usecortex-ai 0.4.0__py3-none-any.whl → 0.5.1__py3-none-any.whl

usecortex_ai/search/raw_client.py

@@ -40,31 +40,18 @@ class RawSearchClient:
         tenant_id: str,
         query: str,
         sub_tenant_id: typing.Optional[str] = OMIT,
-        max_chunks: typing.Optional[int] = OMIT,
+        max_results: typing.Optional[int] = OMIT,
         mode: typing.Optional[RetrieveMode] = OMIT,
         alpha: typing.Optional[Alpha] = OMIT,
         recency_bias: typing.Optional[float] = OMIT,
-        num_related_chunks: typing.Optional[int] = OMIT,
-        personalise_search: typing.Optional[bool] = OMIT,
+        personalise_recall: typing.Optional[bool] = OMIT,
         graph_context: typing.Optional[bool] = OMIT,
-        extra_context: typing.Optional[str] = OMIT,
+        additional_context: typing.Optional[str] = OMIT,
         search_mode: typing.Optional[SearchMode] = OMIT,
+        metadata_filters: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[RetrievalResult]:
         """
-        Search for relevant content within your indexed sources or user memories.
-
-            Results are ranked by relevance and can be customized with parameters like
-            result limits, alpha weighting, and recency preferences.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories (uses inferred content)
-
-            Use `mode` to control retrieval quality:
-            - "fast" (default): Single query, faster response
-            - "accurate": Multi-query generation with reranking, higher quality
-
         Parameters
         ----------
         tenant_id : str
@@ -76,7 +63,7 @@ class RawSearchClient:
         sub_tenant_id : typing.Optional[str]
             Optional sub-tenant identifier used to organize data within a tenant. If omitted, the default sub-tenant created during tenant setup will be used.
 
-        max_chunks : typing.Optional[int]
+        max_results : typing.Optional[int]
             Maximum number of results to return
 
         mode : typing.Optional[RetrieveMode]
@@ -88,21 +75,21 @@ class RawSearchClient:
         recency_bias : typing.Optional[float]
             Preference for newer content (0.0 = no bias, 1.0 =             strong recency preference)
 
-        num_related_chunks : typing.Optional[int]
-            Number of related content chunks to include
-
-        personalise_search : typing.Optional[bool]
+        personalise_recall : typing.Optional[bool]
             Enable personalized search results based on user preferences
 
         graph_context : typing.Optional[bool]
             Enable graph context for search results
 
-        extra_context : typing.Optional[str]
+        additional_context : typing.Optional[str]
             Additional context provided by the user to guide retrieval
 
         search_mode : typing.Optional[SearchMode]
             What to search: 'sources' for documents or 'memories' for user memories
 
+        metadata_filters : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
+            Optional key-value pairs to filter search results by tenant metadata fields. Keys must match fields defined in tenant_metadata_schema during tenant creation. Supports exact match filtering on indexed metadata fields. Example: {'category': 'engineering', 'priority': 'high'}
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -112,21 +99,21 @@ class RawSearchClient:
             Successful Response
         """
         _response = self._client_wrapper.httpx_client.request(
-            "search/hybrid-search",
+            "recall/full_recall",
             method="POST",
             json={
                 "tenant_id": tenant_id,
                 "sub_tenant_id": sub_tenant_id,
                 "query": query,
-                "max_chunks": max_chunks,
+                "max_results": max_results,
                 "mode": mode,
                 "alpha": convert_and_respect_annotation_metadata(object_=alpha, annotation=Alpha, direction="write"),
                 "recency_bias": recency_bias,
-                "num_related_chunks": num_related_chunks,
-                "personalise_search": personalise_search,
+                "personalise_recall": personalise_recall,
                 "graph_context": graph_context,
-                "extra_context": extra_context,
+                "additional_context": additional_context,
                 "search_mode": search_mode,
+                "metadata_filters": metadata_filters,
             },
             headers={
                 "content-type": "application/json",
@@ -244,19 +231,11 @@ class RawSearchClient:
         query: str,
         sub_tenant_id: typing.Optional[str] = OMIT,
         operator: typing.Optional[Bm25OperatorType] = OMIT,
-        max_chunks: typing.Optional[int] = OMIT,
+        max_results: typing.Optional[int] = OMIT,
         search_mode: typing.Optional[SearchMode] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[RetrievalResult]:
         """
-        Perform full text search for exact matches within your indexed sources or memories.
-            Choose between 'OR' and 'AND' operators to control how search terms are combined
-            for precise text matching.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories
-
         Parameters
         ----------
         tenant_id : str
@@ -271,7 +250,7 @@ class RawSearchClient:
         operator : typing.Optional[Bm25OperatorType]
             How to combine search terms (OR or AND)
 
