vector-graph-rag 0.1.3__tar.gz → 0.1.4__tar.gz

{vector_graph_rag-0.1.3 → vector_graph_rag-0.1.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vector-graph-rag
-Version: 0.1.3
+Version: 0.1.4
 Summary: A Graph RAG implementation using pure vector search with Milvus
 Project-URL: Homepage, https://github.com/zilliztech/vector-graph-rag
 Project-URL: Documentation, https://zilliztech.github.io/vector-graph-rag/
@@ -31,14 +31,14 @@ Requires-Dist: scipy>=1.10.0
 Requires-Dist: setuptools<71
 Requires-Dist: tenacity>=8.2.0
 Requires-Dist: tiktoken>=0.5.0
-Requires-Dist: torch>=2.0.0
 Requires-Dist: tqdm>=4.65.0
-Requires-Dist: transformers>=4.30.0
 Provides-Extra: all
 Requires-Dist: fastapi>=0.109.0; extra == 'all'
 Requires-Dist: pytest-asyncio>=0.21.0; extra == 'all'
 Requires-Dist: pytest>=7.0.0; extra == 'all'
 Requires-Dist: ruff>=0.1.0; extra == 'all'
+Requires-Dist: torch>=2.0.0; extra == 'all'
+Requires-Dist: transformers>=4.30.0; extra == 'all'
 Requires-Dist: uvicorn[standard]>=0.27.0; extra == 'all'
 Provides-Extra: api
 Requires-Dist: fastapi>=0.109.0; extra == 'api'
@@ -49,6 +49,9 @@ Requires-Dist: pytest>=7.0.0; extra == 'dev'
 Requires-Dist: ruff>=0.1.0; extra == 'dev'
 Provides-Extra: docs
 Requires-Dist: mkdocs-material>=9.0.0; extra == 'docs'
+Provides-Extra: hf
+Requires-Dist: torch>=2.0.0; extra == 'hf'
+Requires-Dist: transformers>=4.30.0; extra == 'hf'
 Provides-Extra: loaders
 Requires-Dist: lxml-html-clean; extra == 'loaders'
 Requires-Dist: markitdown[docx,pdf]>=0.1.4; extra == 'loaders'
@@ -108,6 +111,17 @@ uv add "vector-graph-rag[loaders]"
 
 </details>
 
+<details>
+<summary><b>With local HuggingFace embedding models</b></summary>
+
+```bash
+pip install "vector-graph-rag[hf]"
+# or
+uv add "vector-graph-rag[hf]"
+```
+
+</details>
+
 ## 🚀 Quick Start
 
 ```python

{vector_graph_rag-0.1.3 → vector_graph_rag-0.1.4}/README.md

@@ -51,6 +51,17 @@ uv add "vector-graph-rag[loaders]"
 
 </details>
 
+<details>
+<summary><b>With local HuggingFace embedding models</b></summary>
+
+```bash
+pip install "vector-graph-rag[hf]"
+# or
+uv add "vector-graph-rag[hf]"
+```
+
+</details>
+
 ## 🚀 Quick Start
 
 ```python

{vector_graph_rag-0.1.3 → vector_graph_rag-0.1.4}/docs/faq.md

@@ -70,7 +70,7 @@ Frequently asked questions about Vector Graph RAG — covering when to use it, c
     ```
 
 ??? note "Can I use my own embeddings?"
-    Vector Graph RAG uses OpenAI embedding models by default (`text-embedding-3-large`), but you can configure the embedding model via the `embedding_model` parameter. Any model accessible through the OpenAI-compatible API will work. If you are using a local or custom embedding endpoint, set the appropriate base URL and model name. The embedding dimensionality is detected automatically. Note that all entities, relations, and passages in a single graph must use the same embedding model — mixing models within one collection prefix is not supported.
+    Vector Graph RAG uses OpenAI embedding models by default (`text-embedding-3-large`), but you can configure the embedding model via the `embedding_model` parameter. Any model accessible through the OpenAI-compatible API will work. If you are using a local HuggingFace embedding model, install the optional dependencies with `pip install "vector-graph-rag[hf]"`. The embedding dimensionality is detected automatically. Note that all entities, relations, and passages in a single graph must use the same embedding model — mixing models within one collection prefix is not supported.
 
 ??? note "How do I deploy to production?"
     For production deployments, use a remote Milvus instance instead of Milvus Lite for better performance, scalability, and persistence. Run the FastAPI backend behind a reverse proxy (e.g., Nginx) with appropriate rate limiting and authentication. The frontend can be built as static files and served from any CDN or static file server.

{vector_graph_rag-0.1.3 → vector_graph_rag-0.1.4}/docs/getting-started.md

@@ -20,6 +20,12 @@
     pip install "vector-graph-rag[loaders]"
     ```
 
+=== "With local embeddings"
+
+    ```bash
+    pip install "vector-graph-rag[hf]"
+    ```
+
 !!! note "Prerequisites"
     - Python 3.9+
     - An OpenAI API key (set `OPENAI_API_KEY` environment variable)

{vector_graph_rag-0.1.3 → vector_graph_rag-0.1.4}/evaluation/README.md

@@ -74,7 +74,7 @@ These triplets were extracted using GPT-3.5-Turbo (1106) with HippoRAG's OpenIE
 
 1. Install dependencies:
 ```bash
-uv sync
+uv sync --extra hf
 ```
 
 2. Set environment variables:

{vector_graph_rag-0.1.3 → vector_graph_rag-0.1.4}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "vector-graph-rag"
-version = "0.1.3"
+version = "0.1.4"
 description = "A Graph RAG implementation using pure vector search with Milvus"
 readme = "README.md"
 license = { text = "MIT" }
@@ -32,8 +32,6 @@ dependencies = [
     "tqdm>=4.65.0",
     "tiktoken>=0.5.0",
     "tenacity>=8.2.0",
-    "transformers>=4.30.0",
-    "torch>=2.0.0",
     "langchain-core>=0.3.0",
     "python-multipart>=0.0.22",
     "setuptools<71",  # Required for pkg_resources used by milvus-lite
@@ -49,8 +47,12 @@ api = [
     "fastapi>=0.109.0",
     "uvicorn[standard]>=0.27.0",
 ]
+hf = [
+    "transformers>=4.30.0",
+    "torch>=2.0.0",
+]
 all = [
-    "vector-graph-rag[dev,api]",
+    "vector-graph-rag[dev,api,hf]",
 ]
 loaders = [
     "markitdown[docx,pdf]>=0.1.4",

{vector_graph_rag-0.1.3 → vector_graph_rag-0.1.4}/src/vector_graph_rag/__init__.py

@@ -6,19 +6,19 @@ using only vector similarity search, without requiring a separate graph database
 """
 
 from vector_graph_rag.config import Settings
-from vector_graph_rag.models import Document, Triplet, Entity, Relation, Passage
-from vector_graph_rag.llm.extractor import TripletExtractor
-from vector_graph_rag.storage.embeddings import EmbeddingModel
-from vector_graph_rag.storage.milvus import MilvusStore
 from vector_graph_rag.graph.builder import GraphBuilder
-from vector_graph_rag.graph.retriever import GraphRetriever
-from vector_graph_rag.graph.knowledge_graph import SubGraph
 from vector_graph_rag.graph.graph import Graph
+from vector_graph_rag.graph.knowledge_graph import SubGraph
+from vector_graph_rag.graph.retriever import GraphRetriever
+from vector_graph_rag.llm.cache import LLMCache, get_llm_cache
+from vector_graph_rag.llm.extractor import TripletExtractor
 from vector_graph_rag.llm.reranker import LLMReranker
+from vector_graph_rag.models import Document, Entity, Passage, Relation, Triplet
 from vector_graph_rag.rag import VectorGraphRAG, create_rag
-from vector_graph_rag.llm.cache import LLMCache, get_llm_cache
+from vector_graph_rag.storage.embeddings import EmbeddingModel
+from vector_graph_rag.storage.milvus import MilvusStore
 
-__version__ = "0.1.3"
+__version__ = "0.1.4"
 
 __all__ = [
     "Settings",

{vector_graph_rag-0.1.3 → vector_graph_rag-0.1.4}/src/vector_graph_rag/api/app.py

@@ -9,34 +9,35 @@ Provides RESTful API endpoints for:
 - Graph exploration (stats, neighbors)
 """
 
-import os
-import tempfile
 import shutil
+import tempfile
 from pathlib import Path
-from typing import List, Optional, Dict, Any
-from fastapi import FastAPI, HTTPException, Query, File, UploadFile, Form
+from typing import Any, Dict, List, Optional
+
+from fastapi import FastAPI, File, Form, HTTPException, Query, UploadFile
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.staticfiles import StaticFiles
 from fastapi.responses import FileResponse
+from fastapi.staticfiles import StaticFiles
 from pydantic import BaseModel, Field
 
 from vector_graph_rag import VectorGraphRAG, __version__
 from vector_graph_rag.config import Settings, get_settings
-from vector_graph_rag.storage.milvus import MilvusStore
 from vector_graph_rag.graph.graph import Graph
-from vector_graph_rag.models import Triplet
-
+from vector_graph_rag.storage.milvus import MilvusStore
 
 # ==================== Request/Response Schemas ====================
 
+
 class HealthResponse(BaseModel):
     """Health check response."""
+
     status: str = Field(default="ok", description="Service status")
     version: str = Field(default=__version__, description="API version")
 
 
 class GraphInfo(BaseModel):
     """Information about a graph."""
+
     name: str = Field(..., description="Graph name (collection prefix)")
     entity_collection: str = Field(..., description="Entity collection name")
     relation_collection: str = Field(..., description="Relation collection name")
@@ -46,11 +47,13 @@ class GraphInfo(BaseModel):
 
 class ListGraphsResponse(BaseModel):
     """Response for listing graphs."""
+
     graphs: List[GraphInfo] = Field(default_factory=list, description="Available graphs")
 
 
 class TripletInput(BaseModel):
     """Input for a triplet."""
+
     subject: str = Field(..., description="Subject entity")
     predicate: str = Field(..., description="Predicate/relationship")
     object: str = Field(..., description="Object entity")
@@ -58,17 +61,20 @@ class TripletInput(BaseModel):
 
 class AddDocumentsRequest(BaseModel):
     """Request to add documents."""
+
     documents: List[str] = Field(..., description="List of document texts")
     ids: Optional[List[str]] = Field(default=None, description="Optional list of document IDs")
-    extract_triplets: bool = Field(default=True, description="Whether to extract triplets using LLM")
+    extract_triplets: bool = Field(
+        default=True, description="Whether to extract triplets using LLM"
+    )
     triplets: Optional[List[List[TripletInput]]] = Field(
-        default=None,
-        description="Optional pre-extracted triplets for each document"
+        default=None, description="Optional pre-extracted triplets for each document"
     )
 
 
 class AddDocumentsResponse(BaseModel):
     """Response after adding documents."""
+
     num_documents: int = Field(..., description="Number of documents added")
     num_entities: int = Field(..., description="Number of entities created")
     num_relations: int = Field(..., description="Number of relations created")
@@ -77,10 +83,11 @@ class AddDocumentsResponse(BaseModel):
 
 class ImportRequest(BaseModel):
     """Request to import documents from various sources."""
+
     sources: List[str] = Field(
         ...,
         description="List of file paths or URLs to import",
-        examples=["/path/to/doc.pdf", "https://example.com/article"]
+        examples=["/path/to/doc.pdf", "https://example.com/article"],
     )
     chunk_documents: bool = Field(True, description="Whether to chunk large documents")
     chunk_size: int = Field(1000, description="Max characters per chunk")
@@ -91,6 +98,7 @@ class ImportRequest(BaseModel):
 
 class ImportResponse(BaseModel):
     """Response from document import."""
+
     success: bool
     num_sources: int = Field(..., description="Number of input sources")
     num_documents: int = Field(..., description="Number of documents after conversion")
@@ -102,15 +110,19 @@ class ImportResponse(BaseModel):
 
 class QueryRequest(BaseModel):
     """Request to query the knowledge base."""
+
     question: str = Field(..., description="Question to answer")
     use_reranking: bool = Field(default=True, description="Whether to use LLM reranking")
     entity_top_k: Optional[int] = Field(default=None, description="Number of entities to retrieve")
-    relation_top_k: Optional[int] = Field(default=None, description="Number of relations to retrieve")
+    relation_top_k: Optional[int] = Field(
+        default=None, description="Number of relations to retrieve"
+    )
     expansion_degree: Optional[int] = Field(default=None, description="Subgraph expansion degree")
 
 
 class EntitySchema(BaseModel):
     """Schema for an entity in the subgraph."""
+
     id: str = Field(..., description="Entity ID")
     name: str = Field(..., description="Entity name")
     relation_ids: List[str] = Field(default_factory=list, description="Connected relation IDs")
@@ -119,61 +131,84 @@ class EntitySchema(BaseModel):
 
 class RelationSchema(BaseModel):
     """Schema for a relation in the subgraph."""
+
     id: str = Field(..., description="Relation ID")
     text: str = Field(..., description="Full relation text")
     subject: str = Field(..., description="Subject entity name")
     predicate: str = Field(..., description="Predicate/relationship")
     object: str = Field(..., description="Object entity name")
-    entity_ids: List[str] = Field(default_factory=list, description="Connected entity IDs [subject_id, object_id]")
+    entity_ids: List[str] = Field(
+        default_factory=list, description="Connected entity IDs [subject_id, object_id]"
+    )
     passage_ids: List[str] = Field(default_factory=list, description="Source passage IDs")
 
 
 class PassageSchema(BaseModel):
     """Schema for a passage in the subgraph."""
+
     id: str = Field(..., description="Passage ID")
     text: str = Field(..., description="Passage text")
 
 
 class ExpansionStepSchema(BaseModel):
     """Schema for an expansion step in the history."""
+
     step: int = Field(..., description="Step number")
     operation: str = Field(..., description="Operation type")
     description: Optional[str] = Field(default=None, description="Operation description")
-    added_entity_ids: List[str] = Field(default_factory=list, description="Entity IDs added in this step")
-    added_relation_ids: List[str] = Field(default_factory=list, description="Relation IDs added in this step")
+    added_entity_ids: List[str] = Field(
+        default_factory=list, description="Entity IDs added in this step"
+    )
+    added_relation_ids: List[str] = Field(
+        default_factory=list, description="Relation IDs added in this step"
+    )
     total_entities: int = Field(default=0, description="Total entities after this step")
     total_relations: int = Field(default=0, description="Total relations after this step")
 
 
 class SubGraphSchema(BaseModel):
     """Schema for the expanded subgraph."""
+
     entity_ids: List[str] = Field(default_factory=list, description="All entity IDs in subgraph")
-    relation_ids: List[str] = Field(default_factory=list, description="All relation IDs in subgraph")
+    relation_ids: List[str] = Field(
+        default_factory=list, description="All relation IDs in subgraph"
+    )
     passage_ids: List[str] = Field(default_factory=list, description="All passage IDs in subgraph")
     entities: List[EntitySchema] = Field(default_factory=list, description="Entity details")
     relations: List[RelationSchema] = Field(default_factory=list, description="Relation details")
     passages: List[PassageSchema] = Field(default_factory=list, description="Passage details")
-    expansion_history: List[ExpansionStepSchema] = Field(default_factory=list, description="Expansion history")
+    expansion_history: List[ExpansionStepSchema] = Field(
+        default_factory=list, description="Expansion history"
+    )
 
 
 class RetrievalDetailSchema(BaseModel):
     """Schema for retrieval details."""
+
     entity_ids: List[str] = Field(default_factory=list, description="Retrieved entity IDs")
     entity_texts: List[str] = Field(default_factory=list, description="Retrieved entity names")
     entity_scores: List[float] = Field(default_factory=list, description="Entity similarity scores")
     relation_ids: List[str] = Field(default_factory=list, description="Retrieved relation IDs")
     relation_texts: List[str] = Field(default_factory=list, description="Retrieved relation texts")
-    relation_scores: List[float] = Field(default_factory=list, description="Relation similarity scores")
+    relation_scores: List[float] = Field(
+        default_factory=list, description="Relation similarity scores"
+    )
 
