Update app.py

app.py

@@ -73,6 +73,10 @@ def process_image_task(image_path: str, task_text: str, agent: CellposeAgent) ->
     """
     Process a user task with the CellposeAgent.
     
+    The image_path is now passed SEPARATELY to agent.run() so it gets stored
+    in global context. This prevents the LLM from needing to reproduce the 
+    exact path string (which it often corrupts).
+    
     Args:
         image_path: Path to the uploaded image file
         task_text: User's text prompt/question
@@ -84,25 +88,18 @@ def process_image_task(image_path: str, task_text: str, agent: CellposeAgent) ->
     if not image_path:
         return "⚠️ Please upload an image first.", None
     
-    # Store the actual image path in agent context/memory BEFORE running
-    # This ensures the agent has access to the correct path
-    #agent.agent.memory.clear()  # Optional: clear previous context
-    
-    # Build task with clear, unambiguous path format
+    # Build task WITHOUT embedding the full path
+    # The agent instructions tell it to use image_path="" and the tool will resolve from context
     if not task_text or task_text.strip() == "":
-        # Use a more explicit format that's harder for LLM to corrupt
-        task = f"""Analyze the image at this exact path: `{image_path}`
-                What parameters would work best for this image?"""
+        task = "Analyze the uploaded image and recommend optimal segmentation parameters. Then run segmentation."
     else:
-        if image_path not in task_text:
-            task = f"""Image path: `{image_path}`
-                    Task: {task_text}"""
-        else:
-            task = task_text
-    
+        # Don't include the path in the task - it will be available via context
+        task = task_text
     
     try:
-        result = agent.run(task)
+        # Pass image_path SEPARATELY - this stores it in global context
+        # Tools will retrieve the correct path from context instead of relying on LLM
+        result = agent.run(task=task, image_path=image_path)
         get_client().flush()
         
         # Extract output image path from agent's memory (more reliable than parsing text)
@@ -171,7 +168,7 @@ def create_gradio_interface():
                 # Task input
                 task_input = gr.Textbox(
                     label="Task / Question",
-                    placeholder="e.g., 'What parameters would work best?' or 'Run segmentation' (image path added automatically)",
+                    placeholder="e.g., 'What parameters would work best?' or 'Run segmentation' (image path handled automatically)",
                     lines=3
                 )
                 
@@ -233,6 +230,11 @@ def create_gradio_interface():
             4. Runs segmentation and checks quality
             5. Refines parameters if needed (up to 2 iterations)
             6. **Displays the segmented overlay image with colored cell masks**
+            
+            ### ⚙️ Technical Note
+            
+            Image paths are handled automatically by the system - the agent doesn't need to 
+            reproduce exact file paths, preventing common path corruption issues with LLMs.
             """
         )