Remote control your Windows computer using Telegram
Updated on 23/01/2023
If you work in an office, you are probably told to always lock your computer before you go AFK as a measure of security. In my case, my colleagues will not hesitate to prank whoever forgets his screen unlocked as a punishment.
I am a very caring person, so things like these should not happen to me. I must find a way to lock my screen from a distance and preferably via my smartphone.
To do that, we are going to use python to build our own Telegram bot to control our Windows PC.
python-telegram-bot is an awesome interface for the Telegram Bot API that enables us to do so much cool stuff using telegram.ext submodule.
Final product
The final result looks like the image below: the /start command pops up the several buttons that you can tap to perform several tasks such as checking the screen status and locking it. At first, that was the main purpose of the bot, but well.. things escalated quickly.
This bot is also capable of taking screenshots, paste the clipboard, list and kill running processes, navigate file system as well execute shell commands.
Step 1: Get the API token and the chat ID
Creating a Telegram bot using BotFather is a simple process. Here are the steps to create a Telegram bot using BotFather:
- Open Telegram and search for @BotFather.
- Start a chat with BotFather by clicking on the search result.
- Send the command ‘/newbot’ to BotFather to create a new bot.
- BotFather will ask you to choose a name for your bot. Pick a name that you like and send it to BotFather.
- BotFather will then ask you to choose a username for your bot. The username must end in “bot” (e.g., mynewbot) and should be unique.
- Once you have chosen a username, BotFather will give you a token. This token is used to authenticate your bot with the Telegram servers. Keep this token safe, as you will need it later.
In order to get your Telegram chat ID, you will need to do the following:
- Start a chat with the bot @RawDataBot.
- Send the command /start to the bot.
- The bot should respond with your Telegram informations. Grab the chat ID that you need.
Step2: Get coding
Now we get to the fun part. First, we need to import all the necessary libraries
from telegram.ext import *
from telegram import KeyboardButton, ReplyKeyboardMarkup
from mss import mss # To to capture screenshots
import tempfile
import os
import psutil # To list and kill running processes
import ctypes # To lock the screen
import webbrowser # To to open URLs in the webbrowser of the computer from the phone
import pyperclip # To access the clipboard
import subprocess # To execute shell commands
import json
Now let’s define our telegram bot class. Of course, this can be done without using a class, but I like to keep things organized. I stored the chat id and the access token in auth.json file so I can access it in the __init__ method.
class telegramBOT:
def __init__(self):
f = open('auth.json')
auth = json.load(f)
self.TOKEN = auth["TOKEN"]
self.CHAT_ID = auth["CHAT_ID"]
Next we define start_command() method that will handle /start command typed by the user in the mobile app. A menu appears containing multiple buttons to command the bot.
def start_command(self, update, context):
buttons = [[KeyboardButton("⚠ Screen status")], [KeyboardButton("🔒 Lock screen")], [KeyboardButton("📸 Take screenshot")],
[KeyboardButton("✂ Paste clipboard")], [KeyboardButton(
"📄 List process")], [KeyboardButton("💤 Sleep")],
[KeyboardButton("💡 More commands")]]
context.bot.send_message(
chat_id=self.CHAT_ID, text="I will do what you command.", reply_markup=ReplyKeyboardMarkup(buttons))
We implement now handle_message() method. When the user taps a button or type a command, this method will handle it and execute the appropriate task.
def handle_message(self, update, input_text):
usr_msg = input_text.split()
if input_text == 'screen status':
for proc in psutil.process_iter():
if (proc.name() == "LogonUI.exe"):
return 'Screen is Locked'
return 'Screen is Unlocked'
if input_text == 'lock screen':
try:
ctypes.windll.user32.LockWorkStation()
return "Screen locked successfully"
except:
return "Error while locking screen"
We can check the status of the screen by searching for the logonUI.exe in the running processes. If it exists, then the screen is locked. ctypes enable us to lock the screen without any problem.
I think taking screenshots is a useful feature to have. Here we define take_screenshot() method to handle this task.
if input_text == "take screenshot":
update.message.bot.send_photo(
chat_id=self.CHAT_ID, photo=open(self.take_screenshot(), 'rb'))
return None
def take_screenshot(self):
TEMPDIR = tempfile.gettempdir()
os.chdir(TEMPDIR)
with mss() as sct:
sct.shot(mon=-1)
return os.path.join(TEMPDIR, 'monitor-0.png')
Getting the clipboard content is pretty straightforward with the pyperclip module.
if input_text == "paste clipboard":
return pyperclip.paste()
Here we handle the part of process listing and killing. To kill a process a user must manually type the keyword kill proceeded by the name of the process (capitalization matters).
if input_text == "list process":
try:
proc_list = []
for proc in psutil.process_iter():
if proc.name() not in proc_list:
proc_list.append(proc.name())
processes = "\n".join(proc_list)
except (psutil.NoSuchProcess, psutil.AccessDenied, psutil.ZombieProcess):
pass
return processes
if usr_msg[0] == 'kill':
proc_list = []
for proc in psutil.process_iter():
p = proc_list.append([proc.name(), str(proc.pid)])
try:
for p in proc_list:
if p[0] == usr_msg[1]:
psutil.Process(int(p[1])).terminate()
return 'Process terminated successfully'
except:
return 'Error occured while killing the process'
We can also open URLs using the command url proceeded by the URL, navigate the file system using cd <path> and download files from our computer with download <name_of_the_file>.
if usr_msg[0] == 'url':
try:
webbrowser.open(usr_msg[1])
return 'Link opened successfully'
except:
return 'Error occured while opening link'
if usr_msg[0] == "cd":
if usr_msg[1]:
try:
os.chdir(usr_msg[1])
except:
return "Directory not found !"
res = os.getcwd()
if res:
return res
if usr_msg[0] == "download":
if usr_msg[1]:
if os.path.exists(usr_msg[1]):
try:
document = open(usr_msg[1], 'rb')
update.message.bot.send_document(
self.CHAT_ID, document)
except:
return "Something went wrong !"
The subprocess module enables us to run shell commands on our computer. We just need to type cmd followed by the command we want to execute and the its output will be received.
if usr_msg[0] == "cmd":
res = subprocess.Popen(
usr_msg[1:], shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.DEVNULL)
stdout = res.stdout.read().decode("utf-8", 'ignore').strip()
stderr = res.stderr.read().decode("utf-8", 'ignore').strip()
if stdout:
return (stdout)
elif stderr:
return (stderr)
else:
return ''
We can put the computer to sleep, but we will lose our control on it.
if input_text == "sleep":
try:
os.system("rundll32.exe powrprof.dll,SetSuspendState 0,1,0")
return "Windows was put to sleep"
except:
return "Cannot put Windows to sleep"
We define the send_response() method to further process the data sent to the user. The limit of characters we can send with send_message method is 4096 characters, so we need to split larger amount of data into chunks and send them separately.
We also limit the use of this bot to only one account username (your account) for security reasons. So you need to change YOUR_USERNAME to your account username.
def send_response(self, update, context):
user_message = update.message.text
# Please modify this
if update.message.chat["username"] != "YOUR_USERNAME":
print("[!] " + update.message.chat["username"] +
' tried to use this bot')
context.bot.send_message(
chat_id=self.CHAT_ID, text="Nothing to see here.")
else:
user_message = user_message.encode(
'ascii', 'ignore').decode('ascii').strip(' ')
user_message = user_message[0].lower() + user_message[1:]
response = self.handle_message(update, user_message)
if response:
if (len(response) > 4096):
for i in range(0, len(response), 4096):
context.bot.send_message(
chat_id=self.CHAT_ID, text=response[i:4096+i])
else:
context.bot.send_message(
chat_id=self.CHAT_ID, text=response)
It is always nice to have readable error when something goes wrong.
def error(self, update, context):
print(f"Update {update} caused error {context.error}")
And finally the main method start_bot() where we add our handlers and start our bot.
def start_bot(self):
updater = Updater(self.TOKEN, use_context=True)
dp = updater.dispatcher
dp.add_handler(CommandHandler("start", self.start_command))
dp.add_handler(MessageHandler(Filters.text, self.send_response))
dp.add_error_handler(self.error)
updater.start_polling()
print("[+] BOT has started")
updater.idle()
To start the bot we instantiate the TelegramBot class and call the start_bot() method.
bot = TelegramBot()
bot.start_bot()
The entire code is available on github here.
Final step
Once you run this python script on the computer you want to control, you need to search for the bot in the telegram app by the name you defined in the first step.
You can always add more features to this bot. Remember, you are only limited by your imagination (and your coding skills lol).
python-telgram-bot documentation here.
HAPPY HACKING.