Bug Description

Hey - browser-use was not able to navigate shadow-roots and nested i-frames. Please fix this.

`Call log:

• waiting for locator("html > body > div > div:nth-of-type(2) > iframe[id="mons-body-container"][src="https://sellercentral.amazon.in/help/center?redirectSource=HelpHub"]").content_frame.locator("button.button[type="button"]")

WARNING [controller] Element not clickable with index 32 - most likely the page changed
INFO [agent] 📍 Step 18
INFO [agent] ⚠ Eval: Failed - The 'Continue' button was not clicked due to an error.
INFO [agent] 🧠 Memory: Entered 'Missing Inventory in FC' in the textarea. Need to click 'Continue' again due to previous error.
INFO [agent] 🎯 Next goal: Attempt to click the 'Continue' button again.
INFO [agent] 🛠️ Action 1/1: {"click_element":{"index":32}}
ERROR [browser] Failed to locate element: Locator.element_handle: Error: strict mode violation: locator("button.button[type="button"]") resolved to 2 elements:
1) … aka locator("kat-button").filter(has_text="Show more").locator("button")
2) … aka get_by_role("button", name="Continue")`

Reproduction Steps

• example provided in the prompt

Code Sample

import os, asyncio
from browser_use import Agent
from pydantic import SecretStr
from langchain_openai import AzureChatOpenAI, ChatOpenAI

# Initialize the model
llm = ChatOpenAI(
    model="gpt-4o",
    temperature=0.0,
)

async def main():
    # Create agent with the model
    agent = Agent(
        task=''' 
        1. go to sellercentral.amazon.in
        2. login via credentials username: xxx password: xxx
        3. wait for the otp i'll provide it to you
        4. after this select the sepoy account and then select india and click select account
        5. hover over help section on the top right and wait for the dropdown & select get help and resources
        6. click on my issue is not listed - wait for the textbox to appear - and in the textbox type "Missing Inventory in FC"
        7. click continue
        ''',
        llm=llm,
    )
    result = await agent.run()
    print(result)

asyncio.run(main())

Version

0.1.35

LLM Model

GPT-4

Operating System

macOS

Relevant Log Output