Feat: Make Tavily search timeout configurable

src/open_deep_research/configuration.py

@@ -117,6 +117,18 @@ class Configuration(BaseModel):
             }
         }
     )
+    tavily_timeout: int = Field(
+        default=60,
+        metadata={
+            "x_oap_ui_config": {
+                "type": "number",
+                "default": 60,
+                "min": 1,
+                "max": 300,
+                "description": "Timeout for Tavily search in seconds."
+            }
+        }
+    )
     # Model Configuration
     summarization_model: str = Field(
         default="openai:gpt-4.1-mini",

src/open_deep_research/utils.py

@@ -169,8 +169,13 @@ async def tavily_search_async(
     ]
 
     # Execute all search queries in parallel and return results
-    search_results = await asyncio.gather(*search_tasks)
-    return search_results
+    configurable = Configuration.from_runnable_config(config)
+    try:
+        search_results = await asyncio.wait_for(asyncio.gather(*search_tasks), timeout=configurable.tavily_timeout)
+        return search_results
+    except asyncio.TimeoutError:
+        logging.warning(f"Tavily search timed out after {configurable.tavily_timeout} seconds")
+        return []
 
 async def summarize_webpage(model: BaseChatModel, webpage_content: str) -> str:
     """Summarize webpage content using AI model with timeout protection.