 
 class RerankResultSchema(BaseModel):
     """Schema for rerank results."""
-    selected_relation_ids: List[str] = Field(default_factory=list, description="Selected relation IDs")
-    selected_relation_texts: List[str] = Field(default_factory=list, description="Selected relation texts")
+
+    selected_relation_ids: List[str] = Field(
+        default_factory=list, description="Selected relation IDs"
+    )
+    selected_relation_texts: List[str] = Field(
+        default_factory=list, description="Selected relation texts"
+    )
 
 
 class EvictionResultSchema(BaseModel):
     """Schema for eviction results (when relations exceed threshold)."""
+
     occurred: bool = Field(default=False, description="Whether eviction occurred")
     before_count: int = Field(default=0, description="Number of relations before eviction")
     after_count: int = Field(default=0, description="Number of relations after eviction")
@@ -181,44 +216,62 @@ class EvictionResultSchema(BaseModel):
 
 class QueryResponse(BaseModel):
     """Response for a query."""
+
     question: str = Field(..., description="Original question")
     answer: str = Field(..., description="Generated answer")
-    query_entities: List[str] = Field(default_factory=list, description="Entities extracted from query")
-    subgraph: Optional[SubGraphSchema] = Field(default=None, description="Expanded subgraph for visualization")
+    query_entities: List[str] = Field(
+        default_factory=list, description="Entities extracted from query"
+    )
+    subgraph: Optional[SubGraphSchema] = Field(
+        default=None, description="Expanded subgraph for visualization"
+    )
     retrieved_passages: List[str] = Field(default_factory=list, description="Retrieved passages")
     stats: Dict[str, Any] = Field(default_factory=dict, description="Query statistics")
-    retrieval_detail: Optional[RetrievalDetailSchema] = Field(default=None, description="Initial retrieval details")
-    rerank_result: Optional[RerankResultSchema] = Field(default=None, description="LLM rerank results")
-    eviction_result: Optional[EvictionResultSchema] = Field(default=None, description="Eviction results if relations exceeded threshold")
+    retrieval_detail: Optional[RetrievalDetailSchema] = Field(
+        default=None, description="Initial retrieval details"
+    )
+    rerank_result: Optional[RerankResultSchema] = Field(
+        default=None, description="LLM rerank results"
+    )
+    eviction_result: Optional[EvictionResultSchema] = Field(
+        default=None, description="Eviction results if relations exceeded threshold"
+    )
 
 
 class DocumentResponse(BaseModel):
     """Response for a single document."""
+
     id: str = Field(..., description="Document ID")
     text: str = Field(..., description="Document text")
     entity_ids: List[str] = Field(default_factory=list, description="Entity IDs in this document")
-    relation_ids: List[str] = Field(default_factory=list, description="Relation IDs from this document")
+    relation_ids: List[str] = Field(
+        default_factory=list, description="Relation IDs from this document"
+    )
 
 
 class ListDocumentsResponse(BaseModel):
     """Response for listing documents."""
+
     documents: List[DocumentResponse] = Field(default_factory=list, description="List of documents")
     total: int = Field(..., description="Total number of documents returned")
 
 
 class UpdateDocumentRequest(BaseModel):
     """Request to update a document."""
+
     text: Optional[str] = Field(default=None, description="New document text")
 
 
 class DeleteResponse(BaseModel):
     """Response for delete operations."""
+
     success: bool = Field(..., description="Whether deletion succeeded")
     message: str = Field(default="", description="Additional message")
 
 
 class GraphStatsResponse(BaseModel):
     """Response for graph statistics."""
+
     graph_name: str = Field(..., description="Graph name")
     entity_count: int = Field(default=0, description="Number of entities")
     relation_count: int = Field(default=0, description="Number of relations")
@@ -227,13 +280,17 @@ class GraphStatsResponse(BaseModel):
 
 class NeighborResponse(BaseModel):
     """Response for entity neighbors."""
+
     entity_id: str = Field(..., description="Central entity ID")
     neighbors: List[EntitySchema] = Field(default_factory=list, description="Neighbor entities")
-    relations: List[RelationSchema] = Field(default_factory=list, description="Connecting relations")
+    relations: List[RelationSchema] = Field(
+        default_factory=list, description="Connecting relations"
+    )
 
 
 class SettingsResponse(BaseModel):
     """Response for system settings."""
+
     llm_model: str = Field(..., description="LLM model name")
     embedding_model: str = Field(..., description="Embedding model name")
     embedding_dimension: int = Field(..., description="Embedding dimension")
@@ -245,6 +302,7 @@ class SettingsResponse(BaseModel):
 
