aiagents4pharma 1.45.1__py3-none-any.whl → 1.46.1__py3-none-any.whl

aiagents4pharma/talk2knowledgegraphs/utils/extractions/milvus_multimodal_pcst.py

@@ -2,8 +2,8 @@
 Exctraction of multimodal subgraph using Prize-Collecting Steiner Tree (PCST) algorithm.
 """
 
+import asyncio
 import logging
-import pickle
 import platform
 import subprocess
 from typing import NamedTuple
@@ -14,8 +14,8 @@ import pcst_fast
 from pymilvus import Collection
 
 try:
-    import cudf
-    import cupy as cp
+    import cudf  # type: ignore
+    import cupy as cp  # type: ignore
 
     CUDF_AVAILABLE = True
 except ImportError:
@@ -217,6 +217,110 @@ class MultimodalPCSTPruning(NamedTuple):
 
         return colls
 
+    async def load_edge_index_async(self, cfg: dict, _connection_manager=None) -> np.ndarray:
+        """
+        Load edge index using hybrid async/sync approach to avoid event loop issues.
+
+        This method queries the edges collection to get head_index and tail_index,
+        eliminating the need for pickle caching and reducing memory usage.
+
+        Args:
+            cfg: The configuration dictionary containing the Milvus setup.
+            _connection_manager: Unused parameter for interface compatibility.
+
+        Returns:
+            numpy.ndarray: Edge index array with shape [2, num_edges]
+        """
+        logger.log(logging.INFO, "Loading edge index from Milvus collection (hybrid)")
+
+        def load_edges_sync():
+            """Load edges synchronously to avoid event loop issues."""
+
+            collection_name = f"{cfg.milvus_db.database_name}_edges"
+            edges_collection = Collection(name=collection_name)
+            edges_collection.load()
+
+            # Query all edges in batches
+            batch_size = getattr(cfg.milvus_db, "query_batch_size", 10000)
+            total_entities = edges_collection.num_entities
+            logger.log(logging.INFO, "Total edges to process: %d", total_entities)
+
+            head_list = []
+            tail_list = []
+
+            for start in range(0, total_entities, batch_size):
+                end = min(start + batch_size, total_entities)
+                logger.debug("Processing edge batch: %d to %d", start, end)
+
+                batch = edges_collection.query(
+                    expr=f"triplet_index >= {start} and triplet_index < {end}",
+                    output_fields=["head_index", "tail_index"],
+                )
+
+                head_list.extend([r["head_index"] for r in batch])
+                tail_list.extend([r["tail_index"] for r in batch])
+
+            # Convert to numpy array format expected by PCST
+            edge_index = self.loader.py.array([head_list, tail_list])
+            logger.log(
+                logging.INFO,
+                "Edge index loaded (hybrid): shape %s",
+                str(edge_index.shape),
+            )
+
+            return edge_index
+
+        # Run in thread to avoid event loop conflicts
+        return await asyncio.to_thread(load_edges_sync)
+
+    def load_edge_index(self, cfg: dict) -> np.ndarray:
+        """
+        Load edge index synchronously from Milvus collection.
+
+        This method queries the edges collection to get head_index and tail_index.
+
+        Args:
+            cfg: The configuration dictionary containing the Milvus setup.
+
+        Returns:
+            numpy.ndarray: Edge index array with shape [2, num_edges]
+        """
+        logger.log(logging.INFO, "Loading edge index from Milvus collection (sync)")
+
+        collection_name = f"{cfg.milvus_db.database_name}_edges"
+        edges_collection = Collection(name=collection_name)
+        edges_collection.load()
+
+        # Query all edges in batches
+        batch_size = getattr(cfg.milvus_db, "query_batch_size", 10000)
+        total_entities = edges_collection.num_entities
+        logger.log(logging.INFO, "Total edges to process: %d", total_entities)
+
+        head_list = []
+        tail_list = []
+
+        for start in range(0, total_entities, batch_size):
+            end = min(start + batch_size, total_entities)
+            logger.debug("Processing edge batch: %d to %d", start, end)
+
+            batch = edges_collection.query(
+                expr=f"triplet_index >= {start} and triplet_index < {end}",
+                output_fields=["head_index", "tail_index"],
+            )
+
+            head_list.extend([r["head_index"] for r in batch])
+            tail_list.extend([r["tail_index"] for r in batch])
+
+        # Convert to numpy array format expected by PCST
+        edge_index = self.loader.py.array([head_list, tail_list])
+        logger.log(
+            logging.INFO,
+            "Edge index loaded (sync): shape %s",
+            str(edge_index.shape),
+        )
+
+        return edge_index
+
     def _compute_node_prizes(self, query_emb: list, colls: dict) -> dict:
         """
         Compute the node prizes based on the similarity between the query and nodes.
@@ -263,6 +367,56 @@ class MultimodalPCSTPruning(NamedTuple):
 
         return n_prizes
 
+    async def _compute_node_prizes_async(
+        self,
+        query_emb: list,
+        collection_name: str,
+        connection_manager,
+        use_description: bool = False,
+    ) -> dict:
+        """
+        Compute the node prizes asynchronously using connection manager.
+
+        Args:
+            query_emb: The query embedding
+            collection_name: Name of the collection to search
+            connection_manager: The MilvusConnectionManager instance
+            use_description: Whether to use description embeddings
+
+        Returns:
+            The prizes of the nodes
+        """
+        # Get collection stats for initialization
+        stats = await connection_manager.async_get_collection_stats(collection_name)
+        num_entities = stats["num_entities"]
+
+        # Initialize prizes array
+        topk = min(self.topk, num_entities)
+        n_prizes = self.loader.py.zeros(num_entities, dtype=self.loader.py.float32)
+
+        # Get the actual metric type to use
+        actual_metric_type = self.metric_type or self.loader.metric_type
+
+        # Determine search field based on use_description
+        anns_field = "desc_emb" if use_description else "feat_emb"
+
+        # Perform async search
+        results = await connection_manager.async_search(
+            collection_name=collection_name,
+            data=[query_emb],
+            anns_field=anns_field,
+            param={"metric_type": actual_metric_type},
+            limit=topk,
+            output_fields=["node_id"],
+        )
+
+        # Update the prizes based on the search results
+        if results and len(results) > 0:
+            result_ids = [hit["id"] for hit in results[0]]
+            n_prizes[result_ids] = self.loader.py.arange(topk, 0, -1).astype(self.loader.py.float32)
+
+        return n_prizes
+
     def _compute_edge_prizes(self, text_emb: list, colls: dict):
         """
         Compute the edge prizes based on the similarity between the query and edges.
@@ -305,6 +459,65 @@ class MultimodalPCSTPruning(NamedTuple):
 
