Update system prompt and enhance image generation response handling

app.py

@@ -13,6 +13,7 @@ import openai
 from openai import OpenAI
 import pdfplumber
 
+
 ## utilty functions
 def is_image_extension(filename: str) -> bool:
     IMAGE_EXTS = {'.jpg', '.jpeg', '.png', '.gif', '.bmp', '.tiff', '.webp', '.svg'}
@@ -45,7 +46,7 @@ def load_file(path: str) -> list | dict:
         return {"raw document text": text, "audio path": path} 
     else:
         return {"raw document text": text, "file path": path}
-
+    
 
 ## tools definition
 @tool
@@ -126,8 +127,8 @@ def generate_image(prompt: str, neg_prompt: str) -> Image.Image:
     image_data = base64.b64decode(completion.to_dict()['data'][0]['b64_json'])
     image = BytesIO(image_data)
     image = Image.open(image).convert("RGB") 
-    return gr.Image(value=image)
 
+    return gr.Image(value=image, label="Generated Image")
 
 
 ## agent definition
@@ -145,15 +146,16 @@ class Agent:
             system_prompt = f.read()
             self.agent.prompt_templates["system_prompt"] = system_prompt
         
-        print("System prompt:", self.agent.prompt_templates["system_prompt"])
+        #print("System prompt:", self.agent.prompt_templates["system_prompt"])
 
     def __call__(self, message: str, images: Optional[list[Image.Image]] = None, files: Optional[str] = None) -> str:
         answer = self.agent.run(message, images = images, additional_args={"files": files})
         return answer
 
 ## gradio functions
-def respond(message, history):
+def respond(message: str, history : dict):
     
+    # input
     text = message.get("text", "")
     if not message.get("files"): # no files uploaded
         print("No files received.")
@@ -168,10 +170,12 @@ def respond(message, history):
             file = load_file(files[0])
             message = agent(text, files=file)
 
+    # output
+    print("Agent response:", message)
+    
     return message
     
 
-
 def initialize_agent():
     agent = Agent()
     print("Agent initialized.")

system_prompt.txt

@@ -237,8 +237,7 @@ derail you from the true variables.
 9. The state persists between code executions: so if in one step you've created 
 variables or imported modules, these will all persist. 
 10. Don't give up! You're in charge of solving the task, not providing directions to 
-solve it. 
- 
+solve it.
 
 ---- 
 
@@ -314,6 +313,19 @@ Additional domain-specific behaviors:
    - Where applicable, the assistant must provide annotated examples and suggest possible reframing or source guidance (e.g., UNESCO media diversity principles).
 ---- 
  
+
+It is MANDATORY to use these rules for the 'final_answer' tool:
+	1.	Always return a Python list. Do not return a dictionary or any other type.
+	2.	The first element of the list must be a Gradio component (for example, gr.Image(…), gr.File(…), etc.).
+	3.	Any text or explanation must come after the component, as string elements in the same list.
+	4.	If there is no component to return, return a list whose only element is the text.
+	5.	Examples of valid returns:
+• [gr.Image(value=“output.png”), “Here is the chart.”]
+• [None, “No media available, here is the text.”]
+• [gr.File(value=“report.pdf”), “Download the report.”]
+	7.	Any deviation (returning a dict, tuple, raw PIL image, etc.) is invalid.
+
+
 Now Begin!