alihan/OmniParser
1
0
Fork 0
mirror of https://github.com/microsoft/OmniParser.git synced 2025-02-18 03:18:33 +03:00
Code Issues Packages Projects Releases Wiki Activity

default click/enter for type

Browse Source
This commit is contained in:
yadonglu
2025-02-12 16:11:20 -08:00
parent d60460d62c
commit 9956befe20
1 changed files with 3 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
omnitool/gradio/tools/computer.py
View File

@@ -170,7 +170,10 @@ class ComputerTool(BaseAnthropicTool):
return ToolResult(output=f"Pressed keys: {text}")
elif action == "type":
# default click before type TODO: check if this is needed
self.send_to_vm("pyautogui.click()")
self.send_to_vm(f"pyautogui.typewrite('{text}', interval={TYPING_DELAY_MS / 1000})")
self.send_to_vm("pyautogui.press('enter')")
screenshot_base64 = (await self.screenshot()).base64_image
return ToolResult(output=text, base64_image=screenshot_base64)