feat: optimize vision tools and image handling

- Add model-aware vision tool loading based on supports_vision flag
- Move view_image_tool from config to builtin tools for dynamic inclusion
- Add timeout to image search to prevent hanging requests
- Optimize image search results format using thumbnails
- Add image validation for reference images in generation
- Improve error handling with detailed messages

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

backend/src/agents/lead_agent/agent.py

@@ -230,7 +230,7 @@ def make_lead_agent(config: RunnableConfig):
     print(f"thinking_enabled: {thinking_enabled}, model_name: {model_name}, is_plan_mode: {is_plan_mode}")
     return create_agent(
         model=create_chat_model(name=model_name, thinking_enabled=thinking_enabled),
-        tools=get_available_tools(),
+        tools=get_available_tools(model_name=model_name),
         middleware=_build_middlewares(config),
         system_prompt=apply_prompt_template(),
         state_schema=ThreadState,

backend/src/community/image_search/tools.py

@@ -46,7 +46,7 @@ def _search_images(
         logger.error("ddgs library not installed. Run: pip install ddgs")
         return []
 
-    ddgs = DDGS()
+    ddgs = DDGS(timeout=30)
 
     try:
         kwargs = {
@@ -119,12 +119,8 @@ def image_search_tool(
     normalized_results = [
         {
             "title": r.get("title", ""),
-            "image_url": r.get("image", ""),
+            "image_url": r.get("thumbnail", ""),
             "thumbnail_url": r.get("thumbnail", ""),
-            "source_url": r.get("url", ""),
-            "source": r.get("source", ""),
-            "width": r.get("width"),
-            "height": r.get("height"),
         }
         for r in results
     ]

backend/src/tools/tools.py

@@ -4,7 +4,7 @@ from langchain.tools import BaseTool
 
 from src.config import get_app_config
 from src.reflection import resolve_variable
-from src.tools.builtins import ask_clarification_tool, present_file_tool
+from src.tools.builtins import ask_clarification_tool, present_file_tool, view_image_tool
 
 logger = logging.getLogger(__name__)
 
@@ -14,7 +14,7 @@ BUILTIN_TOOLS = [
 ]
 
 
-def get_available_tools(groups: list[str] | None = None, include_mcp: bool = True) -> list[BaseTool]:
+def get_available_tools(groups: list[str] | None = None, include_mcp: bool = True, model_name: str | None = None) -> list[BaseTool]:
     """Get all available tools from config.
 
     Note: MCP tools should be initialized at application startup using
@@ -23,6 +23,7 @@ def get_available_tools(groups: list[str] | None = None, include_mcp: bool = Tru
     Args:
         groups: Optional list of tool groups to filter by.
         include_mcp: Whether to include tools from MCP servers (default: True).
+        model_name: Optional model name to determine if vision tools should be included.
 
     Returns:
         List of available tools.
@@ -51,4 +52,16 @@ def get_available_tools(groups: list[str] | None = None, include_mcp: bool = Tru
         except Exception as e:
             logger.error(f"Failed to get cached MCP tools: {e}")
 
-    return loaded_tools + BUILTIN_TOOLS + mcp_tools
+    # Conditionally add view_image_tool only if the model supports vision
+    builtin_tools = BUILTIN_TOOLS.copy()
+
+    # If no model_name specified, use the first model (default)
+    if model_name is None and config.models:
+        model_name = config.models[0].name
+
+    model_config = config.get_model_config(model_name) if model_name else None
+    if model_config is not None and model_config.supports_vision:
+        builtin_tools.append(view_image_tool)
+        logger.info(f"Including view_image_tool for model '{model_name}' (supports_vision=True)")
+
+    return loaded_tools + builtin_tools + mcp_tools