supporting image tool

src/crewai/agents/crew_agent_executor.py

@@ -144,9 +144,13 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
                             formatted_answer
                         )
 
+                        # Directly append the result to the messages if the
+                        # tool is "Add image to content" in case of multimodal
+                        # agents
                         if formatted_answer.tool == "Add image to content":
                             self.messages.append(tool_result.result)
                             continue
+
                         else:
                             if self.step_callback:
                                 self.step_callback(tool_result)

src/crewai/tools/agent_tools/add_image_tool.py

@@ -4,31 +4,37 @@ from crewai.tools.base_tool import BaseTool
 
 class AddImageToolSchema(BaseModel):
     image_url: str = Field(..., description="The URL or path of the image to add")
-    action: str = Field(..., description="The context or purpose of why this image is being added and how it should be used")
+    action: str = Field(
+        default="Please provide a detailed description of this image, including all visual elements, context, and any notable details you can observe.",
+        description="Optional context or question about the image"
+    )
 
 
 class AddImageTool(BaseTool):
     """Tool for adding images to the content"""
 
     name: str = "Add image to content"
-    description: str = "See image to understand it's content"
+    description: str = "See image to understand it's content, you can optionally ask a question about the image"
     args_schema: type[BaseModel] = AddImageToolSchema
 
     def _run(
         self,
         image_url: str,
-        action: str,
+        action: str = None,
         **kwargs,
     ) -> dict:
+        action = action or "Please provide a detailed description of this image, including all visual elements, context, and any notable details you can observe."
+        content = [
+            {"type": "text", "text": action},
+            {
+                "type": "image_url",
+                "image_url": {
+                    "url": image_url,
+                },
+            }
+        ]
+
         return {
             "role": "user",
-            "content": [
-                {"type": "text", "text": action},
-                {
-                    "type": "image_url",
-                    "image_url": {
-                        "url": image_url,
-                    },
-                },
-            ],
+            "content": content
         }

src/crewai/tools/tool_usage.py

@@ -5,6 +5,7 @@ from difflib import SequenceMatcher
 from textwrap import dedent
 from typing import Any, List, Union
 
+from crewai.tools.structured_tool import CrewStructuredTool
 import crewai.utilities.events as events
 from crewai.agents.tools_handler import ToolsHandler
 from crewai.task import Task
@@ -103,6 +104,19 @@ class ToolUsage:
             if self.agent.verbose:
                 self._printer.print(content=f"\n\n{error}\n", color="red")
             return error
+
+        if isinstance(tool, CrewStructuredTool) and tool.name == 'Add image to content':
+            try:
+                result = self._use(tool_string=tool_string, tool=tool, calling=calling)
+                return result
+
+            except Exception as e:
+                error = getattr(e, "message", str(e))
+                self.task.increment_tools_errors()
+                if self.agent.verbose:
+                    self._printer.print(content=f"\n\n{error}\n", color="red")
+                return error
+
         return f"{self._use(tool_string=tool_string, tool=tool, calling=calling)}"  # type: ignore # BUG?: "_use" of "ToolUsage" does not return a value (it only ever returns None)
 
     def _use(