-        max_chunks : typing.Optional[int]
+        max_results : typing.Optional[int]
             Maximum number of results to return
 
         search_mode : typing.Optional[SearchMode]
@@ -286,14 +265,14 @@ class RawSearchClient:
             Successful Response
         """
         _response = self._client_wrapper.httpx_client.request(
-            "search/full-text-search",
+            "recall/boolean_recall",
             method="POST",
             json={
                 "tenant_id": tenant_id,
                 "sub_tenant_id": sub_tenant_id,
                 "query": query,
                 "operator": operator,
-                "max_chunks": max_chunks,
+                "max_results": max_results,
                 "search_mode": search_mode,
             },
             headers={
@@ -424,19 +403,6 @@ class RawSearchClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[QnASearchResponse]:
         """
-        Ask a question and get an AI-generated answer based on your indexed sources or memories.
-
-            The response includes both the AI answer and the source chunks used to generate it,
-            enabling full transparency and citation capabilities.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories
-
-            Use `mode` to control retrieval quality:
-            - "fast" (default): Single query, faster response
-            - "accurate": Multi-query generation with reranking, higher quality
-
         Parameters
         ----------
         tenant_id : str
@@ -487,7 +453,7 @@ class RawSearchClient:
             Successful Response
         """
         _response = self._client_wrapper.httpx_client.request(
-            "search/qna",
+            "recall/qna",
             method="POST",
             json={
                 "tenant_id": tenant_id,
@@ -624,31 +590,18 @@ class AsyncRawSearchClient:
         tenant_id: str,
         query: str,
         sub_tenant_id: typing.Optional[str] = OMIT,
-        max_chunks: typing.Optional[int] = OMIT,
+        max_results: typing.Optional[int] = OMIT,
         mode: typing.Optional[RetrieveMode] = OMIT,
         alpha: typing.Optional[Alpha] = OMIT,
         recency_bias: typing.Optional[float] = OMIT,
-        num_related_chunks: typing.Optional[int] = OMIT,
-        personalise_search: typing.Optional[bool] = OMIT,
+        personalise_recall: typing.Optional[bool] = OMIT,
         graph_context: typing.Optional[bool] = OMIT,
-        extra_context: typing.Optional[str] = OMIT,
+        additional_context: typing.Optional[str] = OMIT,
         search_mode: typing.Optional[SearchMode] = OMIT,
+        metadata_filters: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[RetrievalResult]:
         """
-        Search for relevant content within your indexed sources or user memories.
-
-            Results are ranked by relevance and can be customized with parameters like
-            result limits, alpha weighting, and recency preferences.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories (uses inferred content)
-
-            Use `mode` to control retrieval quality:
-            - "fast" (default): Single query, faster response
-            - "accurate": Multi-query generation with reranking, higher quality
-
         Parameters
         ----------
         tenant_id : str
@@ -660,7 +613,7 @@ class AsyncRawSearchClient:
         sub_tenant_id : typing.Optional[str]
             Optional sub-tenant identifier used to organize data within a tenant. If omitted, the default sub-tenant created during tenant setup will be used.
 
-        max_chunks : typing.Optional[int]
+        max_results : typing.Optional[int]
             Maximum number of results to return
 
         mode : typing.Optional[RetrieveMode]
@@ -672,21 +625,21 @@ class AsyncRawSearchClient:
         recency_bias : typing.Optional[float]
             Preference for newer content (0.0 = no bias, 1.0 =             strong recency preference)
 
-        num_related_chunks : typing.Optional[int]
-            Number of related content chunks to include
-
-        personalise_search : typing.Optional[bool]
+        personalise_recall : typing.Optional[bool]
             Enable personalized search results based on user preferences
 
         graph_context : typing.Optional[bool]
             Enable graph context for search results
 
-        extra_context : typing.Optional[str]
+        additional_context : typing.Optional[str]
             Additional context provided by the user to guide retrieval
 
         search_mode : typing.Optional[SearchMode]
             What to search: 'sources' for documents or 'memories' for user memories
 
+        metadata_filters : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
+            Optional key-value pairs to filter search results by tenant metadata fields. Keys must match fields defined in tenant_metadata_schema during tenant creation. Supports exact match filtering on indexed metadata fields. Example: {'category': 'engineering', 'priority': 'high'}
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -696,21 +649,21 @@ class AsyncRawSearchClient:
             Successful Response
         """
         _response = await self._client_wrapper.httpx_client.request(
-            "search/hybrid-search",
+            "recall/full_recall",
             method="POST",
             json={
                 "tenant_id": tenant_id,
                 "sub_tenant_id": sub_tenant_id,
                 "query": query,
-                "max_chunks": max_chunks,
+                "max_results": max_results,
                 "mode": mode,
                 "alpha": convert_and_respect_annotation_metadata(object_=alpha, annotation=Alpha, direction="write"),
                 "recency_bias": recency_bias,
-                "num_related_chunks": num_related_chunks,
-                "personalise_search": personalise_search,
+                "personalise_recall": personalise_recall,
                 "graph_context": graph_context,
-                "extra_context": extra_context,
+                "additional_context": additional_context,
                 "search_mode": search_mode,
+                "metadata_filters": metadata_filters,
             },
             headers={
                 "content-type": "application/json",
@@ -828,19 +781,11 @@ class AsyncRawSearchClient:
         query: str,
         sub_tenant_id: typing.Optional[str] = OMIT,
         operator: typing.Optional[Bm25OperatorType] = OMIT,
-        max_chunks: typing.Optional[int] = OMIT,
+        max_results: typing.Optional[int] = OMIT,
         search_mode: typing.Optional[SearchMode] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[RetrievalResult]:
         """
-        Perform full text search for exact matches within your indexed sources or memories.
-            Choose between 'OR' and 'AND' operators to control how search terms are combined
-            for precise text matching.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories
-
         Parameters
         ----------
         tenant_id : str
@@ -855,7 +800,7 @@ class AsyncRawSearchClient:
         operator : typing.Optional[Bm25OperatorType]
             How to combine search terms (OR or AND)
 
-        max_chunks : typing.Optional[int]
+        max_results : typing.Optional[int]
             Maximum number of results to return
 
         search_mode : typing.Optional[SearchMode]
@@ -870,14 +815,14 @@ class AsyncRawSearchClient:
             Successful Response
         """
         _response = await self._client_wrapper.httpx_client.request(
-            "search/full-text-search",
+            "recall/boolean_recall",
             method="POST",
             json={
                 "tenant_id": tenant_id,
                 "sub_tenant_id": sub_tenant_id,
                 "query": query,
                 "operator": operator,
-                "max_chunks": max_chunks,
+                "max_results": max_results,
                 "search_mode": search_mode,
             },
             headers={
@@ -1008,19 +953,6 @@ class AsyncRawSearchClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[QnASearchResponse]:
         """
-        Ask a question and get an AI-generated answer based on your indexed sources or memories.
-
-            The response includes both the AI answer and the source chunks used to generate it,
-            enabling full transparency and citation capabilities.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories
-
-            Use `mode` to control retrieval quality:
-            - "fast" (default): Single query, faster response
-            - "accurate": Multi-query generation with reranking, higher quality
-
         Parameters
         ----------
         tenant_id : str
@@ -1071,7 +1003,7 @@ class AsyncRawSearchClient:
             Successful Response
         """
         _response = await self._client_wrapper.httpx_client.request(
-            "search/qna",
+            "recall/qna",
             method="POST",
             json={
                 "tenant_id": tenant_id,

usecortex_ai/sources/client.py

@@ -35,13 +35,6 @@ class SourcesClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> SourceDeleteResponse:
         """
-        Remove documents and content from your knowledge base.
-
-        This endpoint permanently deletes the specified sources from your knowledge base.
-         Once deleted, the content will no longer be available for search or retrieval.
-
-        Use this carefully as the action cannot be undone. The system will confirm successful deletion of each source ID you specify.
-
         Parameters
         ----------
         tenant_id : str
@@ -64,7 +57,7 @@ class SourcesClient:
         from usecortex-ai import CortexAI
 
