How to Automate Search with Python 3 + Selenium 4 via GitHub Actions

Table of Contents

This post is about how to automate search actions with Python 3 + Selenium 4 on GitHub Actions.

If you are using Meilisearch Cloud with its free plan, you will receive an email notification of project deletion after 3 days of inactivity(*). It's possible that the search system on your website is not used that often. An automate search must be useful to prevent the project from being deleted.

*You still have 58 hours after receiving the warning email. The actual minimum required operating interval should be 1 time per 3 days + 58 hours = 130 hours (5.42 days).


  • Python 3.11
  • Selenium 4


The folder structure is as follows;

├─ .github/
|  └─ workflows/
|     └─ run.yml
└─ requirements.txt

Workflow Overview

Set the input element id as id="searchInput".

  • Run on (1 + 3n)th day at 0:00 UTC
  • Exit the program after entering "a" in the search input element
  • Describe dependencies to requirements.txt

Python program itself

  • Specify User-Agent to bypass anti-bot server protection


name: Run Python script on schedule
    - cron: "0 15 1,4,7,10,13,16,19,22,25,28,31 * *"
  # workflow_dispatch: <- Add it for manual execution

    runs-on: ubuntu-latest
      - name: Check out repository code
        uses: actions/checkout@v3
      - name: Set up Python environment
        uses: actions/setup-python@v4
          python-version: 3.11
      - name: Cache pip dependencies
        uses: actions/cache@v3
          path: ~/.cache/pip
          key: ${{ runner.os }}-pip-${{ hashFiles('**/requirements.txt') }}
          restore-keys: |
            ${{ runner.os }}-pip-
      - name: Install dependencies
        run: |
          pip install get-chrome-driver --upgrade
          pip install -r requirements.txt
      - name: Run Python script
        run: python
from selenium import webdriver
from import By
from import WebDriverWait
from import expected_conditions as EC
import time
import random

# User-Agent lists
user_agents = [
  "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36 Edge/B08C3901",
  "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36 Edge/16.16299"

# Randomly select an user agent
user_agent = random.choice(user_agents)

# webdriver
options = webdriver.ChromeOptions()
options.add_argument(f'user-agent={user_agent}') # add the User-Agent
driver = webdriver.Chrome(options=options)

# If element not found, wait 5 seconds
# Wait 5 seconds for element to take effect
wait = WebDriverWait(driver, 5)

# The site to search

# Get id="searchInput" and input "a"
input_element = wait.until(EC.presence_of_element_located((By.ID, "searchInput")))

# Close webdriver

Code Description

Anti-Bot Measures

The server you are using may protect bot access with its WAF. To prevent this, I added User-Agent to the header. Without it, Cloudflare rejected the program.

It might be better to update the versions in the code when a browser version is updated, but the program worked even if the version is not the same as the current version in real.

Operation Interval

If you use this program for Meilisearch Cloud free plan, you still have 58 hours after a warning email. In this program, I run it on (1 + 3n)th day because I don't want to receive their email.

GitHub Actions offers 2,000 minutes of free usage per month for private repositories (as of June 2023). This program takes about 30 seconds to run, which would be a total of 5 minutes per month.

Link About billing for GitHub Actions - GitHub Docs

I added lines to cache the dependencies, but it only saves a few seconds.

If you want to save execution time, reduce the date or wait time.