AI Agents (Claude, GPT)

Combine MeetLogin with AI automation frameworks to build intelligent, multi-account web agents.

Architecture

The pattern is: AI Agent → MeetLogin API (creates/launches profile) → Playwright/Puppeteer connects via CDP → AI controls the browser → profile stops and syncs.

Python Automation Example

import anthropic
from meetlogin import MeetLogin
from playwright.sync_api import sync_playwright

ml = MeetLogin(token="YOUR_API_KEY")
claude = anthropic.Anthropic()

def run_task(profile_id: str, task: str):
    # Launch the profile
    result = ml.profiles.start(profile_id)
    ws_url = result['wsUrl']

    with sync_playwright() as p:
        browser = p.chromium.connect_over_cdp(ws_url)
        page = browser.new_page()

        # Take a screenshot and send to Claude
        screenshot = page.screenshot()
        response = claude.messages.create(
            model="claude-sonnet-4-6",
            max_tokens=1024,
            messages=[{
                "role": "user",
                "content": [
                    {"type": "image", "source": {
                        "type": "base64",
                        "media_type": "image/png",
                        "data": screenshot.decode('base64')
                    }},
                    {"type": "text", "text": task}
                ]
            }]
        )
        print(response.content[0].text)
        browser.close()

    ml.profiles.stop(profile_id)

run_task("PROFILE_ID", "What is the current price of the first product on this page?")

Best Practices

• Always stop profiles after the task completes to trigger cloud sync.
• Add human-like delays between actions to avoid bot detection.
• Use residential proxies for better success rates on sensitive sites.
• Rotate profiles for high-volume tasks instead of reusing a single profile.