linkedin-autobot

linkedin-autobot helps you collect LinkedIn profile data, analyze leads, and automate outreach.

Install

pip install linkedin-autobot

For local browser automation, install automation dependencies manually:

pip install "linkedin-autobot[automation]"
python -m playwright install chromium

Choose Your Mode

1. Hosted mode: recommended for production (stable, backend-driven).
2. Local bot mode: runs browser automation on the user machine.
3. Colab/Jupyter mode: use async bot (AsyncLinkedInBot).

Environment Variables

Hosted:

• LINKEDIN_AUTOBOT_BASE_URL
• LINKEDIN_AUTOBOT_API_KEY (optional if your backend does not require it)

Local bot:

• LINKEDIN_EMAIL
• LINKEDIN_PASSWORD
• GMAIL_APP_PASSWORD (optional, for automatic 2FA email verification handling)

Analytics (optional):

• GOOGLE_API_KEY

2FA Email Verification Handling

When LinkedIn requires email verification (common in automated login scenarios), the bot can automatically fetch and submit the verification code from your Gmail inbox.

Setup:

1. Enable 2-Step Verification on your Google account: https://myaccount.google.com/security
2. Generate a Gmail App Password: https://myaccount.google.com/apppasswords
3. Set GMAIL_APP_PASSWORD environment variable or pass it to LinkedInBot:

with LinkedInBot.from_env(headless=False) as bot:
    # Automatically handles 2FA if LinkedIn requires verification
    bot.login()

How it works:

• Detects LinkedIn's verification challenge
• Connects to Gmail IMAP
• Fetches the 6-digit code from LinkedIn's verification email
• Automatically enters the code and continues login

Requirements:

• Gmail account (same as your LinkedIn email)
• 2-Step Verification enabled on Google account
• Gmail App Password (from https://myaccount.google.com/apppasswords)
• Gmail IMAP access enabled

Tutorial: Hosted Client

1) Create client with from_env(...)

from linkedin_autobot import LinkedInAutobotClient

client = LinkedInAutobotClient.from_env()

2) Collect profile by name with collect_profile_by_name(name)

profile = client.collect_profile_by_name("Satya Nadella")
print(profile.profile_url, profile.full_name, profile.title)

Returned data fields:

• profile.profile_url
• profile.full_name
• profile.title
• profile.location
• profile.about
• profile.experience
• profile.status
• profile.metadata

3) Collect profile by URL with collect_profile_by_url(profile_url)

profile = client.collect_profile_by_url("https://www.linkedin.com/in/satyanadella/")
print(profile.full_name, profile.location)

This method returns the same LinkedInProfile fields listed above.

4) Analyze profile with analyze_profile(profile, prompt)

analysis = client.analyze_profile(
    profile,
    prompt="Score this lead and suggest the best outreach angle."
)
print(analysis.score, analysis.decision_maker_level)

5) Send outreach with send_connection_request(profile_url, message=None)

sent = client.send_connection_request(
    "https://www.linkedin.com/in/satyanadella/",
    message="Hi Satya, I would love to connect."
)
print(sent)

Tutorial: Local Bot

1) Install local requirements, then create bot with from_env(...) and login with login()

from linkedin_autobot import LinkedInBot

with LinkedInBot.from_env(headless=False) as bot:
    bot.login()

Tutorial: Colab / Jupyter (Async)

In notebook environments, use AsyncLinkedInBot (not LinkedInBot sync API).

import asyncio
from linkedin_autobot import AsyncLinkedInBot

async def run():
    async with AsyncLinkedInBot.from_env(headless=True) as bot:
        await bot.login()
        profile = await bot.collect_profile_by_name("Satya Nadella")
        print(profile.profile_url, profile.full_name)

await run()

2) Open a profile with visit_profile(profile_url)

with LinkedInBot.from_env(headless=False) as bot:
    bot.login()
    bot.visit_profile("https://www.linkedin.com/in/satyanadella/")

3) Find profile URL with find_profile_url_by_name(name)

with LinkedInBot.from_env(headless=False) as bot:
    bot.login()
    profile_url = bot.find_profile_url_by_name("Satya Nadella")
    print(profile_url)

4) Collect profile with collect_profile(profile_url)

with LinkedInBot.from_env(headless=False) as bot:
    bot.login()
    profile = bot.collect_profile("https://www.linkedin.com/in/satyanadella/")
    print(profile.full_name, profile.title)

Returned data fields:

• profile.profile_url
• profile.full_name
• profile.title
• profile.location
• profile.about
• profile.experience
• profile.status
• profile.metadata

5) Collect by name with collect_profile_by_name(name)

with LinkedInBot.from_env(headless=False) as bot:
    bot.login()
    profile = bot.collect_profile_by_name("Satya Nadella")
    print(profile.profile_url)

This method returns the same LinkedInProfile fields listed above.

6) Send request with send_connection_request(profile_url, message=None)

with LinkedInBot.from_env(headless=False) as bot:
    bot.login()
    sent = bot.send_connection_request(
        "https://www.linkedin.com/in/satyanadella/",
        message="Hi Satya, let's connect."
    )
    print(sent)

7) Close bot manually with close()

from linkedin_autobot import LinkedInBot

bot = LinkedInBot.from_env(headless=False)
bot.login()
# ... your actions ...
bot.close()

Tutorial: Local Analytics Engine

Analyze with LeadAnalyticsEngine.analyze(profile, prompt)

from linkedin_autobot import LeadAnalyticsEngine, LinkedInProfile

profile = LinkedInProfile(
    profile_url="https://www.linkedin.com/in/example/",
    full_name="Jane Founder",
    title="Founder & CEO",
    about="B2B growth and sales.",
)

engine = LeadAnalyticsEngine()  # uses GOOGLE_API_KEY if available
result = engine.analyze(profile, "Score this lead and propose outreach.")
print(result.score, result.insights_strategic)

Tutorial: Django Helpers

profile_from_lead_model(lead)

from linkedin_autobot.django import profile_from_lead_model

profile = profile_from_lead_model(lead_instance)

analyze_lead_model(lead, prompt, api_key=None)

from linkedin_autobot.django import analyze_lead_model

analysis = analyze_lead_model(lead_instance, "Score this lead.")
print(analysis.score)

Data Models

• LinkedInCredentials(email, password)
• LinkedInProfile(...)
• AnalysisResult(...)

Exceptions

• LinkedinAutobotError
• MissingDependencyError
• AuthenticationError
• ScrapingError

Practical Notes

• LinkedInBot.login() performs preflight checks and raises clear instructions when Chromium is missing.
• If local browser automation is blocked by OS permissions, use hosted mode.
• For production-grade stability, hosted mode is recommended.