag/constants.py	741	# workers); other waiters back off exponentially up to the DEFERRED cap.	COMMENT
LOW	lightrag/constants.py	781	ROLLBACK_REPORT_SAMPLE_CAP = 32	COMMENT
LOW	lightrag/operate.py	2141	if all_keywords	COMMENT
LOW	lightrag/operate.py	2821	if relation_chunks_storage is not None:	COMMENT
LOW	lightrag/operate.py	2921	# of distinct contributing sources. On reprocess/resume the same source	COMMENT
LOW	lightrag/operate.py	4021	chunk_key = chunk_key_dp[0]	COMMENT
LOW	lightrag/operate.py	4401	# every run of the same document. That permutation reaches persisted state	COMMENT
LOW	lightrag/utils.py	2701	#	COMMENT
LOW	lightrag/utils.py	2961	# ---------------------------------------------------------------------------	COMMENT
LOW	lightrag/utils.py	3021	# the event loop: ``write_json`` for the JSON KV / doc-status stores,	COMMENT
LOW	lightrag/utils.py	3241	# reliable token-to-character mapping; see ``TiktokenTokenizer`` below for	COMMENT
LOW	lightrag/utils.py	3461		COMMENT
LOW	lightrag/utils.py	3661		COMMENT
LOW	lightrag/utils.py	5481	if not text:	COMMENT
LOW	lightrag/utils.py	6721		COMMENT
LOW	lightrag/utils.py	7061		COMMENT
LOW	lightrag/utils.py	7221	return None	COMMENT
LOW	lightrag/pipeline.py	161	# §7.3) — never as a side effect of an unrelated upload or rescan.	COMMENT
226 more matches not shown…

Redundant / Tautological Comments140 hits · 191 pts

Severity	File	Line	Snippet	Context
LOW	docker-build-push.sh	38	# Check if buildx builder exists, create if not	COMMENT
LOW	lightrag/lightrag.py	4285	# Check if query_result is None	COMMENT
LOW	lightrag/operate.py	3042	# Check if this is a placeholder record	COMMENT
LOW	lightrag/operate.py	3952	# Check if JSON structured output mode is enabled	COMMENT
LOW	lightrag/operate.py	4395	# Check if any task raised an exception and ensure all exceptions are retrieved.	COMMENT
LOW	lightrag/operate.py	4762	# Check if pre-defined keywords are already provided	COMMENT
LOW	lightrag/operate.py	6349	# Check if any relations still have chunks after deduplication	COMMENT
LOW	lightrag/utils.py	305	# Check if the resolved path is within the base directory	COMMENT
LOW	lightrag/utils.py	474	# Check if record has the required attributes for an access log	COMMENT
LOW	lightrag/utils.py	651	# Check if func is already an EmbeddingFunc instance and unwrap it	COMMENT
LOW	lightrag/utils.py	674	# Check if user provided embedding_dim parameter	COMMENT
LOW	lightrag/utils.py	698	# Check if underlying function supports max_token_size and inject if not provided	COMMENT
LOW	lightrag/utils.py	711	# Check if total elements can be evenly divided by embedding_dim	COMMENT
LOW	lightrag/utils.py	1398	# Check if task was cancelled before worker started	COMMENT
LOW	lightrag/utils.py	4171	# Check if we already have identical content cached	COMMENT
LOW	lightrag/utils.py	5846	# Check if there are still unused chunks	COMMENT
LOW	lightrag/utils.py	6117	# Check if results are in the new index-based format	COMMENT
LOW	lightrag/pipeline.py	2895	# Check if corresponding content exists in full_docs. Strict where	COMMENT
LOW	lightrag/pipeline.py	2904	# Check if this is a failed document that should be preserved	COMMENT
LOW	lightrag/storage_migrations.py	34	# Check if migration is needed:	COMMENT
LOW	lightrag/storage_migrations.py	54	# Check if full_entities and full_relations are empty	COMMENT
LOW	lightrag/utils_graph.py	343	# Check if the entity exists	COMMENT
LOW	lightrag/utils_graph.py	444	# Check if the relation exists	COMMENT
LOW	lightrag/utils_graph.py	1266	# Check if storage has existing data	COMMENT
LOW	lightrag/utils_graph.py	1419	# Check if the normalized entity already exists.	COMMENT
LOW	lightrag/utils_graph.py	1564	# Check if both entities exist	COMMENT
LOW	lightrag/utils_graph.py	1573	# Check if relation already exists	COMMENT
LOW	lightrag/base.py	261	# Check if model_name exists (model_name is optional in EmbeddingFunc)	COMMENT
LOW	lightrag/tools/clean_llm_query_cache.py	207	# Check if config.ini has configuration	COMMENT
LOW	lightrag/tools/clean_llm_query_cache.py	971	# Check if choice is valid	COMMENT
LOW	lightrag/tools/clean_llm_query_cache.py	1069	# Check if user cancelled	COMMENT
LOW	lightrag/tools/clean_llm_query_cache.py	1131	# Check if there are any records to delete	COMMENT
LOW	lightrag/tools/download_cache.py	131	# Check if TIKTOKEN_CACHE_DIR is already set in environment	COMMENT
LOW	lightrag/tools/check_initialization.py	101	# Print results	COMMENT
LOW⚡	lightrag/tools/migrate_llm_cache.py	237	# Check if storage requires configuration	COMMENT
LOW⚡	lightrag/tools/migrate_llm_cache.py	244	# Check if has environment variables	COMMENT
LOW⚡	lightrag/tools/migrate_llm_cache.py	249	# Check if has config.ini configuration	COMMENT
LOW	lightrag/tools/migrate_llm_cache.py	216	# Check if config.ini has configuration	COMMENT
LOW	lightrag/tools/migrate_llm_cache.py	1068	# Check if choice is valid	COMMENT
LOW	lightrag/tools/migrate_llm_cache.py	1438	# Check if user cancelled (setup_storage returns None for all fields)	COMMENT
LOW	lightrag/tools/migrate_llm_cache.py	1442	# Check if there are at least 2 storage types available	COMMENT
LOW	lightrag/llm/jina.py	28	# Check if the error response is HTML (common for 502, 503, etc.)	COMMENT
LOW	lightrag/llm/gemini.py	262	# Check if this part is thought content using the 'thought' attribute	COMMENT
LOW	lightrag/llm/binding_options.py	296	# Check if this is a dataclass and use dataclass fields	COMMENT
LOW⚡	lightrag/llm/openai.py	589	# Check if this chunk has usage information (final chunk)	COMMENT
LOW⚡	lightrag/llm/openai.py	596	# Check if choices exists and is not empty	COMMENT
LOW⚡	lightrag/llm/openai.py	604	# Check if delta exists	COMMENT
LOW	lightrag/llm/openai.py	53	# Check if required Langfuse environment variables are set	COMMENT
LOW	lightrag/llm/openai.py	820	# Check if we should include reasoning content	COMMENT
LOW	lightrag/parser/docx/utils.py	373	# Check if it mentions billing which indicates permanent quota issue	COMMENT
LOW	lightrag/kg/qdrant_impl.py	329	# Check if legacy collection has workspace_id to determine migration strategy	COMMENT
LOW	lightrag/kg/qdrant_impl.py	440	# Check if we've reached the end	COMMENT
LOW	lightrag/kg/networkx_impl.py	349	# Check if data needs to be reloaded	COMMENT
LOW	lightrag/kg/networkx_impl.py	734	# Check if graph is truncated	COMMENT
LOW	lightrag/kg/networkx_impl.py	745	# Check if node exists	COMMENT
LOW⚡	lightrag/kg/networkx_impl.py	796	# Add neighbors to the queue with their degrees	COMMENT
LOW⚡	lightrag/kg/networkx_impl.py	801	# Check if there are unexplored neighbors (skipped due to depth limit)	COMMENT
LOW⚡	lightrag/kg/networkx_impl.py	809	# Check if we've reached max_nodes	COMMENT
LOW⚡	lightrag/kg/networkx_impl.py	818	# Check if graph is truncated - either due to max_nodes limit or depth limit	COMMENT
LOW	lightrag/kg/networkx_impl.py	947	# Check if storage was updated by another process	COMMENT
80 more matches not shown…

Self-Referential Comments46 hits · 138 pts

Severity	File	Line	Snippet	Context
MEDIUM	reproduce/Step_1_openai_compatible.py	77	# Initialize RAG instance	COMMENT
MEDIUM	reproduce/Step_1.py	42	# Initialize RAG instance	COMMENT
MEDIUM⚡	lightrag/lightrag.py	1374	# Create a NEW EmbeddingFunc instance with the wrapped func to avoid mutating the caller's object	COMMENT
MEDIUM	lightrag/lightrag.py	4091	# Create a copy of param to avoid modifying the original	COMMENT
MEDIUM	lightrag/operate.py	6564	# Create a preliminary system prompt with empty content_data to calculate overhead	COMMENT
MEDIUM	lightrag/utils.py	6900	# Create a list of (file_path, count, first_index) tuples	COMMENT
MEDIUM	lightrag/pipeline.py	1351	# Create a new record with unique ID for this duplicate attempt	COMMENT
MEDIUM	lightrag/tools/migrate_llm_cache.py	716	# Create a snapshot of matching items while holding the lock	COMMENT
MEDIUM	lightrag/llm/openai.py	229	# Create a merged config dict with precedence: explicit params > client_configs	COMMENT
MEDIUM	lightrag/llm/openai.py	263	# Create a merged config dict with precedence: explicit params > client_configs > defaults	COMMENT
MEDIUM	lightrag/llm/openai.py	443	# Create the OpenAI client (supports both OpenAI and Azure)	COMMENT
MEDIUM	lightrag/llm/openai.py	1166	# Create the OpenAI client (supports both OpenAI and Azure)	COMMENT
MEDIUM	lightrag/llm/bedrock.py	347	# Create a session that will be used throughout the streaming process	COMMENT
MEDIUM	lightrag/llm/bedrock.py	358	# Define the generator function that will manage the client lifecycle	COMMENT
MEDIUM⚡	lightrag/kg/mongo_impl.py	3747	# Create the improved search index (async, no waiting)	COMMENT
MEDIUM	lightrag/kg/mongo_impl.py	489	# Create a copy of v for $set operation, excluding create_time to avoid conflicts	COMMENT
MEDIUM	lightrag/kg/mongo_impl.py	4149	# Define the aggregation pipeline with the converted query vector	COMMENT
MEDIUM	lightrag/kg/postgres_impl.py	2170	# Define the field changes needed	COMMENT
MEDIUM	lightrag/kg/faiss_impl.py	269	# Create an empty Faiss index for inner product (useful for normalized vectors = cosine similarity).	COMMENT
MEDIUM	lightrag/kg/shared_storage.py	62	# Define a direct print function for critical logs that must be visible in all processes	COMMENT
MEDIUM	lightrag/kg/shared_storage.py	1671	# Create a shared list object for history_messages	COMMENT
MEDIUM	lightrag/kg/shared_storage.py	3554	# Create a simple mutable object to store boolean value for compatibility with mutiprocess	COMMENT
MEDIUM	lightrag/api/run_with_gunicorn.py	147	# Define a custom application class that loads our config	COMMENT
MEDIUM	lightrag/api/run_with_gunicorn.py	277	# Create the application	COMMENT
MEDIUM	lightrag/api/lightrag_server.py	1024	# Create the EmbeddingFunc instance (now returns complete EmbeddingFunc with max_token_size)	COMMENT
MEDIUM	lightrag/api/routers/ollama_api.py	210	# Create an instance of the model	COMMENT
MEDIUM	lightrag/api/routers/query_routes.py	547	# Create a mapping from reference_id to chunk content	COMMENT
MEDIUM	lightrag/api/routers/document_routes.py	4698	# Create the bounded job record BEFORE publishing the manual intent	COMMENT
MEDIUM	tests/kg/test_graph_storage.py	152	# Initialize the storage instance	COMMENT
MEDIUM	tests/kg/postgres_impl/test_postgres_index_name.py	84	# Create a table name that results in exactly 63 bytes	COMMENT
MEDIUM	tests/chunker/test_rerank_chunking.py	90	# Create a very long document that exceeds character limit	COMMENT
MEDIUM	examples/graph_visual_with_html.py	15	# Create a Pyvis network	COMMENT
MEDIUM	examples/lightrag_openai_compatible_demo.py	150	# Initialize RAG instance	COMMENT
MEDIUM	examples/lightrag_openai_mongodb_graph_demo.py	72	# Initialize RAG instance	COMMENT
MEDIUM	examples/lightrag_openai_demo.py	119	# Initialize RAG instance	COMMENT
MEDIUM	examples/lightrag_ollama_demo.py	139	# Initialize RAG instance	COMMENT
MEDIUM	examples/graph_visual_with_neo4j.py	159	# Create a Neo4j driver	STRING
MEDIUM	…mples/unofficial-sample/lightrag_embedding_prefixes.py	156	# Initialize RAG instance	COMMENT
MEDIUM	…/unofficial-sample/lightrag_llamaindex_litellm_demo.py	103	# Initialize RAG instance	COMMENT
MEDIUM	…cial-sample/lightrag_openai_neo4j_milvus_redis_demo.py	73	# Initialize RAG instance	COMMENT
MEDIUM	…s/unofficial-sample/lightrag_llamaindex_direct_demo.py	101	# Initialize RAG instance	COMMENT
MEDIUM	…ficial-sample/lightrag_llamaindex_litellm_opik_demo.py	114	# Initialize RAG instance	COMMENT
MEDIUM	examples/unofficial-sample/lightrag_cloudflare_demo.py	251	# Initialize RAG instance	COMMENT
MEDIUM	examples/unofficial-sample/lightrag_lmdeploy_demo.py	69	# Initialize RAG instance	COMMENT
MEDIUM	examples/unofficial-sample/lightrag_nvidia_demo.py	123	# Initialize RAG instance	COMMENT
MEDIUM	scripts/setup/apple-container.sh	272	# This file is a copy of the user's .env (real API keys) — keep it private.	COMMENT

Cross-Language Confusion21 hits · 96 pts

Severity	File	Line	Snippet	Context
HIGH	lightrag/utils.py	6507	``{"chunk_ids": null}``), which mean UNKNOWN. Only in the UNKNOWN case may a	STRING
HIGH	lightrag/utils.py	7277	``SET n += {k: null}`` silently deletes the property.	STRING
HIGH	lightrag/parser/docx/parser.py	149	# ``positions: [{"type": "paraid", "range": null}]``.	COMMENT
HIGH	lightrag/parser/docx/smart_heading/title_block.py	1129	- If false: set all six text fields to null, and classify EVERY index — {indices} — into exactly one of "headings" (a re	CODE
HIGH	lightrag/parser/docx/smart_heading/title_block.py	1132	{{"is_title_block": true\|false, "main_title": string\|null, "sub_title": string\|null, "doc_number": string\|null, "classif	CODE
HIGH	lightrag/kg/pgtable_impl.py	753	properties = lightrag_graph_nodes.properties \|\| EXCLUDED.properties,	CODE
HIGH	lightrag/kg/pgtable_impl.py	1513	properties = lightrag_graph_nodes.properties \|\| EXCLUDED.properties,	CODE
HIGH	lightrag/kg/mongo_impl.py	1351	# every doc with a real (non-null, existing) value.	COMMENT
HIGH	lightrag/kg/postgres_impl.py	7231	" $1::text \|\| E'\\x01' \|\|"	CODE
HIGH	lightrag/kg/postgres_impl.py	5936	CURRENT_TIMESTAMP) sort FIRST and encode as ``[null, id]`` so the	STRING
HIGH	lightrag/api/lightrag_server.py	139	if (!data \|\| data.type !== 'lightrag:set-docs-theme') return;	CODE
HIGH	lightrag/api/routers/graph_routes.py	374	"merge_error": null,	STRING
HIGH	lightrag/api/routers/graph_routes.py	376	"target_entity": null,	STRING
HIGH	lightrag/api/routers/graph_routes.py	402	"merge_error": null,	STRING
HIGH⚡	tests/kg/postgres_impl/test_pg_scheduling_pages.py	199	the cursor encodes [null, id] instead of failing closed — a row-value	STRING
HIGH	…sts/kg/postgres_impl/test_postgres_cypher_injection.py	202	"formula": "x < 5 && y > 3",	CODE
HIGH	…sts/kg/postgres_impl/test_postgres_cypher_injection.py	213	assert '`formula`: "x < 5 && y > 3"' in call["sql"]	CODE
HIGH	…s/kg/postgres_impl/test_postgres_upsert_edge_cypher.py	183	assert "$1::text \|\| E'\\x01' \|\|" in lock_sql	CODE
HIGH	tests/kg/neo4j_impl/test_search_labels_sanitize.py	34	("foo && bar", "foo bar"),	CODE
HIGH⚡	tests/kg/pgtable_impl/test_pgtable_graph_storage.py	350	assert "lightrag_graph_nodes.properties \|\| EXCLUDED.properties" in sql	CODE
HIGH⚡	tests/kg/pgtable_impl/test_pgtable_graph_storage.py	366	assert "lightrag_graph_nodes.properties \|\| EXCLUDED.properties" in sql	CODE

Fake / Example Data66 hits · 62 pts

Severity	File	Line	Snippet	Context
LOW	lightrag/api/routers/document_routes.py	1149	"metadata": {"author": "John Doe", "year": 2025},	CODE
LOW	lightrag/api/routers/document_routes.py	1213	"metadata": {"author": "John Doe"},	CODE
LOW	lightrag/api/routers/document_routes.py	1255	"metadata": {"author": "John Doe", "year": 2025},	CODE
LOW	lightrag/api/routers/document_routes.py	1373	"metadata": {"author": "John Doe", "year": 2025},	CODE
LOW	tests/setup/test_collect.py	860	prompt_secret_until_valid_with_default() {{ printf 'fake-key'; }}	CODE
LOW	tests/parser/docx/test_smart_heading_flow.py	3758	logs = [m for m in _log_messages(caplog) if "placeholder" in m]	CODE
LOW	tests/kg/json_impl/test_write_json_optimization.py	30	"name": "John Doe",	CODE
LOW⚡	tests/api/config/test_api_config_vlm_enable.py	76	monkeypatch.setenv("VLM_LLM_BINDING_API_KEY", "placeholder")	CODE
LOW⚡	tests/api/routes/test_description_api_validation.py	229	["entity", "Alice", "Acme Corp", "founded", "Alice founded Acme Corp."],	CODE
LOW	tests/api/routes/test_description_api_validation.py	241	assert relation["tgt_id"] == "Acme Corp"	CODE
LOW	tests/api/routes/test_description_api_validation.py	262	["entity", "Alice", "Acme Corp", "founded", " "],	CODE
LOW	tests/api/routes/test_description_api_validation.py	277	["edge", "Alice", "Acme Corp", "founded", "Alice founded Acme Corp."],	CODE
LOW	tests/extraction/test_relation_evidence_floor_repair.py	265	@pytest.mark.parametrize("placeholder", sorted(RELATION_NO_EVIDENCE_SOURCE_IDS))	CODE
LOW	tests/extraction/test_relation_evidence_floor_repair.py	327	@pytest.mark.parametrize("placeholder", sorted(RELATION_NO_EVIDENCE_SOURCE_IDS))	CODE
LOW⚡	tests/extraction/test_entity_extraction_stability.py	510	assert set(nodes) == {"Alice", "Acme Corp"}	CODE
LOW⚡	tests/extraction/test_entity_extraction_stability.py	511	assert ("Alice", "Acme Corp") in edges	CODE
LOW⚡	tests/extraction/test_entity_extraction_stability.py	702	assert set(entities.keys()) == {"Alice", "Acme Corp"}	CODE
LOW⚡	tests/extraction/test_entity_extraction_stability.py	706	assert relation_data["tgt_id"] == "Acme Corp"	CODE
LOW	tests/extraction/test_entity_extraction_stability.py	185	"name": "Acme Corp",	CODE
LOW	tests/extraction/test_entity_extraction_stability.py	193	"target": "Acme Corp",	CODE
LOW	tests/extraction/test_entity_extraction_stability.py	660	assert next(iter(relationships.keys())) == ("Alice", "Acme Corp")	CODE
LOW	tests/extraction/test_entity_extraction_stability.py	683	assert relation_data["tgt_id"] == "Acme Corp"	CODE
LOW	lightrag_webui/src/locales/zh.json	390	"placeholder": "页面内搜索节点...",	CODE
LOW	lightrag_webui/src/locales/zh.json	397	"placeholder": "搜索节点名称...",	CODE
LOW	lightrag_webui/src/locales/zh.json	436	"placeholder": "输入查询内容 (支持模式前缀: /<Query Mode>)",	CODE
LOW	lightrag_webui/src/locales/zh.json	493	"placeholder": "请输入 API Key",	CODE
LOW	lightrag_webui/src/locales/ja.json	390	"placeholder": "ページ内のノードを検索...",	CODE
LOW	lightrag_webui/src/locales/ja.json	397	"placeholder": "ノード名を検索...",	CODE
LOW	lightrag_webui/src/locales/ja.json	436	"placeholder": "クエリを入力（プレフィックス対応: /<クエリモード>）",	CODE
LOW	lightrag_webui/src/locales/ja.json	493	"placeholder": "APIキーを入力",	CODE
LOW	lightrag_webui/src/locales/de.json	390	"placeholder": "Knoten auf der Seite suchen...",	CODE
LOW	lightrag_webui/src/locales/de.json	397	"placeholder": "Knotenname suchen...",	CODE
LOW	lightrag_webui/src/locales/de.json	436	"placeholder": "Geben Sie Ihre Abfrage ein (Präfix unterstützt: /<Abfragemodus>)",	CODE
LOW	lightrag_webui/src/locales/de.json	493	"placeholder": "API-Schlüssel eingeben",	CODE
LOW	lightrag_webui/src/locales/ru.json	390	"placeholder": "Поиск узлов на странице...",	CODE
LOW	lightrag_webui/src/locales/ru.json	397	"placeholder": "Поиск имени узла...",	CODE
LOW	lightrag_webui/src/locales/ru.json	436	"placeholder": "Введите ваш запрос (Поддержка префикса: /<Режим запроса>)",	CODE
LOW	lightrag_webui/src/locales/ru.json	493	"placeholder": "Введите ваш API ключ",	CODE
LOW	lightrag_webui/src/locales/uk.json	390	"placeholder": "Шукати вузли на сторінці...",	CODE
LOW	lightrag_webui/src/locales/uk.json	397	"placeholder": "Шукати ім'я вузла...",	CODE
LOW	lightrag_webui/src/locales/uk.json	436	"placeholder": "Введіть ваш запит (Підтримка префіксу: /<Режим запиту>)",	CODE
LOW	lightrag_webui/src/locales/uk.json	493	"placeholder": "Введіть ваш API ключ",	CODE
LOW	lightrag_webui/src/locales/en.json	390	"placeholder": "Search nodes in page...",	CODE
LOW	lightrag_webui/src/locales/en.json	397	"placeholder": "Search node name...",	CODE
LOW	lightrag_webui/src/locales/en.json	436	"placeholder": "Enter your query (Support prefix: /<Query Mode>)",	CODE
LOW	lightrag_webui/src/locales/en.json	493	"placeholder": "Enter your API key",	CODE
LOW	lightrag_webui/src/locales/fr.json	390	"placeholder": "Rechercher des nœuds dans la page...",	CODE
LOW	lightrag_webui/src/locales/fr.json	397	"placeholder": "Rechercher le nom du nœud...",	CODE
LOW	lightrag_webui/src/locales/fr.json	436	"placeholder": "Tapez votre requête (Préfixe de requête : /<Query Mode>)",	CODE
LOW	lightrag_webui/src/locales/fr.json	493	"placeholder": "Entrez votre clé API",	CODE
LOW	lightrag_webui/src/locales/ko.json	390	"placeholder": "페이지 내 노드 검색...",	CODE
LOW	lightrag_webui/src/locales/ko.json	397	"placeholder": "노드 이름 검색...",	CODE
LOW	lightrag_webui/src/locales/ko.json	436	"placeholder": "질문을 입력하세요 (접두사 지원: /<쿼리 모드>)",	CODE
LOW	lightrag_webui/src/locales/ko.json	493	"placeholder": "API 키 입력",	CODE
LOW	lightrag_webui/src/locales/vi.json	390	"placeholder": "Tìm kiếm nút trong trang...",	CODE
LOW	lightrag_webui/src/locales/vi.json	397	"placeholder": "Tìm kiếm tên nút...",	CODE
LOW	lightrag_webui/src/locales/vi.json	436	"placeholder": "Nhập truy vấn của bạn (Hỗ trợ tiền tố: /<Chế Độ Truy Vấn>)",	CODE
LOW	lightrag_webui/src/locales/vi.json	493	"placeholder": "Nhập khóa API của bạn",	CODE
LOW	lightrag_webui/src/locales/zh_TW.json	390	"placeholder": "頁面內搜尋節點...",	CODE
LOW	lightrag_webui/src/locales/zh_TW.json	397	"placeholder": "搜尋節點名稱...",	CODE
6 more matches not shown…

Structural Annotation Overuse31 hits · 48 pts

Severity	File	Line	Snippet	Context
LOW⚡	lightrag/lightrag.py	1373	# Step 2: Apply priority wrapper decorator to EmbeddingFunc's inner func	COMMENT
LOW	lightrag/lightrag.py	1355	# Step 1: Capture embedding_func and max_token_size before applying rate_limit decorator	COMMENT
LOW⚡	lightrag/operate.py	6138	# Step 5: Batch retrieve chunk data	COMMENT
LOW⚡	lightrag/operate.py	6144	# Step 6: Build result chunks with valid data and update chunk tracking	COMMENT
LOW⚡	lightrag/operate.py	6433	# Step 5: Batch retrieve chunk data	COMMENT
LOW⚡	lightrag/operate.py	6439	# Step 6: Build result chunks with valid data and update chunk tracking	COMMENT
LOW	lightrag/operate.py	6027	# Step 1: Collect all text chunks for each entity	COMMENT
LOW	lightrag/operate.py	6054	# Step 2: Count chunk occurrences and deduplicate (keep chunks from earlier positioned entities)	COMMENT
LOW	lightrag/operate.py	6071	# Step 3: Sort chunks for each entity by occurrence count (higher count = higher priority)	COMMENT
LOW	lightrag/operate.py	6084	# Step 4: Apply the selected chunk selection algorithm	COMMENT
LOW	lightrag/operate.py	6278	# Step 1: Collect all text chunks for each relationship	COMMENT
LOW	lightrag/operate.py	6313	# Step 2: Count chunk occurrences and deduplicate (keep chunks from earlier positioned relationships)	COMMENT
LOW	lightrag/operate.py	6362	# Step 3: Sort chunks for each relationship by occurrence count (higher count = higher priority)	COMMENT
LOW	lightrag/operate.py	6377	# Step 4: Apply the selected chunk selection algorithm	COMMENT
LOW⚡	lightrag/utils.py	6465	# Step 1: Remove chunks that are no longer needed	COMMENT
LOW⚡	lightrag/utils.py	6471	# Step 2: Append genuine additions only (preserving order from new_chunk_ids).	COMMENT
LOW⚡	lightrag/api/lightrag_server.py	787	# Step 3: Create optimized embedding function (calls underlying function directly)	COMMENT
LOW	lightrag/api/lightrag_server.py	660	# Step 1: Import provider function and extract default attributes	COMMENT
LOW	lightrag/api/lightrag_server.py	767	# Step 2: Apply priority (user config > provider default)	COMMENT
LOW	lightrag/api/lightrag_server.py	983	# Step 4: Wrap in EmbeddingFunc and return	COMMENT
LOW	lightrag/chunker/paragraph_semantic.py	1351	# Step 1: expand each oversized table paragraph into row-bounded	COMMENT
LOW	lightrag/chunker/paragraph_semantic.py	1375	# Step 2: greedy-pack pieces into chunks ≤ target_max. A piece	COMMENT
LOW	tests/kg/postgres_impl/test_postgres_migration.py	662	# Step 1: Simulate workspace_a initialization (Case 3 - only legacy exists)	COMMENT
LOW	tests/kg/postgres_impl/test_postgres_migration.py	760	# Step 2: Simulate workspace_b initialization (Case 3 - both exist, but legacy has B's data)	COMMENT
LOW	docs/OfflineDeployment.md	212	### Step 1: Prepare in Online Environment	COMMENT
LOW	docs/OfflineDeployment.md	231	### Step 2: Transfer to Offline Environment	COMMENT
LOW	docs/OfflineDeployment.md	241	### Step 3: Install in Offline Environment	COMMENT
LOW	docs/OfflineDeployment.md	262	### Step 4: Verify Installation	COMMENT
LOW⚡	examples/lightrag_ag2_multiagent_demo.py	267	# Step 1: Set up LightRAG (async, runs on the background loop)	COMMENT
LOW⚡	examples/lightrag_ag2_multiagent_demo.py	270	# Step 2: Create AG2 agents with LightRAG tools	COMMENT
LOW⚡	examples/lightrag_ag2_multiagent_demo.py	273	# Step 3: Ask a complex question	COMMENT

Verbosity Indicators27 hits · 43 pts

Severity	File	Line	Snippet	Context
LOW⚡	lightrag/lightrag.py	1373	# Step 2: Apply priority wrapper decorator to EmbeddingFunc's inner func	COMMENT
LOW	lightrag/lightrag.py	1355	# Step 1: Capture embedding_func and max_token_size before applying rate_limit decorator	COMMENT
LOW⚡	lightrag/operate.py	6138	# Step 5: Batch retrieve chunk data	COMMENT
LOW⚡	lightrag/operate.py	6144	# Step 6: Build result chunks with valid data and update chunk tracking	COMMENT
LOW⚡	lightrag/operate.py	6433	# Step 5: Batch retrieve chunk data	COMMENT
LOW⚡	lightrag/operate.py	6439	# Step 6: Build result chunks with valid data and update chunk tracking	COMMENT
LOW	lightrag/operate.py	6027	# Step 1: Collect all text chunks for each entity	COMMENT
LOW	lightrag/operate.py	6054	# Step 2: Count chunk occurrences and deduplicate (keep chunks from earlier positioned entities)	COMMENT
LOW	lightrag/operate.py	6071	# Step 3: Sort chunks for each entity by occurrence count (higher count = higher priority)	COMMENT
LOW	lightrag/operate.py	6084	# Step 4: Apply the selected chunk selection algorithm	COMMENT
LOW	lightrag/operate.py	6278	# Step 1: Collect all text chunks for each relationship	COMMENT
LOW	lightrag/operate.py	6313	# Step 2: Count chunk occurrences and deduplicate (keep chunks from earlier positioned relationships)	COMMENT
LOW	lightrag/operate.py	6362	# Step 3: Sort chunks for each relationship by occurrence count (higher count = higher priority)	COMMENT
LOW	lightrag/operate.py	6377	# Step 4: Apply the selected chunk selection algorithm	COMMENT
LOW⚡	lightrag/utils.py	6465	# Step 1: Remove chunks that are no longer needed	COMMENT
LOW⚡	lightrag/utils.py	6471	# Step 2: Append genuine additions only (preserving order from new_chunk_ids).	COMMENT
LOW⚡	lightrag/api/lightrag_server.py	787	# Step 3: Create optimized embedding function (calls underlying function directly)	COMMENT
LOW	lightrag/api/lightrag_server.py	660	# Step 1: Import provider function and extract default attributes	COMMENT
LOW	lightrag/api/lightrag_server.py	767	# Step 2: Apply priority (user config > provider default)	COMMENT
LOW	lightrag/api/lightrag_server.py	983	# Step 4: Wrap in EmbeddingFunc and return	COMMENT
LOW	lightrag/chunker/paragraph_semantic.py	1351	# Step 1: expand each oversized table paragraph into row-bounded	COMMENT
LOW	lightrag/chunker/paragraph_semantic.py	1375	# Step 2: greedy-pack pieces into chunks ≤ target_max. A piece	COMMENT
LOW	tests/kg/postgres_impl/test_postgres_migration.py	662	# Step 1: Simulate workspace_a initialization (Case 3 - only legacy exists)	COMMENT
LOW	tests/kg/postgres_impl/test_postgres_migration.py	760	# Step 2: Simulate workspace_b initialization (Case 3 - both exist, but legacy has B's data)	COMMENT
LOW⚡	examples/lightrag_ag2_multiagent_demo.py	267	# Step 1: Set up LightRAG (async, runs on the background loop)	COMMENT
LOW⚡	examples/lightrag_ag2_multiagent_demo.py	270	# Step 2: Create AG2 agents with LightRAG tools	COMMENT
LOW⚡	examples/lightrag_ag2_multiagent_demo.py	273	# Step 3: Ask a complex question	COMMENT

AI Slop Vocabulary15 hits · 40 pts

Severity	File	Line	Snippet	Context
MEDIUM	lightrag/tools/clean_llm_query_cache.py	872	"""Print comprehensive cleanup report	STRING
MEDIUM⚡	lightrag/tools/migrate_llm_cache.py	1526	# Print comprehensive migration report	COMMENT
MEDIUM	lightrag/tools/migrate_llm_cache.py	1358	"""Print comprehensive migration report	STRING
MEDIUM	lightrag/kg/mongo_impl.py	3485	"""Try Atlas Search using compound query for comprehensive matching."""	STRING
MEDIUM	lightrag/kg/shared_storage.py	1295	"""Release all locks with comprehensive error handling, protected from cancellation"""	STRING
MEDIUM	lightrag/api/routers/document_routes.py	6575	# Handle both DocStatus enum and string cases for robust deserialization	COMMENT
MEDIUM	tests/tools/test_migrate_graph_storage_networkx.py	53	# the pgtable harness exercises.	COMMENT
LOW	tests/workspace/test_workspace_isolation.py	696	# back in Test 8.2 and is legitimately False here, while ns_a/ns_b were just set.	COMMENT
MEDIUM⚡	tests/parser/docx/test_smart_heading_e2e.py	321	# harness	COMMENT
MEDIUM⚡	tests/kg/postgres_impl/test_postgres_graph_batch.py	16	# Capture harness	COMMENT
MEDIUM	tests/api/routes/test_aquery_data_endpoint.py	493	"mode": "mix", # Use mixed mode to get the most comprehensive results	CODE
MEDIUM⚡	…sts/chunker/test_paragraph_semantic_drop_references.py	122	# Filtering behaviour (assert on content markers, robust to LevelMerge)	COMMENT
MEDIUM⚡	tests/chunker/test_chunking_raw_lightrag_parity.py	62	# Shared fixtures (mirrors the harness used by test_pipeline_release_closure)	COMMENT
MEDIUM	lightrag_webui/src/features/RetrievalView.tsx	366	// Use the new robust COT parsing function	COMMENT
MEDIUM	lightrag_webui/src/features/RetrievalView.tsx	917	// Handle copying message content with robust clipboard support	COMMENT