         client = CortexAI(token="YOUR_TOKEN", )
-        client.sources.delete(tenant_id='tenant_id', sub_tenant_id='sub_tenant_id', source_ids=['source_ids'], )
+        client.sources.delete(tenant_id='tenant_1234', sub_tenant_id='sub_tenant_4567', source_ids=['source_ids'], )
         """
         _response = self._raw_client.delete(
             tenant_id=tenant_id, sub_tenant_id=sub_tenant_id, source_ids=source_ids, request_options=request_options
@@ -96,13 +89,6 @@ class AsyncSourcesClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> SourceDeleteResponse:
         """
-        Remove documents and content from your knowledge base.
-
-        This endpoint permanently deletes the specified sources from your knowledge base.
-         Once deleted, the content will no longer be available for search or retrieval.
-
-        Use this carefully as the action cannot be undone. The system will confirm successful deletion of each source ID you specify.
-
         Parameters
         ----------
         tenant_id : str
@@ -128,7 +114,7 @@ class AsyncSourcesClient:
 
         client = AsyncCortexAI(token="YOUR_TOKEN", )
         async def main() -> None:
-            await client.sources.delete(tenant_id='tenant_id', sub_tenant_id='sub_tenant_id', source_ids=['source_ids'], )
+            await client.sources.delete(tenant_id='tenant_1234', sub_tenant_id='sub_tenant_4567', source_ids=['source_ids'], )
         asyncio.run(main())
         """
         _response = await self._raw_client.delete(

usecortex_ai/sources/raw_client.py

@@ -34,13 +34,6 @@ class RawSourcesClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[SourceDeleteResponse]:
         """
-        Remove documents and content from your knowledge base.
-
-        This endpoint permanently deletes the specified sources from your knowledge base.
-         Once deleted, the content will no longer be available for search or retrieval.
-
-        Use this carefully as the action cannot be undone. The system will confirm successful deletion of each source ID you specify.
-
         Parameters
         ----------
         tenant_id : str
@@ -59,7 +52,7 @@ class RawSourcesClient:
             Successful Response
         """
         _response = self._client_wrapper.httpx_client.request(
-            "delete/delete-sources",
+            "knowledge/delete_knowledge",
             method="POST",
             json={
                 "tenant_id": tenant_id,
@@ -178,13 +171,6 @@ class AsyncRawSourcesClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[SourceDeleteResponse]:
         """
-        Remove documents and content from your knowledge base.
-
-        This endpoint permanently deletes the specified sources from your knowledge base.
-         Once deleted, the content will no longer be available for search or retrieval.
-
-        Use this carefully as the action cannot be undone. The system will confirm successful deletion of each source ID you specify.
-
         Parameters
         ----------
         tenant_id : str
@@ -203,7 +189,7 @@ class AsyncRawSourcesClient:
             Successful Response
         """
         _response = await self._client_wrapper.httpx_client.request(
-            "delete/delete-sources",
+            "knowledge/delete_knowledge",
             method="POST",
             json={
                 "tenant_id": tenant_id,

usecortex_ai/tenant/client.py

@@ -38,47 +38,6 @@ class TenantClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> TenantCreateResponse:
         """
-        Create a tenant for your account.
-
-        Use this endpoint to initialize a tenant space you can use for ingestion, embeddings, and search.
-
-        **Tenant Metadata Schema**
-
-        You can optionally provide a `tenant_metadata_schema` to define custom fields that will be
-        indexed in the vector store. Each field can be configured with:
-
-        - `enable_match`: Enable text filtering on this field
-        - `enable_dense_embedding`: Create dense embeddings for semantic similarity search
-        - `enable_sparse_embedding`: Create sparse embeddings (BM25) for keyword search
-
-        **Example Request:**
-        ```json
-        {
-            "tenant_id": "my-tenant",
-            "tenant_metadata_schema": [
-                {
-                    "name": "category",
-                    "data_type": "VARCHAR",
-                    "max_length": 256,
-                    "enable_match": true
-                },
-                {
-                    "name": "product_description",
-                    "data_type": "VARCHAR",
-                    "max_length": 4096,
-                    "enable_dense_embedding": true,
-                    "enable_sparse_embedding": true
-                }
-            ]
-        }
-        ```
-
-        Expected outcome:
-        - A tenant is created and returned with its identifier.
-        - If tenant_metadata_schema is provided, the vector store collection will include
-          the specified custom fields with their configured search capabilities.
-        - If the tenant already exists, you receive a success message with the existing identifier.
-
         Parameters
         ----------
         tenant_id : str
@@ -106,7 +65,7 @@ class TenantClient:
         from usecortex-ai import CortexAI
 
         client = CortexAI(token="YOUR_TOKEN", )
-        client.tenant.create_tenant(tenant_id='tenant_id', )
+        client.tenant.create_tenant(tenant_id='tenant_1234', )
         """
         _response = self._raw_client.create_tenant(
             tenant_id=tenant_id,
@@ -119,17 +78,6 @@ class TenantClient:
 
     def stats(self, *, tenant_id: str, request_options: typing.Optional[RequestOptions] = None) -> TenantStatsResponse:
         """
-        Retrieve usage stats for your tenant.
-
-        Use this endpoint to check whether a tenant exists and view core metrics like total
-        indexed objects and vector dimension. This helps you validate
-        setup and monitor ingestion.
-
-        Expected outcome
-
-        You receive the current object count and vector dimension for the tenant.
-        If the tenant does not exist, you get a not-found error.
-
         Parameters
         ----------
         tenant_id : str
@@ -148,7 +96,7 @@ class TenantClient:
         from usecortex-ai import CortexAI
 
         client = CortexAI(token="YOUR_TOKEN", )
-        client.tenant.stats(tenant_id='tenant_id', )
+        client.tenant.stats(tenant_id='tenant_1234', )
         """
         _response = self._raw_client.stats(tenant_id=tenant_id, request_options=request_options)
         return _response.data
@@ -179,47 +127,6 @@ class AsyncTenantClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> TenantCreateResponse:
         """
-        Create a tenant for your account.
-
-        Use this endpoint to initialize a tenant space you can use for ingestion, embeddings, and search.
-
-        **Tenant Metadata Schema**
-
-        You can optionally provide a `tenant_metadata_schema` to define custom fields that will be
-        indexed in the vector store. Each field can be configured with:
-
-        - `enable_match`: Enable text filtering on this field
-        - `enable_dense_embedding`: Create dense embeddings for semantic similarity search
-        - `enable_sparse_embedding`: Create sparse embeddings (BM25) for keyword search
-
-        **Example Request:**
-        ```json
-        {
-            "tenant_id": "my-tenant",
-            "tenant_metadata_schema": [
-                {
-                    "name": "category",
-                    "data_type": "VARCHAR",
-                    "max_length": 256,
-                    "enable_match": true
-                },
-                {
-                    "name": "product_description",
-                    "data_type": "VARCHAR",
-                    "max_length": 4096,
-                    "enable_dense_embedding": true,
-                    "enable_sparse_embedding": true
-                }
-            ]
-        }
-        ```
-
-        Expected outcome:
-        - A tenant is created and returned with its identifier.
-        - If tenant_metadata_schema is provided, the vector store collection will include
-          the specified custom fields with their configured search capabilities.
-        - If the tenant already exists, you receive a success message with the existing identifier.
-
         Parameters
         ----------
         tenant_id : str
@@ -250,7 +157,7 @@ class AsyncTenantClient:
 
         client = AsyncCortexAI(token="YOUR_TOKEN", )
         async def main() -> None:
-            await client.tenant.create_tenant(tenant_id='tenant_id', )
+            await client.tenant.create_tenant(tenant_id='tenant_1234', )
         asyncio.run(main())
         """
         _response = await self._raw_client.create_tenant(
@@ -266,17 +173,6 @@ class AsyncTenantClient:
         self, *, tenant_id: str, request_options: typing.Optional[RequestOptions] = None
     ) -> TenantStatsResponse:
         """
-        Retrieve usage stats for your tenant.
-
-        Use this endpoint to check whether a tenant exists and view core metrics like total
-        indexed objects and vector dimension. This helps you validate
-        setup and monitor ingestion.
-
-        Expected outcome
-
-        You receive the current object count and vector dimension for the tenant.
-        If the tenant does not exist, you get a not-found error.
-
         Parameters
         ----------
         tenant_id : str
@@ -298,7 +194,7 @@ class AsyncTenantClient:
 
         client = AsyncCortexAI(token="YOUR_TOKEN", )
         async def main() -> None:
-            await client.tenant.stats(tenant_id='tenant_id', )
+            await client.tenant.stats(tenant_id='tenant_1234', )
         asyncio.run(main())
         """
         _response = await self._raw_client.stats(tenant_id=tenant_id, request_options=request_options)