 # ==================== Application Factory ====================
 
+
 def create_app(settings: Optional[Settings] = None) -> FastAPI:
     """
     Create and configure the FastAPI application.
@@ -314,9 +372,7 @@ def create_app(settings: Optional[Settings] = None) -> FastAPI:
             milvus_token=app.state.settings.milvus_token,
             milvus_db=app.state.settings.milvus_db,
         )
-        return ListGraphsResponse(
-            graphs=[GraphInfo(**g) for g in graphs]
-        )
+        return ListGraphsResponse(graphs=[GraphInfo(**g) for g in graphs])
 
     @app.get("/settings", response_model=SettingsResponse, tags=["System"])
     async def get_system_settings():
@@ -362,19 +418,15 @@ def create_app(settings: Optional[Settings] = None) -> FastAPI:
             if deleted:
                 return DeleteResponse(
                     success=True,
-                    message=f"Successfully deleted graph '{graph_name}' and all its collections"
+                    message=f"Successfully deleted graph '{graph_name}' and all its collections",
                 )
             else:
                 return DeleteResponse(
-                    success=False,
-                    message=f"Graph '{graph_name}' not found or already deleted"
+                    success=False, message=f"Graph '{graph_name}' not found or already deleted"
                 )
 
         except Exception as e:
-            raise HTTPException(
-                status_code=500,
-                detail=f"Failed to delete graph: {str(e)}"
-            )
+            raise HTTPException(status_code=500, detail=f"Failed to delete graph: {str(e)}")
 
     @app.post("/add_documents", response_model=AddDocumentsResponse, tags=["Documents"])
     async def add_documents(
@@ -395,10 +447,9 @@ def create_app(settings: Optional[Settings] = None) -> FastAPI:
             for i, text in enumerate(request.documents):
                 doc_data = {
                     "passage": text,
-                    "triplets": [
-                        [t.subject, t.predicate, t.object]
-                        for t in request.triplets[i]
-                    ] if i < len(request.triplets) else [],
+                    "triplets": [[t.subject, t.predicate, t.object] for t in request.triplets[i]]
+                    if i < len(request.triplets)
+                    else [],
                 }
                 if request.ids and i < len(request.ids):
                     doc_data["id"] = request.ids[i]
@@ -575,9 +626,9 @@ def create_app(settings: Optional[Settings] = None) -> FastAPI:
         if result.subgraph:
             sg = result.subgraph
             subgraph_schema = SubGraphSchema(
-                entity_ids=list(sg.entity_ids) if hasattr(sg, 'entity_ids') else [],
-                relation_ids=list(sg.relation_ids) if hasattr(sg, 'relation_ids') else [],
-                passage_ids=list(sg.passage_ids) if hasattr(sg, 'passage_ids') else [],
+                entity_ids=list(sg.entity_ids) if hasattr(sg, "entity_ids") else [],
+                relation_ids=list(sg.relation_ids) if hasattr(sg, "relation_ids") else [],
+                passage_ids=list(sg.passage_ids) if hasattr(sg, "passage_ids") else [],
                 entities=[
                     EntitySchema(
                         id=e.id,
@@ -585,7 +636,7 @@ def create_app(settings: Optional[Settings] = None) -> FastAPI:
                         relation_ids=e.relation_ids,
                         passage_ids=e.passage_ids,
                     )
-                    for e in (sg.entities if hasattr(sg, 'entities') else [])
+                    for e in (sg.entities if hasattr(sg, "entities") else [])
                 ],
                 relations=[
                     RelationSchema(
@@ -597,23 +648,27 @@ def create_app(settings: Optional[Settings] = None) -> FastAPI:
                         entity_ids=r.entity_ids,
                         passage_ids=r.passage_ids,
                     )
-                    for r in (sg.relations if hasattr(sg, 'relations') else [])
+                    for r in (sg.relations if hasattr(sg, "relations") else [])
                 ],
                 passages=[
                     PassageSchema(id=p.id, text=p.text)
-                    for p in (sg.passages if hasattr(sg, 'passages') else [])
+                    for p in (sg.passages if hasattr(sg, "passages") else [])
                 ],
                 expansion_history=[
                     ExpansionStepSchema(
-                        step=h.get('step', i),
-                        operation=h.get('operation', 'unknown'),
-                        description=h.get('description'),
-                        added_entity_ids=h.get('added_entity_ids', h.get('new_entity_ids', [])),
-                        added_relation_ids=h.get('added_relation_ids', h.get('new_relation_ids', [])),
-                        total_entities=h.get('total_entities', 0),
-                        total_relations=h.get('total_relations', 0),
+                        step=h.get("step", i),
+                        operation=h.get("operation", "unknown"),
+                        description=h.get("description"),
+                        added_entity_ids=h.get("added_entity_ids", h.get("new_entity_ids", [])),
+                        added_relation_ids=h.get(
+                            "added_relation_ids", h.get("new_relation_ids", [])
+                        ),
+                        total_entities=h.get("total_entities", 0),
+                        total_relations=h.get("total_relations", 0),
+                    )
+                    for i, h in enumerate(
+                        sg.expansion_history if hasattr(sg, "expansion_history") else []
                     )
-                    for i, h in enumerate(sg.expansion_history if hasattr(sg, 'expansion_history') else [])
                 ],
             )
 
@@ -681,11 +736,11 @@ def create_app(settings: Optional[Settings] = None) -> FastAPI:
             stats = graph.get_stats()
             return GraphStatsResponse(
                 graph_name=graph_name,
-                entity_count=stats.get('entity_count', 0),
-                relation_count=stats.get('relation_count', 0),
-                passage_count=stats.get('passage_count', 0),
+                entity_count=stats.get("entity_count", 0),
+                relation_count=stats.get("relation_count", 0),
+                passage_count=stats.get("passage_count", 0),
             )
-        except Exception as e:
+        except Exception:
             # If stats not available, return zeros
             return GraphStatsResponse(
                 graph_name=graph_name,
@@ -694,7 +749,9 @@ def create_app(settings: Optional[Settings] = None) -> FastAPI:
                 passage_count=0,
             )
 
-    @app.get("/graph/{graph_name}/neighbors/{entity_id}", response_model=NeighborResponse, tags=["Graph"])
+    @app.get(
+        "/graph/{graph_name}/neighbors/{entity_id}", response_model=NeighborResponse, tags=["Graph"]
+    )
     async def get_entity_neighbors(
         graph_name: str,
         entity_id: str,
@@ -729,12 +786,14 @@ def create_app(settings: Optional[Settings] = None) -> FastAPI:
             for nid in list(neighbor_ids)[:limit]:
                 neighbor = graph.get_entity(nid)
                 if neighbor:
-                    neighbors.append(EntitySchema(
-                        id=neighbor.id,
-                        name=neighbor.name,
-                        relation_ids=neighbor.relation_ids,
-                        passage_ids=neighbor.passage_ids,
-                    ))
+                    neighbors.append(
+                        EntitySchema(
+                            id=neighbor.id,
+                            name=neighbor.name,
+                            relation_ids=neighbor.relation_ids,
+                            passage_ids=neighbor.passage_ids,
+                        )
+                    )
 
             # Convert relations to schema
             relation_schemas = [
@@ -786,7 +845,9 @@ def create_app(settings: Optional[Settings] = None) -> FastAPI:
     @app.get("/documents", response_model=ListDocumentsResponse, tags=["Documents"])
     async def list_documents(
         graph_name: Optional[str] = Query(default=None, description="Graph name to use"),
-        query: Optional[str] = Query(default=None, description="Optional search query for vector similarity"),
+        query: Optional[str] = Query(
+            default=None, description="Optional search query for vector similarity"
+        ),
         top_k: int = Query(default=10, description="Number of documents to return"),
     ):
         """
@@ -921,6 +982,7 @@ def run_server(host: str = "0.0.0.0", port: int = 8000, reload: bool = False):
         reload: Whether to enable auto-reload.
     """
     import uvicorn
+
     uvicorn.run("vector_graph_rag.api.app:app", host=host, port=port, reload=reload)