Modern Structural Boilerplate39 hits · 39 pts

Severity	File	Line	Snippet	Context
LOW	lightrag/llm_roles.py	139	def set_role_llm_metadata(self, role: str, **metadata: Any) -> None:	CODE
LOW⚡	lightrag/lightrag.py	1028	def _set_runtime_addon_params(self, addon_params: Mapping[str, Any] \| None) -> None:	CODE
LOW	lightrag/__init__.py	5	__all__ = [	CODE
LOW⚡	lightrag/multimodal_context.py	85	logger = logging.getLogger(__name__)	CODE
LOW	lightrag/multimodal_context.py	1060	__all__ = [	CODE
LOW	lightrag/chunk_schema.py	376	__all__ = [	CODE
LOW	lightrag/prompt_multimodal.py	352	__all__ = [	STRING
LOW	lightrag/llm/gemini.py	839	__all__ = [	CODE
LOW	lightrag/llm/azure_openai.py	18	__all__ = [	CODE
LOW	lightrag/parser/_html_table.py	208	__all__ = [	CODE
LOW	lightrag/parser/_markdown.py	68	__all__ = [	CODE
LOW	lightrag/parser/markdown/ir_builder.py	228	__all__ = ["NativeMarkdownIRBuilder"]	CODE
LOW	lightrag/parser/markdown/raw_cache.py	249	__all__ = ["NativeImageRawCache", "native_md_options_signature"]	CODE
LOW	lightrag/parser/legacy/__init__.py	14	__all__ = ["LegacyExtractionError", "extract_text"]	CODE
LOW	lightrag/parser/external/_manifest.py	159	__all__ = [	CODE
LOW	lightrag/parser/external/_zip.py	99	__all__ = ["result_bundle_limits", "safe_extract_zip"]	CODE
LOW	lightrag/parser/external/_common.py	326	__all__ = [	CODE
LOW	lightrag/parser/external/__init__.py	35	__all__ = [	CODE
LOW	lightrag/parser/external/docling/ir_builder.py	1069	__all__ = ["DoclingIRBuilder"]	CODE
LOW	lightrag/parser/external/docling/manifest.py	126	__all__ = [	CODE
LOW	lightrag/parser/external/docling/client.py	523	__all__ = [	CODE
LOW	lightrag/parser/external/docling/cache.py	251	__all__ = [	CODE
LOW	lightrag/parser/external/docling/__init__.py	33	__all__ = [	CODE
LOW	lightrag/parser/external/mineru/ir_builder.py	785	__all__ = ["MinerUIRBuilder"]	CODE
LOW	lightrag/parser/external/mineru/manifest.py	155	__all__ = [	CODE
LOW	lightrag/parser/external/mineru/client.py	742	__all__ = ["MinerURawClient", "CONTENT_LIST_FILENAME"]	CODE
LOW	lightrag/parser/external/mineru/cache.py	421	__all__ = [	CODE
LOW	lightrag/parser/external/mineru/__init__.py	21	__all__ = [	CODE
LOW	lightrag/parser/docx/ir_builder.py	335	__all__ = ["NativeDocxIRBuilder"]	CODE
LOW	lightrag/parser/docx/zip_budget.py	199	__all__ = [	CODE
LOW	lightrag/api/routers/__init__.py	14	__all__ = ["OllamaAPI"]	CODE
LOW	lightrag/evaluation/__init__.py	16	__all__ = ["RAGEvaluator"]	CODE
LOW	lightrag/sidecar/__init__.py	29	__all__ = [	CODE
LOW	lightrag/chunker/__init__.py	68	__all__ = [	CODE
LOW⚡	tests/llm/bedrock_impl/test_bedrock_llm.py	643	def set_role_llm_metadata(self, _role: str, **_metadata) -> None:	CODE
LOW	tests/parser/docx/_native_docx_fixtures.py	329	__all__ = ["Scenario", "SCENARIOS", "_block"]	CODE
LOW	…ts/parser/docx/test_extract_blocks_characterization.py	54	def _set_para_id(para, hex_id: str) -> None:	CODE
LOW	tests/parser/docx/test_smart_heading_features.py	40	def _set_doc_default_size(doc: Document, half_points: int) -> None:	CODE
LOW	tests/api/test_lightrag_ollama_chat.py	70	def set_verbose(cls, verbose: bool) -> None:	CODE

AI Structural Patterns30 hits · 22 pts

Severity	File	Line	Context
LOW	lightrag/llm_roles.py	276	CODE
LOW	lightrag/llm_roles.py	360	CODE
LOW	lightrag/llm_roles.py	395	CODE
LOW	lightrag/operate.py	2745	CODE
LOW	lightrag/operate.py	3477	CODE
LOW	lightrag/utils.py	5051	CODE
LOW	lightrag/pipeline.py	664	CODE
LOW	lightrag/llm/gemini.py	288	CODE
LOW	lightrag/llm/gemini.py	704	CODE
LOW	lightrag/llm/openai.py	305	CODE
LOW	lightrag/llm/openai.py	1032	CODE
LOW	lightrag/llm/openai.py	1054	CODE
LOW	lightrag/llm/openai.py	1216	CODE
LOW	lightrag/llm/openai.py	1298	CODE
LOW	lightrag/llm/openai.py	1307	CODE
LOW	lightrag/llm/llama_index_impl.py	201	CODE
LOW	lightrag/llm/bedrock.py	187	CODE
LOW	lightrag/llm/bedrock.py	531	CODE
LOW	lightrag/llm/hf.py	176	CODE
LOW	lightrag/parser/external/mineru/cache.py	278	CODE
LOW	lightrag/parser/docx/utils.py	752	CODE
LOW	lightrag/parser/docx/utils.py	803	CODE
LOW	lightrag/parser/docx/smart_heading/heading_flow.py	391	CODE
LOW	lightrag/parser/docx/smart_heading/title_block.py	504	CODE
LOW	lightrag/kg/opensearch_impl.py	5063	CODE
LOW	lightrag/kg/scan_job_store.py	258	CODE
LOW	lightrag/chunker/semantic_vector.py	197	CODE
LOW	tests/parser/external/docling/test_client.py	65	CODE
LOW	tests/kg/mongo_impl/test_mongo_scheduling_pages.py	102	CODE
LOW	tests/api/routes/test_document_routes_docx_archive.py	126	CODE

Slop Phrases2 hits · 6 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	examples/unofficial-sample/lightrag_cloudflare_demo.py	32	WORKING_DIR = "../dickens" # you can change output as desired	CODE
MEDIUM	examples/unofficial-sample/lightrag_nvidia_demo.py	112	# so you can adjust to be able to fit the NVIDIA model (future work)	COMMENT

Example Usage Blocks3 hits · 5 pts

Severity	File	Line	Snippet	Context
LOW	docker-compose.podman.yml	3	# Usage:	COMMENT
LOW⚡	lightrag/llm/binding_options.py	792	# Usage:	COMMENT
LOW	scripts/setup/apple-container.sh	35	# Usage:	COMMENT

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings