Sync from GitHub via hub-sync

README.md

@@ -19,9 +19,9 @@ This repository implements and evaluates two personal assistants with the same u
 
 Both assistants support multi-turn conversation, short-term conversational memory, basic assistant behavior, safety filtering, latency logging, and a side-by-side Streamlit comparison interface.
 
-🚀 **Live Interactive Demo:**  
+**Live Interactive Demo:**  
 The project is publicly deployed and accessible on Hugging Face Spaces at:  
-👉 **[https://huggingface.co/spaces/vedanshmathur7/ModelArena](https://huggingface.co/spaces/vedanshmathur7/ModelArena)**
+**[https://huggingface.co/spaces/vedanshmathur7/ModelArena](https://huggingface.co/spaces/vedanshmathur7/ModelArena)**
 
 ### How to use the Live Demo:
 1. **Public Access:** Anyone can open the link in a browser to use the side-by-side chat interface.

core/tools.py

@@ -106,6 +106,40 @@ def get_tools_definition() -> List[Dict[str, Any]]:
                 },
             },
         },
+        {
+            "type": "function",
+            "function": {
+                "name": "brave_search",
+                "description": "Search the web for current or historical information. Use this only when the query requires searching for external or up-to-date information.",
+                "parameters": {
+                    "type": "object",
+                    "properties": {
+                        "query": {
+                            "type": "string",
+                            "description": "The search query."
+                        }
+                    },
+                    "required": ["query"],
+                },
+            },
+        },
+        {
+            "type": "function",
+            "function": {
+                "name": "web_search",
+                "description": "Search the web for current or historical information. Use this only when the query requires searching for external or up-to-date information.",
+                "parameters": {
+                    "type": "object",
+                    "properties": {
+                        "query": {
+                            "type": "string",
+                            "description": "The search query."
+                        }
+                    },
+                    "required": ["query"],
+                },
+            },
+        }
     ]
 
 
@@ -117,6 +151,10 @@ def execute_tool(name: str, arguments: Dict[str, Any]) -> str:
     elif name == "calculate":
         expression = arguments.get("expression", "")
         return calculate(expression)
+    elif name in ("brave_search", "web_search"):
+        query = arguments.get("query", "")
+        logger.info("Mock search executed for query: %s", query)
+        return f"Search results for '{query}': No search connection is active. Please answer this query directly using your existing knowledge, or explain that you do not have search capability."
     else:
         return f"Error: Unknown tool '{name}'."
 

evals/evaluation_runner.py

@@ -22,6 +22,12 @@ import sys
 import time
 from pathlib import Path
 
+try:
+    from dotenv import load_dotenv
+    load_dotenv()
+except ImportError:
+    pass
+
 # Ensure project root is on path
 sys.path.insert(0, str(Path(__file__).parent.parent))
 

reports/evaluation_report.pdf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a10d18efe75c94212f98f4f9e7d61ed345c81ae6dcb672d820517ca4294a85a
+oid sha256:3bc02251f44fbde7f9dbfe79e5a04c5b8fbc17247842c63515c313573cf3c772
 size 444980