browser-use

uv pip install browser-use
uv run patchright install

 
from browser_use import Agent, Browser, BrowserConfig, Controller
from langchain_openai import ChatOpenAI
import asyncio
from typing import List
from pydantic import BaseModel
import json
from langchain_google_genai import ChatGoogleGenerativeAI
 
from dotenv import load_dotenv
load_dotenv()
 
 
class Company(BaseModel):
    name: str
 
 
class Companies(BaseModel):
    companies: List[Company]
    source_urls: List[str]
 
 
BATCH = "X25"
ACCELERATOR = "Y Combinator"
 
import os
 
batch_nospace = BATCH.replace(" ", "_")
accelerator_nospace = ACCELERATOR.replace(" ", "_")
output_file = f"{accelerator_nospace}_{batch_nospace}_companies.json"
 
TASK = (
    f"Find the official page(s) for the {ACCELERATOR} {BATCH} batch. "
    f"Extract all the names of companies from those pages and return them as a JSON object with two fields: "
    f"'companies' (a list of objects with 'name'), and 'source_urls' (a list of the URLs you extracted from)."
)
 
async def main():
    initial_actions = [
        {'open_tab': {'url': 'https://google.com'}},
    ]
 
    browser = Browser(
        config=BrowserConfig(
            browser_binary_path='/Applications/Google Chrome.app/Contents/MacOS/Google Chrome',
        )
    )
 
    controller = Controller(output_model=Companies)
 
    agent = Agent(
        task=TASK,
        llm=ChatGoogleGenerativeAI(model='gemini-2.0-flash'),
        browser=browser,
        controller=controller,
        initial_actions=initial_actions
    )
 
    history = await agent.run()
 
    result = history.final_result()
    if result:
        parsed: Companies = Companies.model_validate_json(result)
        print("\nExtracted companies:")
        for company in parsed.companies:
            print(f"- {company.name}")
        print("\nSource URLs:")
        for url in parsed.source_urls:
            print(f"- {url}")
 
        output_dir = os.path.dirname(output_file)
        if output_dir:
            os.makedirs(output_dir, exist_ok=True)
        with open(output_file, "w") as f:
            json.dump({
                "companies": [c.model_dump() for c in parsed.companies],
                "source_urls": parsed.source_urls
            }, f, indent=2)
        print(f"\nSaved {len(parsed.companies)} companies to {output_file}")
    else:
        print('No result')
 
    input('Press Enter to close the browser...')
    await browser.close()
 
if __name__ == '__main__':
    asyncio.run(main())