优化操作逻辑

gradio_ui/app.py

@@ -7,6 +7,7 @@ from datetime import datetime
 from enum import StrEnum
 from functools import partial
 from pathlib import Path
+import time
 from typing import cast
 import argparse
 import gradio as gr
@@ -170,7 +171,8 @@ def process_input(user_input, state, vision_agent_state):
     # Reset the stop flag
     if state["stop"]:
         state["stop"] = False
-    
+    # wait 3 seconds
+    time.sleep(3)
     config = Config()
     config.set_openai_config(base_url=state["base_url"], api_key=state["api_key"], model=state["model"])
     # Append the user message to state["messages"]

gradio_ui/loop.py

@@ -1,10 +1,8 @@
 """
 Agentic sampling loop that calls the Anthropic API and local implenmentation of anthropic-defined computer use tools.
 """
-import base64
 from collections.abc import Callable
-from time import time
-
+from time import sleep
 import cv2
 from gradio_ui.agent.vision_agent import VisionAgent
 from gradio_ui.tools.screen_capture import get_screenshot
@@ -54,7 +52,7 @@ def sampling_loop_sync(
     while True:
         parsed_screen = parse_screen(vision_agent)
         tools_use_needed, __ = task_run_agent(task_plan=plan, parsed_screen=parsed_screen)
-        time.sleep(1)
+        sleep(2)
         for message, tool_result_content in executor(tools_use_needed, messages):
             yield message
         if not tool_result_content:

gradio_ui/tools/computer.py

@@ -1,6 +1,5 @@
 import base64
 import time
-from enum import StrEnum
 from typing import Literal, TypedDict
 from PIL import Image
 from util import tool
@@ -175,6 +174,8 @@ class ComputerTool(BaseAnthropicTool):
                     pyautogui.click()
                 elif action == "right_click":
                     pyautogui.rightClick()
+                    # 等待5秒，等待菜单弹出
+                    time.sleep(5)
                 elif action == "middle_click":
                     pyautogui.middleClick()
                 elif action == "double_click":