         return e_prizes
 
+    async def _compute_edge_prizes_async(
+        self, text_emb: list, collection_name: str, connection_manager
+    ) -> dict:
+        """
+        Compute the edge prizes asynchronously using connection manager.
+
+        Args:
+            text_emb: The textual description embedding
+            collection_name: Name of the edges collection
+            connection_manager: The MilvusConnectionManager instance
+
+        Returns:
+            The prizes of the edges
+        """
+        # Get collection stats for initialization
+        stats = await connection_manager.async_get_collection_stats(collection_name)
+        num_entities = stats["num_entities"]
+
+        # Initialize prizes array
+        topk_e = min(self.topk_e, num_entities)
+        e_prizes = self.loader.py.zeros(num_entities, dtype=self.loader.py.float32)
+
+        # Get the actual metric type to use
+        actual_metric_type = self.metric_type or self.loader.metric_type
+
+        # Perform async search
+        results = await connection_manager.async_search(
+            collection_name=collection_name,
+            data=[text_emb],
+            anns_field="feat_emb",
+            param={"metric_type": actual_metric_type},
+            limit=topk_e,
+            output_fields=["head_id", "tail_id"],
+        )
+
+        # Update the prizes based on the search results
+        if results and len(results) > 0:
+            result_ids = [hit["id"] for hit in results[0]]
+            result_scores = [hit["distance"] for hit in results[0]]  # Use distance/score
+            e_prizes[result_ids] = result_scores
+
+        # Process edge prizes using helper method
+        return self._process_edge_prizes(e_prizes, topk_e)
+
+    def _process_edge_prizes(self, e_prizes, topk_e):
+        """Helper method to process edge prizes and reduce complexity."""
+        unique_prizes, inverse_indices = self.loader.py.unique(e_prizes, return_inverse=True)
+        sorted_indices = self.loader.py.argsort(-unique_prizes)[:topk_e]
+        topk_e_values = unique_prizes[sorted_indices]
+        last_topk_e_value = topk_e
+
+        for k in range(topk_e):
+            indices = inverse_indices == (unique_prizes == topk_e_values[k]).nonzero()[0]
+            value = min((topk_e - k) / indices.sum().item(), last_topk_e_value)
+            e_prizes[indices] = value
+            last_topk_e_value = value * (1 - self.c_const)
+
+        return e_prizes
+
     def compute_prizes(self, text_emb: list, query_emb: list, colls: dict) -> dict:
         """
         Compute the node prizes based on the cosine similarity between the query and nodes,
@@ -331,6 +544,27 @@ class MultimodalPCSTPruning(NamedTuple):
 
         return {"nodes": n_prizes, "edges": e_prizes}
 
+    async def compute_prizes_async(
+        self, text_emb: list, query_emb: list, cfg: dict, modality: str
+    ) -> dict:
+        """
+        Compute node and edge prizes asynchronously in parallel using sync fallback.
+
+        Args:
+            text_emb: The textual description embedding
+            query_emb: The query embedding
+            cfg: The configuration dictionary containing the Milvus setup
+            modality: The modality to use for the subgraph extraction
+
+        Returns:
+            The prizes of the nodes and edges
+        """
+        logger.log(logging.INFO, "Computing prizes in parallel (hybrid async/sync)")
+
+        # Use existing sync method wrapped in asyncio.to_thread
+        colls = self.prepare_collections(cfg, modality)
+        return await asyncio.to_thread(self.compute_prizes, text_emb, query_emb, colls)
+
     def compute_subgraph_costs(self, edge_index, num_nodes: int, prizes: dict):
         """
         Compute the costs in constructing the subgraph proposed by G-Retriever paper.
@@ -481,11 +715,9 @@ class MultimodalPCSTPruning(NamedTuple):
         logger.log(logging.INFO, "Preparing collections")
         colls = self.prepare_collections(cfg, modality)
 
-        # Load cache edge index
-        logger.log(logging.INFO, "Loading cache edge index")
-        with open(cfg.milvus_db.cache_edge_index_path, "rb") as f:
-            edge_index = pickle.load(f)
-            edge_index = self.loader.py.array(edge_index)
+        # Load edge index directly from Milvus (replaces pickle cache)
+        logger.log(logging.INFO, "Loading edge index from Milvus")
+        edge_index = self.load_edge_index(cfg)
 
         # Assert the topk and topk_e values for subgraph retrieval
         assert self.topk > 0, "topk must be greater than or equal to 0"

aiagents4pharma/talk2scholars/configs/app/frontend/default.yaml

@@ -1,36 +1,72 @@
+_target_: app.frontend.streamlit_app_talk2scholars
+default_user: "talk2scholars_user"
+
+# File upload configuration
+upload_data_dir: "../files"
+pdf_allowed_file_types:
+  - "pdf"
+
+# OpenAI configuration - can use custom base_url for enterprise/Azure deployments
+openai_api_key: ${oc.env:OPENAI_API_KEY}
+openai_base_url: ${oc.env:OPENAI_BASE_URL,null} # Optional: custom OpenAI endpoint
+openai_llms:
+  - "OpenAI/gpt-4o-mini"
+openai_embeddings:
+  - "text-embedding-ada-002"
+  - "text-embedding-3-small"
+# Rate limiting and retry configuration
+llm_max_retries: 5  # Number of retries on rate limit or transient errors
+llm_timeout: 60     # Timeout in seconds for LLM requests
+embedding_max_retries: 3  # Number of retries for embedding requests
+embedding_timeout: 30     # Timeout in seconds for embedding requests
+
+# NVIDIA configuration
+nvidia_api_key: ${oc.env:NVIDIA_API_KEY}
+nvidia_llms:
+  - "NVIDIA/llama-3.3-70b-instruct"
+  - "NVIDIA/llama-3.1-405b-instruct"
+  - "NVIDIA/llama-3.1-70b-instruct"
+nvidia_embeddings:
+  - "NVIDIA/llama-3.2-nv-embedqa-1b-v2"
+
+# Azure OpenAI configuration
+azure_openai_endpoint: ${oc.env:AZURE_OPENAI_ENDPOINT,null} # Azure OpenAI endpoint
+azure_openai_deployment: ${oc.env:AZURE_OPENAI_DEPLOYMENT,null} # Azure deployment name
+azure_openai_api_version: ${oc.env:AZURE_OPENAI_API_VERSION,"2024-02-01"} # Azure API version
+azure_openai_model_name: ${oc.env:AZURE_OPENAI_MODEL_NAME,null} # Model name for analytics
+azure_openai_model_version: ${oc.env:AZURE_OPENAI_MODEL_VERSION,null} # Model version
+# Azure AD authentication (uses AZURE_CLIENT_ID, AZURE_TENANT_ID, AZURE_CLIENT_SECRET)
+azure_client_id: ${oc.env:AZURE_CLIENT_ID,null}
+azure_tenant_id: ${oc.env:AZURE_TENANT_ID,null}
+azure_client_secret: ${oc.env:AZURE_CLIENT_SECRET,null}
+azure_openai_llms:
+  - "Azure/gpt-4o-mini" # Will map to Azure deployment
+azure_openai_embeddings:
+  - "Azure/text-embedding-ada-002"
+
+# Ollama configuration (for local deployment)
+ollama_llms:
+  - "Ollama/llama3.1:8b"
+ollama_embeddings:
+  - "nomic-embed-text"
+
+# Default models
+default_llm_provider: "openai"
+default_embedding_model: "openai"
+
+# App settings
+temperature: 0.1
+streaming: False
+
+# Logo configuration
+logo_paths:
+  container: "/app/docs/assets/VPE.png"
+  local: "docs/assets/VPE.png"
+  relative: "../../docs/assets/VPE.png"
+logo_link: "https://github.com/VirtualPatientEngine"
+
 # Page configuration
 page:
   title: "Talk2Scholars"
   icon: "🤖"
   layout: "wide"
-
-# Available LLM models
-llms:
-  available_models:
-    - "OpenAI/gpt-4o-mini"
-    - "NVIDIA/llama-3.3-70b-instruct"
-# # Chat UI configuration
-# chat:
-#   assistant_avatar: "🤖"
-#   user_avatar: "👩🏻‍💻"
-#   input_placeholder: "Say something ..."
-#   spinner_text: "Fetching response ..."
-
-api_keys:
-  openai_key: "OPENAI_API_KEY"
-  nvidia_key: "NVIDIA_API_KEY"
-# # Feedback configuration
-# feedback:
-#   type: "thumbs"
-#   text_label: "[Optional] Please provide an explanation"
-#   success_message: "Your feedback is on its way to the developers. Thank you!"
-#   success_icon: "🚀"
-
-# # Layout configuration
-# layout:
-#   column_ratio: [3, 7] # Ratio for main_col1 and main_col2
-#   chat_container_height: 575
-#   sidebar_container_height: 500
-#
-# # Project name prefix
-# project_name_prefix: "Talk2Scholars-"

{aiagents4pharma-1.45.1.dist-info → aiagents4pharma-1.46.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aiagents4pharma
-Version: 1.45.1
+Version: 1.46.1
 Summary: AI Agents for drug discovery, drug development, and other pharmaceutical R&D.
 License-File: LICENSE
 Classifier: License :: OSI Approved :: MIT License
@@ -8,6 +8,7 @@ Classifier: Operating System :: OS Independent
 Classifier: Programming Language :: Python :: 3
 Requires-Python: >=3.12
 Requires-Dist: anndata==0.11.3
+Requires-Dist: azure-identity==1.24.0
 Requires-Dist: beautifulsoup4==4.13.4
 Requires-Dist: cell2sentence==1.1.0
 Requires-Dist: cloudscraper==1.2.71
@@ -192,6 +193,14 @@ export LANGCHAIN_API_KEY=...        # Optional for all agents
 
 4. **Launch the app:**
 
+> System Dependency: libmagic (for secure uploads)
+> For accurate file MIME-type detection used by our secure upload validation, install the libmagic system library. This is recommended across all providers (OpenAI, Azure OpenAI, NVIDIA) because it runs locally in the Streamlit apps.
+>
+> - Linux (Debian/Ubuntu): `sudo apt-get install libmagic1`
+> - macOS (Homebrew): `brew install libmagic`
+> - Windows: Use the `python-magic`/`python-magic-bin` package; libmagic is bundled
+>   If libmagic is not available, the apps fall back to extension-based detection. For best security, keep libmagic installed.
+
 **Option A: Using UV (recommended)**
 
 ```sh

{aiagents4pharma-1.45.1.dist-info → aiagents4pharma-1.46.1.dist-info}/RECORD

@@ -7,9 +7,12 @@ aiagents4pharma/talk2aiagents4pharma/install.md,sha256=YYzY1vIEA6RrVUvNuv-h_YTar
 aiagents4pharma/talk2aiagents4pharma/agents/__init__.py,sha256=NpNI6Vr9XIr5m0ZaO32c6NEUTDOZvJUqd8gKzNZhcSw,130
 aiagents4pharma/talk2aiagents4pharma/agents/main_agent.py,sha256=1nRIhj3huv9eVT7v3nhSvx1dDOEEaiApPi7AMRDviXE,2750
 aiagents4pharma/talk2aiagents4pharma/configs/__init__.py,sha256=hwLAR-uhZGEbD5R7mp4kiltSvxuKkG6-_ac17sF-4xU,68
-aiagents4pharma/talk2aiagents4pharma/configs/config.yaml,sha256=Y7mZzvcaJA3vt4Pcf8dBG4WX0hX4unoRSH72lwhDIeM,52
+aiagents4pharma/talk2aiagents4pharma/configs/config.yaml,sha256=bJpU9_L88dOSqI6-BL3-X5tlR9eznXXNgXCi0qzPdEs,78
 aiagents4pharma/talk2aiagents4pharma/configs/agents/__init__.py,sha256=8a9wAMF8zcfKYVZw-lQWsPsQtOPL_HQzu62F-EYekvw,72
 aiagents4pharma/talk2aiagents4pharma/configs/agents/main_agent/default.yaml,sha256=jISv2CazHsK_OuQNgRG2uu3i9i2OeWm1Kxzh-2sPokU,2424
+aiagents4pharma/talk2aiagents4pharma/configs/app/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+aiagents4pharma/talk2aiagents4pharma/configs/app/frontend/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+aiagents4pharma/talk2aiagents4pharma/configs/app/frontend/default.yaml,sha256=BpDv_Bau0F7xDPzMHVYxedKpZQAPdo0rAsRnfH32-U8,3470
 aiagents4pharma/talk2aiagents4pharma/docker-compose/cpu/.env.example,sha256=twsuiEN-FczXfka40_bVoSQXaMDR_3J3ESKCqU4qSkg,585
 aiagents4pharma/talk2aiagents4pharma/docker-compose/cpu/docker-compose.yml,sha256=2I8jyzaFXSGsBMMFOZ4Hz7hAH3A-ELpoYCphkpT_8rA,2631
 aiagents4pharma/talk2aiagents4pharma/docker-compose/gpu/.env.example,sha256=twsuiEN-FczXfka40_bVoSQXaMDR_3J3ESKCqU4qSkg,585
@@ -17,7 +20,7 @@ aiagents4pharma/talk2aiagents4pharma/docker-compose/gpu/docker-compose.yml,sha25
 aiagents4pharma/talk2aiagents4pharma/states/__init__.py,sha256=VuVAFmoH8p2erE-mYiaa0uoqTuFynBcXia5Y8lmjI34,109
 aiagents4pharma/talk2aiagents4pharma/states/state_talk2aiagents4pharma.py,sha256=ahquXi8hg6Bk1ttskl9QaYdHq4rDWa_bK7hu81E75M4,468
 aiagents4pharma/talk2aiagents4pharma/tests/__init__.py,sha256=U3PsTiUZaUBD1IZanFGkDIOdFieDVJtGKQ5-woYUo8c,45
-aiagents4pharma/talk2aiagents4pharma/tests/test_main_agent.py,sha256=0yx3Ij95M8EdeephO4LWKpLGqMHtQhm-qLai5GRk-IM,7621
+aiagents4pharma/talk2aiagents4pharma/tests/test_main_agent.py,sha256=oLRYlnkSBiD7trIUf-NZBTJDKX0X0JIK8hhMB9nMu9s,10945
 aiagents4pharma/talk2biomodels/.dockerignore,sha256=-hAM7RzkGbjDeU411-kXOmYzNfl3Z9OlLWvN9zMDAXE,89
 aiagents4pharma/talk2biomodels/Dockerfile,sha256=BXIqnV5HWHtCsS5dzvN-lL6S8wSzDmKawcIuyLscaqw,3537
 aiagents4pharma/talk2biomodels/README.md,sha256=0eGxj7jxi_LrCvX-4I4KrQv-7T2ivo3pqLslG7suaCk,74
@@ -25,15 +28,17 @@ aiagents4pharma/talk2biomodels/__init__.py,sha256=DxARZJu91m4WHW4PBSZvlMb1MCbjvk
 aiagents4pharma/talk2biomodels/install.md,sha256=9YAEeW_vG5hv7WiMnNEzgKQIgVyHnpk1IIWXg_jhLxE,1520
 aiagents4pharma/talk2biomodels/agents/__init__.py,sha256=4wPy6hWRJksX6z8qX1cVjFctZLpsja8JMngKHqn49N4,129
 aiagents4pharma/talk2biomodels/agents/t2b_agent.py,sha256=b_9pjnY3VGPRrzy-LCFrbJCkP1PT3M30NiQq6gbavaA,3267
-aiagents4pharma/talk2biomodels/api/__init__.py,sha256=2fKfvp7tj6fr6mhesQo2flaNpW5cINjzx_o_zUpTzZc,98
-aiagents4pharma/talk2biomodels/api/kegg.py,sha256=znaXYtw6HZe5XoX7wyl0hi1anHd2vzKRgVzt0HZ_TU0,2836
+aiagents4pharma/talk2biomodels/api/__init__.py,sha256=KGok9mCa6RT8whDj3jT3kcpFO1yHxk5vVD8IExsI5Bc,92
 aiagents4pharma/talk2biomodels/api/ols.py,sha256=QSvbsD0V07-w0OU-wPQ4EypXi9bn_xl0NyliZQxRvCU,2173
 aiagents4pharma/talk2biomodels/api/uniprot.py,sha256=jXoyd7BhIQA9JNaGMVPzORpQ5k1Ix9iYYduMv6YG7hw,1147
 aiagents4pharma/talk2biomodels/configs/__init__.py,sha256=WgGZbEtZiVR8EVJeH3iIJ6LxBGFafAjfCftptnTm2Ds,75
-aiagents4pharma/talk2biomodels/configs/config.yaml,sha256=ANNi5gKbHpozGHttG3EfTsW8KFmUmHqH8FgJf9jJ1vU,151
+aiagents4pharma/talk2biomodels/configs/config.yaml,sha256=9I9L99j2DxD3MzNX3Ch-x4Q38nsKzh33SrAqv6ftpnk,177
 aiagents4pharma/talk2biomodels/configs/agents/__init__.py,sha256=6CfgJ15-smSZ4HP4uIlk3QgfVzzpDunBZOuqpuz9CIY,71
 aiagents4pharma/talk2biomodels/configs/agents/t2b_agent/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2biomodels/configs/agents/t2b_agent/default.yaml,sha256=oMC7OK3t1tMI2LLJHJcQQcteOoqO6gR12aKJWyZ08_E,535
+aiagents4pharma/talk2biomodels/configs/app/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+aiagents4pharma/talk2biomodels/configs/app/frontend/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+aiagents4pharma/talk2biomodels/configs/app/frontend/default.yaml,sha256=dAnLyA8pd02Ko9IURh3xeYTmbC4RK9j1kuZQJwUPqWQ,2478
 aiagents4pharma/talk2biomodels/configs/tools/__init__.py,sha256=m3Kvvc0YNyJ2_uiLpBVJmEJvJ0WyoPrdokdR7TKQR5I,106
 aiagents4pharma/talk2biomodels/configs/tools/ask_question/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2biomodels/configs/tools/ask_question/default.yaml,sha256=9AlSgfb6XSFaH3N80BMQ-CcI_x-ZWfF-52QNgpdtlhE,1296
@@ -49,10 +54,10 @@ aiagents4pharma/talk2biomodels/states/state_talk2biomodels.py,sha256=kKtVBfSK92h
 aiagents4pharma/talk2biomodels/tests/BIOMD0000000449_url.xml,sha256=RkWbstfLrT1mAfOtZf7JsBz6poyWg6-5G7H_IdVXEXg,72630
 aiagents4pharma/talk2biomodels/tests/__init__.py,sha256=U3PsTiUZaUBD1IZanFGkDIOdFieDVJtGKQ5-woYUo8c,45
 aiagents4pharma/talk2biomodels/tests/article_on_model_537.pdf,sha256=rfBnG9XSGRZodq-NQsouQQ3dvm4JKcrAqEkoAQJmuDc,470738
-aiagents4pharma/talk2biomodels/tests/test_api.py,sha256=0Gyakd5Yaa0CxUH18040xl-TBrNKGi6aDaIyVFF1d9A,1724
+aiagents4pharma/talk2biomodels/tests/test_api.py,sha256=rfH2_K2pC5fm3DcyKJSHOBcN6Mzz3D37oJEpFCE1Jek,970
 aiagents4pharma/talk2biomodels/tests/test_ask_question.py,sha256=lMrFI5URqUEz0Cv3WOoj8vhR_ciwDOgoJlqh1dOdaik,1523
 aiagents4pharma/talk2biomodels/tests/test_basico_model.py,sha256=F_1AbFRcXIDahQKcS-Nq0v45bEozLUOYx_oyyFN0hxw,2429
-aiagents4pharma/talk2biomodels/tests/test_get_annotation.py,sha256=T5RvNfiyz9Z49LBo6FkOOaD7f3vSalklF6Oi1Qj3-Kk,8179
+aiagents4pharma/talk2biomodels/tests/test_get_annotation.py,sha256=L4fTQaY2rnkfHwfPSC6GdXD2T3NGTIqNoF2dRklBIQA,8173
 aiagents4pharma/talk2biomodels/tests/test_getmodelinfo.py,sha256=PlVDIQZaZw8a94_Muw5GMa2Kop6foEBGhp3x-412cQA,3258
 aiagents4pharma/talk2biomodels/tests/test_integration.py,sha256=t8jR45pX7hKBGOjXySQ-XG5wJpTt41jGba74T8zuIdI,4838
 aiagents4pharma/talk2biomodels/tests/test_load_biomodel.py,sha256=8nVSDa8_z85dyvxa8aYGQR0YGZDtpzLF5HhBmifCk6w,895
@@ -65,7 +70,7 @@ aiagents4pharma/talk2biomodels/tests/test_sys_bio_model.py,sha256=poMxOsKhg8USnp
 aiagents4pharma/talk2biomodels/tools/__init__.py,sha256=j6wTW09BOwFMzHERfJbsajctDsNxJDtWJHUE3FmTu-A,279
 aiagents4pharma/talk2biomodels/tools/ask_question.py,sha256=IbolM6zbYKHd_UCfLMa8bawt9fJH59cCUtkLB_wtxKI,4495
 aiagents4pharma/talk2biomodels/tools/custom_plotter.py,sha256=dk5HUmPwSTIRp2sbd8Q8__fwSE8m13UseonvcpyDs00,6636
-aiagents4pharma/talk2biomodels/tools/get_annotation.py,sha256=spP_EVDlkbg7DfpzZw4YyIIYAsiax5EoJxB0f0gjVbM,13312
+aiagents4pharma/talk2biomodels/tools/get_annotation.py,sha256=oHERHdY4KinQFg9udufEgJP3tE3x0gtoWWy4Kna9H78,12854
 aiagents4pharma/talk2biomodels/tools/get_modelinfo.py,sha256=mVAqO2TRWIlD93ZsMggs2N3629sxHZWnOisW7r_yBU0,5987
 aiagents4pharma/talk2biomodels/tools/load_arguments.py,sha256=LZQNkAikXhG0AKRnfLUqqpa5hNAyVGSwPQ4_nBu-DSw,4009
 aiagents4pharma/talk2biomodels/tools/load_biomodel.py,sha256=025-E5qo2uiJVvHIhyeDh1tfmXTeIguSgS0KIY0LiyY,1208
@@ -90,27 +95,29 @@ aiagents4pharma/talk2knowledgegraphs/.dockerignore,sha256=-hAM7RzkGbjDeU411-kXOm
 aiagents4pharma/talk2knowledgegraphs/Dockerfile,sha256=qGy6I4oBvQonpDEANQaCW-5JMtHPdd2HKc-JJgoCyYQ,3384
 aiagents4pharma/talk2knowledgegraphs/README.md,sha256=0eGxj7jxi_LrCvX-4I4KrQv-7T2ivo3pqLslG7suaCk,74
 aiagents4pharma/talk2knowledgegraphs/__init__.py,sha256=ZztaRzRlovSXtVX3i9Rvf84ivIjPn8RMPiYRkbkEJ0E,114
-aiagents4pharma/talk2knowledgegraphs/entrypoint.sh,sha256=O9iXAmfj3vzdp5CxzpRUDb4FmXumDsjZfYK2S7RkAMY,6098
+aiagents4pharma/talk2knowledgegraphs/entrypoint.sh,sha256=EK_jGau1VuW1uTmFWZcKhLMK9VanC5l3q9axF4ZYgmI,5758
 aiagents4pharma/talk2knowledgegraphs/install.md,sha256=6eZe9czZ8nRrsOHbRmn9WaAk9xgl2kxDZac4Exs1WqU,6022
-aiagents4pharma/talk2knowledgegraphs/milvus_data_dump.py,sha256=0eMX0lTFfATR6wZE-sT9n-heXdH7RbD22f41DPj8uzE,35577
+aiagents4pharma/talk2knowledgegraphs/milvus_data_dump.py,sha256=nLChVpn9KwJ2F5zgtb51zG2s81vOzc6o5Ut-xhkjpJI,35550
 aiagents4pharma/talk2knowledgegraphs/agents/__init__.py,sha256=ugUvVYEdjbZ3y_dogfF5hpQ3lFPFrAvLSydlcpbkGo0,93
 aiagents4pharma/talk2knowledgegraphs/agents/t2kg_agent.py,sha256=GDeSjJNhAqQWagZOxAWUKqDhzUohHViSsu444W9SzRQ,3240
 aiagents4pharma/talk2knowledgegraphs/configs/__init__.py,sha256=H-yhTbJ_RXBLe3XSto5x6FmVrgbi7y1WKEfiwmKzLAk,87
-aiagents4pharma/talk2knowledgegraphs/configs/config.yaml,sha256=FjloiqLeDx73kM2DVhCbLEuj3N1th2idFNy6p3CJkT4,418
+aiagents4pharma/talk2knowledgegraphs/configs/config.yaml,sha256=W-yjtavcqGE3hZdGwG5o1qDnvjEChQ-BjuU8Y2Salk8,453
 aiagents4pharma/talk2knowledgegraphs/configs/agents/t2kg_agent/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2knowledgegraphs/configs/agents/t2kg_agent/default.yaml,sha256=4NcNS8f9A6e9_8DcdYjC_Q5sjgmGA6GZFAwbsmEcPiA,4824
 aiagents4pharma/talk2knowledgegraphs/configs/app/__init__.py,sha256=JoSZV6N669kGMv5zLDszwf0ZjcRHx9TJfIqGhIIdPXE,70
 aiagents4pharma/talk2knowledgegraphs/configs/app/frontend/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
-aiagents4pharma/talk2knowledgegraphs/configs/app/frontend/default.yaml,sha256=eW-pJXgL6uHByOIggMg-N76HHAvP0XRcKKPSzGrAZ_I,2301
+aiagents4pharma/talk2knowledgegraphs/configs/app/frontend/default.yaml,sha256=c5ufnM3rU5affg1s-JEkQ_8LUpnc6EwjIdLRNvQMBOA,3245
 aiagents4pharma/talk2knowledgegraphs/configs/tools/__init__.py,sha256=lGyiRtLtwBn0pTa7AV2ZzjyVck2lVrVO3D9FQF22s7w,125
 aiagents4pharma/talk2knowledgegraphs/configs/tools/graphrag_reasoning/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2knowledgegraphs/configs/tools/graphrag_reasoning/default.yaml,sha256=bSEqSz70hd4gnCSNZ0vlmtF-X1NO4BaXo3CA6Kdcz1o,1038
 aiagents4pharma/talk2knowledgegraphs/configs/tools/multimodal_subgraph_extraction/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-aiagents4pharma/talk2knowledgegraphs/configs/tools/multimodal_subgraph_extraction/default.yaml,sha256=QEtxF7Fj1DYFEw1qS-JXAptbIgNHc-dnBV6aic0alkk,1330
+aiagents4pharma/talk2knowledgegraphs/configs/tools/multimodal_subgraph_extraction/default.yaml,sha256=pqR9bBQM3sW-V4bBlP7LFB2GpO2RhacHrXzGXmUL6fk,904
 aiagents4pharma/talk2knowledgegraphs/configs/tools/subgraph_extraction/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2knowledgegraphs/configs/tools/subgraph_extraction/default.yaml,sha256=r5T29xiWnBK2C1vfsSlNHDlB0DfTvReFbzQgZqqB5l4,1470
 aiagents4pharma/talk2knowledgegraphs/configs/tools/subgraph_summarization/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2knowledgegraphs/configs/tools/subgraph_summarization/default.yaml,sha256=kiHerHXsGuohu_FLq58nXkY3bwlSFMw3kGqtQ68LMfA,562
+aiagents4pharma/talk2knowledgegraphs/configs/utils/database/milvus/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
+aiagents4pharma/talk2knowledgegraphs/configs/utils/database/milvus/default.yaml,sha256=4J6cIguIy4Z794UM9EHwzwzwrwWv7qvsKTN8c5QucLc,2394
 aiagents4pharma/talk2knowledgegraphs/configs/utils/enrichments/ols_terms/default.yaml,sha256=7Fu03IkClglxbAzwTUdy-z-Tv6MZuo4oIdBYpQ7TPjI,107
 aiagents4pharma/talk2knowledgegraphs/configs/utils/enrichments/reactome_pathways/default.yaml,sha256=gyYLJREMO1jDex8-0CcCPfsuD5KguY6fKDmDCL_S4Uc,125
 aiagents4pharma/talk2knowledgegraphs/configs/utils/enrichments/uniprot_proteins/default.yaml,sha256=JlmawYim5ECgAKCZta3UrPD2YIjkD2d5TvLmR5V6wXM,184
@@ -127,16 +134,17 @@ aiagents4pharma/talk2knowledgegraphs/docker-compose/gpu/docker-compose.yml,sha25
 aiagents4pharma/talk2knowledgegraphs/states/__init__.py,sha256=9X3zHxvtAJFQ0s0VLZ_-iBn4rMVfZWk5CQiWEKJkr0c,109
 aiagents4pharma/talk2knowledgegraphs/states/state_talk2knowledgegraphs.py,sha256=QECNXd8IWDh3WlMvePcd2T6G5XqjOI9EkZdWmICcCT0,1088
 aiagents4pharma/talk2knowledgegraphs/tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-aiagents4pharma/talk2knowledgegraphs/tests/test_agents_t2kg_agent.py,sha256=IwoawiVrVR0oixK66rwwOGMpqyLnML5ds7J36f5YuS4,6749
+aiagents4pharma/talk2knowledgegraphs/tests/test_agents_t2kg_agent.py,sha256=1Y-W2uH3goFTKKB_MeVtrfdMwPy-WGnieawF4KQIjkM,11132
 aiagents4pharma/talk2knowledgegraphs/tests/test_datasets_biobridge_primekg.py,sha256=AHUJ7StFYTBa8l4n_zYgyT0C9_G8j6NY3qHzLBhtkDg,9410
 aiagents4pharma/talk2knowledgegraphs/tests/test_datasets_dataset.py,sha256=nBLMI1TzulCl1pj-svy_mstomSQ6BRumTiXV8lRV40o,570
 aiagents4pharma/talk2knowledgegraphs/tests/test_datasets_primekg.py,sha256=csAR86I2RAxYG4Y0wpz-XLSBQf52x5gnhivvW_S17do,2329
 aiagents4pharma/talk2knowledgegraphs/tests/test_datasets_starkqa_primekg.py,sha256=30hNjv2i8G7VmbU_LLTuwpAkYFuQQdhJguLX_JJ-0MA,4391
 aiagents4pharma/talk2knowledgegraphs/tests/test_tools_graphrag_reasoning.py,sha256=Z2PIXBb_vV91BlC9x6a3YvQ8lvCtXnn1T28A8iwzKwM,11940
-aiagents4pharma/talk2knowledgegraphs/tests/test_tools_milvus_multimodal_subgraph_extraction.py,sha256=eFbFjqwxGkgXGar54JpC0Km2rdP7xWL8z4NEw2h9RnE,29208
+aiagents4pharma/talk2knowledgegraphs/tests/test_tools_milvus_multimodal_subgraph_extraction.py,sha256=U_b8tz7McoX2Vd0rZRkkl_UOmMtiiROvdZZGo59rMjY,52689
 aiagents4pharma/talk2knowledgegraphs/tests/test_tools_multimodal_subgraph_extraction.py,sha256=fIfs1atTZAYdJArPO1SUYmikYXn8pBIzQ_f3qrAo4Yc,5851
 aiagents4pharma/talk2knowledgegraphs/tests/test_tools_subgraph_extraction.py,sha256=zdhBkKj7Ygo-X5DSWSrLa-Xz-z8Va00NNekJXJjYl5Q,5428
 aiagents4pharma/talk2knowledgegraphs/tests/test_tools_subgraph_summarization.py,sha256=dt_wi3bQUDd57HsoVl4_dIOy-EP_KZvvHxJcrdMvt_4,8907
+aiagents4pharma/talk2knowledgegraphs/tests/test_utils_database_milvus_connection_manager.py,sha256=BaHFc5g7rLI3cMaHcgvTA8aI0lGmxZIUzLYyKN_QQtk,25028
 aiagents4pharma/talk2knowledgegraphs/tests/test_utils_embeddings_embeddings.py,sha256=LXrztJ9ICAph-txBbXB2Dw1hPfSI2fYAdKq8CjqaOhA,1521
 aiagents4pharma/talk2knowledgegraphs/tests/test_utils_embeddings_huggingface.py,sha256=C36Dn4ri4qABCY3Wh_amPxxKNTWVp5Xed9QsAAkNLhk,1628
 aiagents4pharma/talk2knowledgegraphs/tests/test_utils_embeddings_nim_molmim.py,sha256=NZF5sP64Q-jV8E-gxtq1NZX6kq3NJcgV74jkcnyM7Jg,2055
@@ -148,19 +156,21 @@ aiagents4pharma/talk2knowledgegraphs/tests/test_utils_enrichments_ols.py,sha256=
 aiagents4pharma/talk2knowledgegraphs/tests/test_utils_enrichments_pubchem.py,sha256=USHsjpfHxAktxjc5Q8INZ7qQJDWn_XFQYsS6mpeTLv0,1704
 aiagents4pharma/talk2knowledgegraphs/tests/test_utils_enrichments_reactome.py,sha256=eseSw3flQLQAmXLS2R-bobN3QGEs_i6iId8l4bLEitE,1658
 aiagents4pharma/talk2knowledgegraphs/tests/test_utils_enrichments_uniprot.py,sha256=8JqptVypftx4IrObzxGPDjQ1C7I24f3zUUx69D8svAo,1615
-aiagents4pharma/talk2knowledgegraphs/tests/test_utils_extractions_milvus_multimodal_pcst.py,sha256=Oo6IvPB9L9lAmdxc4UqFJVX-lYYdFKYv3lNaHrjkvS8,21515
+aiagents4pharma/talk2knowledgegraphs/tests/test_utils_extractions_milvus_multimodal_pcst.py,sha256=vBc1nlAawIyp3NNp9P4JO1cu3uKIA_OiajXBNro-T_s,28591
 aiagents4pharma/talk2knowledgegraphs/tests/test_utils_kg_utils.py,sha256=RDheTkkRxnGCxIoaTmTRKOQA2ewJyOOlKtgxdABgSSA,2397
 aiagents4pharma/talk2knowledgegraphs/tests/test_utils_pubchem_utils.py,sha256=80PhbFXD287LeCwGpXGtPal5vkjqKKCACefR_BPdfbM,3967
 aiagents4pharma/talk2knowledgegraphs/tools/__init__.py,sha256=unrqqDUAmfTpgiJSV65Pag9FWHpnf3eEsE8Cwh59NWI,242
 aiagents4pharma/talk2knowledgegraphs/tools/graphrag_reasoning.py,sha256=cCPBH1tKs9MjX1q9v6BXi-dInz_gxKwMyIVA-XdKocg,5251
 aiagents4pharma/talk2knowledgegraphs/tools/load_arguments.py,sha256=zhmsRp-8vjB5rRekqTA07d3yb-42HWqng9dDMkvK6hM,623
-aiagents4pharma/talk2knowledgegraphs/tools/milvus_multimodal_subgraph_extraction.py,sha256=QIgPJUTPP-AR_bytfmp3WLW8Gqwe6VQn5cYItWsSBE8,23311
+aiagents4pharma/talk2knowledgegraphs/tools/milvus_multimodal_subgraph_extraction.py,sha256=chyHEOlbzLwPd8IXwgA4B2YMWbJAyBikBkf2-x_XH1E,39730
 aiagents4pharma/talk2knowledgegraphs/tools/multimodal_subgraph_extraction.py,sha256=cAsRkBklFxitBDNvhOIqmqd0ZTjtRYKsgb-rySC0PTk,14774
 aiagents4pharma/talk2knowledgegraphs/tools/subgraph_extraction.py,sha256=DSUucfzKf3LoOo1v9snp6-Zk1vTGB9jeH-hshatd0PY,11161
 aiagents4pharma/talk2knowledgegraphs/tools/subgraph_summarization.py,sha256=mXuKTahLXXFYFMS-0HkmiP7o6MSLjE_REEEsxPwCF7c,4372
 aiagents4pharma/talk2knowledgegraphs/utils/__init__.py,sha256=OwsMjDLRsVdHm6jS_oKbUP9tUKOwGt2yOjz5EgWCv_M,144
 aiagents4pharma/talk2knowledgegraphs/utils/kg_utils.py,sha256=IeCekDG__hjvIBXk4geLLBzlrLJukVg2y8IZfXTosQ0,2188
 aiagents4pharma/talk2knowledgegraphs/utils/pubchem_utils.py,sha256=PWjLdXbkdnQig95TkR4zqrhEULbQoGFIQMFK7HIMX8U,3378
+aiagents4pharma/talk2knowledgegraphs/utils/database/__init__.py,sha256=KwsAQO484Ado7GUn0kvymVn5_71h0oxcAkbMxOf6MZQ,154
+aiagents4pharma/talk2knowledgegraphs/utils/database/milvus_connection_manager.py,sha256=IW4FoTAYQvhln4_M8jcRfpXjb-WqaMw65cRe5vvqNOw,21083
 aiagents4pharma/talk2knowledgegraphs/utils/embeddings/__init__.py,sha256=XWFU0_juBq-8oEP8oAgyGLMpowZ1VdgibwjYdc0I3y4,148
 aiagents4pharma/talk2knowledgegraphs/utils/embeddings/embeddings.py,sha256=O11r9uKDKBjsgq9gGeKxoUR-HoBSVX6D16wci7eYW8s,2527
 aiagents4pharma/talk2knowledgegraphs/utils/embeddings/huggingface.py,sha256=rDhksY0g8HfpWiJO1CeKGdvZJ-stFrZ9Q88CXlfXJCU,3724
@@ -175,7 +185,7 @@ aiagents4pharma/talk2knowledgegraphs/utils/enrichments/pubchem_strings.py,sha256
 aiagents4pharma/talk2knowledgegraphs/utils/enrichments/reactome_pathways.py,sha256=8iyrT1p01GJpi7k7FCuwZBqn-hyueRWYxPqabM5U80Q,1990
 aiagents4pharma/talk2knowledgegraphs/utils/enrichments/uniprot_proteins.py,sha256=0mTFX5sVgBZHkEuf057lCFNciTBapn9CqPi2-YublUg,3013
 aiagents4pharma/talk2knowledgegraphs/utils/extractions/__init__.py,sha256=kgioLxc3ZPY46aXe36JMTo2OODchsLnzSzClpz-nwaU,128
-aiagents4pharma/talk2knowledgegraphs/utils/extractions/milvus_multimodal_pcst.py,sha256=KfljjljjBetfGdK05oe6eXTBNYu_uOvsqMIWbHkMIig,21312
+aiagents4pharma/talk2knowledgegraphs/utils/extractions/milvus_multimodal_pcst.py,sha256=QZ_hJpeqre1YKKGTRr1fJpY1cb1y1Ldy8UcdxaTYyLU,30570
 aiagents4pharma/talk2knowledgegraphs/utils/extractions/multimodal_pcst.py,sha256=GD2lsq3yiDf6n-pT9HbJiRd6o0DBgnv6lwq9RuC-zuU,12356
 aiagents4pharma/talk2knowledgegraphs/utils/extractions/pcst.py,sha256=__IB_hPEQ-GyHAsz6FOmYUlXXmcB0T6J9wvMS2UNiko,9324
 aiagents4pharma/talk2scholars/.dockerignore,sha256=-hAM7RzkGbjDeU411-kXOmYzNfl3Z9OlLWvN9zMDAXE,89
@@ -205,7 +215,7 @@ aiagents4pharma/talk2scholars/configs/agents/talk2scholars/zotero_agent/__init__
 aiagents4pharma/talk2scholars/configs/agents/talk2scholars/zotero_agent/default.yaml,sha256=ymVGlho7BZTJlMxT3xMRUsMwd9qNT6ZR7KgB57DgvaY,1002
 aiagents4pharma/talk2scholars/configs/app/__init__.py,sha256=tXpOW3R4eAfNoqvoaHfabSG-DcMHmUGSTg_4zH_vlgw,94
 aiagents4pharma/talk2scholars/configs/app/frontend/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
-aiagents4pharma/talk2scholars/configs/app/frontend/default.yaml,sha256=A6nYjrgzEyRv5JYsGN7oqNX4-tufMBZ6mg-A7bMX6V4,906
+aiagents4pharma/talk2scholars/configs/app/frontend/default.yaml,sha256=Z2oZBqGkJBC_8M5jXCZwrBcJwbllRd5YnCiUtKZdOig,2445
 aiagents4pharma/talk2scholars/configs/tools/__init__.py,sha256=TTLziqMoxBLU5yZpecgptfySKQJH82tc1K1G6iKzmKg,366
 aiagents4pharma/talk2scholars/configs/tools/multi_paper_recommendation/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2scholars/configs/tools/multi_paper_recommendation/default.yaml,sha256=comNgL9hRpH--IWuEsrN6hV5WdrJmh-ZsRh7hbryVhg,631
@@ -318,7 +328,7 @@ aiagents4pharma/talk2scholars/tools/zotero/utils/review_helper.py,sha256=-q-UuzP
 aiagents4pharma/talk2scholars/tools/zotero/utils/write_helper.py,sha256=K1EatPfC-riGyFmkOAS3ReNBaGPY-znne1KqOnFahkI,7339
 aiagents4pharma/talk2scholars/tools/zotero/utils/zotero_path.py,sha256=sKkfJu3u4LKSZjfoQRfeqz26IESHRwBtcSDzLMLlJMo,6311
 aiagents4pharma/talk2scholars/tools/zotero/utils/zotero_pdf_downloader.py,sha256=DBrF5IiF7VRP58hUK8T9LST3lQWLFixLUfnpMSTccoQ,4614
-aiagents4pharma-1.45.1.dist-info/METADATA,sha256=JFqD43WM9t6Qp9zi5odBVfXPhDC5WCsmY_pe4RkbGcU,16285
-aiagents4pharma-1.45.1.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-aiagents4pharma-1.45.1.dist-info/licenses/LICENSE,sha256=IcIbyB1Hyk5ZDah03VNQvJkbNk2hkBCDqQ8qtnCvB4Q,1077
-aiagents4pharma-1.45.1.dist-info/RECORD,,
+aiagents4pharma-1.46.1.dist-info/METADATA,sha256=z8dEOeNUk6HxhM5WQfaFKsnkdrUo-OAbJwf3G8-4Yro,16928
+aiagents4pharma-1.46.1.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+aiagents4pharma-1.46.1.dist-info/licenses/LICENSE,sha256=IcIbyB1Hyk5ZDah03VNQvJkbNk2hkBCDqQ8qtnCvB4Q,1077
+aiagents4pharma-1.46.1.dist-info/RECORD,,