Working Ctrl F tool

app.py

@@ -585,6 +585,7 @@ with gr.Blocks(theme=theme, css=custom_css, js=custom_js) as demo:
                 examples=[
                     "Check the commuting time between Bern and Zurich on Google maps",
                     "Write 'Hello World' in a text editor",
+                    "When was Temple Grandin introduced to the American Academy of Arts and Sciences, according to Wikipedia?",
                     "Search a flight Rome - Berlin for tomorrow",
                     "What' s the name of the pond just south of Château de Fontainebleau in Google maps?",
                     "Go generate a picture of the Golden Gate bridge on a FLUX1.dev space",
@@ -657,7 +658,6 @@ with gr.Blocks(theme=theme, css=custom_css, js=custom_js) as demo:
                 label="Header"
             )
 
-    stop_btn = gr.Button("Stop the agent!", variant="stop")
 
     chatbot_display = gr.Chatbot(
         elem_id="chatbot",
@@ -672,6 +672,8 @@ with gr.Blocks(theme=theme, css=custom_css, js=custom_js) as demo:
 
     agent_ui = EnrichedGradioUI(CodeAgent(tools=[], model=None, name="ok", description="ok"))
 
+    stop_btn = gr.Button("Stop the agent!", variant="huggingface")
+
     def read_log_content(log_file, tail=4):
         """Read the contents of a log file for a specific session"""
         if not log_file:
@@ -696,7 +698,7 @@ with gr.Blocks(theme=theme, css=custom_css, js=custom_js) as demo:
         return update_html(True, request)
 
     def reactivate_stop_btn():
-        return gr.Button("Stop the agent!", variant="stop")
+        return gr.Button("Stop the agent!", variant="huggingface")
 
     is_interactive = gr.Checkbox(value=True, visible=False)
 
@@ -718,8 +720,11 @@ with gr.Blocks(theme=theme, css=custom_css, js=custom_js) as demo:
     )
 
     def interrupt_agent(session_state):
-        session_state["agent"].interrupt()
-        return gr.Button("Stopping agent... (could take time)", variant="secondary")
+        if not session_state["agent"].interrupt_switch:
+            session_state["agent"].interrupt()
+            return gr.Button("Stopping agent... (could take time)", variant="secondary")
+        else:
+            return gr.Button("Stop the agent!", variant="huggingface")
 
     stop_btn.click(fn=interrupt_agent, inputs=[session_state], outputs=[stop_btn])
 

e2bqwen.py

@@ -356,6 +356,24 @@ class E2BVisionAgent(CodeAgent):
             self.logger.log(f"Opening URL: {url}")
             return f"Opened URL: {url}"
 
+        @tool
+        def find_on_page_ctrl_f(search_string: str) -> str:
+            """
+            Scroll the viewport to the first occurrence of the search string. This is equivalent to Ctrl+F.
+            Args:
+                search_string: The string to search for on the page.
+            """
+            self.desktop.press(["ctrl", "f"])
+            time.sleep(0.3)
+            clean_text = normalize_text(search_string)
+            self.desktop.write(clean_text, delay_in_ms=75)
+            time.sleep(0.3)
+            self.desktop.press("enter")
+            time.sleep(0.3)
+            self.desktop.press("esc")
+            output_message = f"Scrolled to the first occurrence of '{clean_text}'"
+            self.logger.log(output_message)
+            return output_message
 
         # Register the tools
         self.tools["click"] = click
@@ -369,6 +387,7 @@ class E2BVisionAgent(CodeAgent):
         self.tools["open_url"] = open_url
         self.tools["go_back"] = go_back
         self.tools["drag_and_drop"] = drag_and_drop
+        self.tools["find_on_page_ctrl_f"] = find_on_page_ctrl_f
 
 
     def take_screenshot_callback(self, memory_step: ActionStep, agent=None) -> None: