Tool desc again

vlm_tools.py

@@ -130,14 +130,19 @@ onnx_path = "vlm_assets/yolov3-8.onnx"
 names_path = "vlm_assets/obj.names"
 
 class ObjectDetectionTool(Tool):
-    description = "Detect objects in a list of frames (images). It takes a list of frames (images) as input and returns a list of detected objects with labels, confidence, and bounding boxes."
+    description = """
+    Detect objects in a list of frames (images). 
+    It takes a list of frames (images) as input and returns 
+    a list of detected objects with labels, confidence, and bounding boxes.
+    The output type will be List[List[str]]
+    """
     name = "object_detection"
     inputs = {
         "frames": {"type": "any", "description": "The list of frames (images) to detect objects in. Must be a List[str] or a List[np.ndarray]"},
         "onnx_path": {"type": "string", "description": "The path to the onnx file"},
         "names_path": {"type": "string", "description": "The path to the names file"}
     }
-    output_type = "List[List[str]]"
+    output_type = "any"
     
     def setup(self):
         # Load ONNX model
@@ -164,12 +169,17 @@ class ObjectDetectionTool(Tool):
         return detected_objects
 
 class OCRTool(Tool):
-    description = "Scan an image for text. It takes a list of frames (images) as input and returns a list of text in the images."
+    description = """
+    Scan an image for text. 
+    It takes a list of frames (images) as input and returns 
+    a list of text in the images.
+    The output type will be List[List[str]]
+    """
     name = "ocr_scan"
     inputs = {
         "frames": {"type": "any", "description": "The list of frames (images) to scan for text. Must be a List[str] or a List[np.ndarray]"}
     }
-    output_type = "List[List[str]]"
+    output_type = "any"
 
     def forward(self, frames: any)->any:
         scanned_text = []