Add URL context to google search

Author: olivier-lacroix

docs/google-gemini-integration.md

@@ -176,13 +176,13 @@ VERTEX_AI_RAG_STORE="projects/your-project/locations/global/collections/default_
 >
 > Future versions may extend these settings to also optimize generated images before upload/display.
 
-## Grounding with Google search
+## Web search and access 
 
-Grounding with Google search is enabled/disabled with the `google_search_tool` feature, which can be switched on/off in a Filter.
+[Grounding with Google search](https://ai.google.dev/gemini-api/docs/google-search) together with the [URL context tool](https://ai.google.dev/gemini-api/docs/url-context) are enabled/disabled together via the `google_search_tool` feature, which can be switched on/off in a Filter.
 
-For instance, the following [Filter (google_search_tool.py)](../filters/google_search_tool.py) will replace Open Web UI default web search function with google search grounding.
+For instance, the following [Filter (google_search_tool.py)](../filters/google_search_tool.py) will replace Open Web UI default web search function with Google search grounding + the URL context tool.
 
-When enabled, sources and google queries used by Gemini will be displayed with the response.
+When enabled, sources and google queries from the search used by Gemini will be displayed with the response.
 
 ## Grounding with Vertex AI Search
 

pipelines/google/google_gemini.py

@@ -28,6 +28,7 @@
   - Intelligent grounding with Google search integration
   - Vertex AI Search grounding for RAG
   - Native tool calling support with automatic signature management
+  - URL context grounding for specified web pages
   - Unified image processing with consolidated helper methods
   - Optimized payload creation for image generation models
   - Configurable image processing parameters (size, quality, compression)
@@ -1411,14 +1412,17 @@ def _configure_generation(
             ]
             gen_config_params |= {"safety_settings": safety_settings}
 
+        # Add various tools to Gemini as required
         features = __metadata__.get("features", {})
+        params = __metadata__.get("params", {})
+        tools = []
+
         if features.get("google_search_tool", False):
             self.log.debug("Enabling Google search grounding")
-            gen_config_params.setdefault("tools", []).append(
-                types.Tool(google_search=types.GoogleSearch())
-            )
+            tools.append(types.Tool(google_search=types.GoogleSearch()))
+            self.log.debug("Enabling URL context grounding")
+            tools.append(types.Tool(url_context=types.UrlContext()))
 
-        params = __metadata__.get("params", {})
         if features.get("vertex_ai_search", False) or (
             self.valves.USE_VERTEX_AI
             and (self.valves.VERTEX_AI_RAG_STORE or os.getenv("VERTEX_AI_RAG_STORE"))
@@ -1429,26 +1433,34 @@ def _configure_generation(
                 or os.getenv("VERTEX_AI_RAG_STORE")
             )
             if vertex_rag_store:
-                self.log.debug(f"Enabling Vertex AI Search grounding: {vertex_rag_store}")
-                gen_config_params.setdefault("tools", []).append(
+                self.log.debug(
+                    f"Enabling Vertex AI Search grounding: {vertex_rag_store}"
+                )
+                tools.append(
                     types.Tool(
                         retrieval=types.Retrieval(
-                            vertex_ai_search=types.VertexAISearch(datastore=vertex_rag_store)
+                            vertex_ai_search=types.VertexAISearch(
+                                datastore=vertex_rag_store
+                            )
                         )
                     )
                 )
             else:
                 self.log.warning(
                     "Vertex AI Search requested but vertex_rag_store not provided in params, valves, or env"
                 )
+
         if __tools__ is not None and params.get("function_calling") == "native":
             for name, tool_def in __tools__.items():
                 if not name.startswith("_"):
                     tool = tool_def["callable"]
                     self.log.debug(
                         f"Adding tool '{name}' with signature {tool.__signature__}"
                     )
-                    gen_config_params.setdefault("tools", []).append(tool)
+                    tools.append(tool)
+
+        if tools:
+            gen_config_params["tools"] = tools
 
         # Filter out None values for generation config
         filtered_params = {k: v for k, v in gen_config_params.items() if v is not None}
@@ -1470,7 +1482,9 @@ def _format_grounding_chunks_as_sources(
                             "uri": getattr(context, "uri", None),
                         },
                         "document": [getattr(context, "chunk_text", None) or ""],
-                        "metadata": [{"source": getattr(context, "title", None) or "Document"}],
+                        "metadata": [
+                            {"source": getattr(context, "title", None) or "Document"}
+                        ],
                     }
                 )
             elif hasattr(chunk, "web") and chunk.web: