Current config:

id: 852404243
language: en
captchaType: button
timeGiven: 120
adminLocked: false
restrict: false
noChannelLinks: false
deleteEntryMessages: false
greetsUsers: true
customCaptchaMessage: false
strict: true
deleteGreetingTime: 0
banUsers: false
deleteEntryOnKick: false
cas: true
underAttack: false
noAttack: true
buttonText: Not set
allowInvitingBots: false
greetingButtons:
Not set

For example: /trust @b0g3r

I have some bots that ban people before and after they enter the group. In the last case, Shieldy still presents a welcome for them, waiting for the time I set to expire. It would be nice if Shieldy detect those bans or kicks and delete its one's messages for them.

When a user joins, Shieldy will send a captcha, if the user fails the captcha, Shieldy will delete all messages the user sent and kick him.

Except when the user sends an inline bot message, like this one :

The captcha logic still works and the user is kicked, but the inline message is not deleted

Hello, when spamer adds to group and making fast message it shows to all users and captcha doesnt work.

Yes, after timeout user where kicked or banned, but before this action spamer can make a spam.

It will be great to delete all messages of spam user with kicking after failed captcha.

It's kinda like issue #75, but with the greeting message. First, it won't evoke mention if the new user doesn't have a username. Also If we don't use $username variable in the message, $fullname (if the new user only has it) or $username (if the new user has both) appears after the custom message, isolated by a empty line from the rest of the message.

Hi,
Current config:

id: XXXXXXXXX
language: en
captchaType: button
timeGiven: 30
adminLocked: true
restrict: true
noChannelLinks: false
deleteEntryMessages: true
greetsUsers: true
customCaptchaMessage: true
strict: true
deleteGreetingTime: 3
banUsers: true
deleteEntryOnKick: true
cas: true
underAttack: false
noAttack: false
buttonText: I'm not a robot ✅
allowInvitingBots: false
greetingButtons:
Not set

Bot is of course admin in group. After new user joins group - he is forced to do Digits CAPTCHA instead of a simple button press.

Example:

(10 + 5) $username, please, send the solution to the arithmetic operation provided within the time amount specified to this group, otherwise you will be kicked. Thank you! (60 sec)

As you can see - timeGiven should be 30 seconds instead of 60, according to my bot configuration.

Or am I doing something wrong?

Also - HUGE thank you for this bot - awesome job! Keep it up, proud of You!

If I make my spam bot with the typical click bait text + URL then it will display in each group that has joined the group. Which achieves the spamming, without needing to actually post anything.
Shield would have to ban & delete these bots before the captcha phase though.

You could add buttons (like "-" and "+" and/or "sec", "min", "hour") so we can customize it in a better way. The present option is kicking users too fast before they participate. My group is a small one, so people subscribe, see that it's no active convo and go somewhere else and don't see that it's been asked them to take some action.

User is banned to blacklist even if /banUsers@shieldy_bot is set to kick only.
FYI: I use the same user to test different options.

Hello can you add french language plz?
Translation:
Digits:(Number+Number)(Name), merci d'envoyer la solution de l'équation dans le temps imparti sur ce groupe où vous serez banni.merci! (Time)
Simple: (name)merci d'envoyer un message dans le temps imparti où vous serez banni(time)
Button: (name)merci de cliquer sur le bouton dans le temps imparti où vous serez banni (time)
Thank you ;)

such as zh-CN

In addition, how do you feel about adding mathematical formulas?

For example, in Russian:

@username, пожалуйста, отправьте любое сообщение в этот чат в течение указанного времени, иначе вы будете кикнуты. Спасибо! (240 сек)

↓

@username, пожалуйста, отправьте любое сообщение в этот чат в течение 240 сек, иначе вы будете кикнуты. Спасибо!

If users are added by an admin, they should be whitelisted.

The bot stopped working.
I wrote directly and it doesnt answer.
Something went wrong.
Can you check your deploy?

My group needs to evaluate members activity periodically. So, we need some kind of members activity statistics. But we don't want to use those fancy and over-complicated stats-bot like the ones popular out there. We simply need these informations:

1. Number of total messages sent into the group within a time period.
2. Number of total members involved in the conversation within a time period.
3. A few of most active members in the group within a time period.
4. A few of most inactive members in the group within a time period.

Say, the command is /stats which would produce output like this:

Group Statistics

Within the last 30 days, there are 871 messages sent into this group by 26 members involved in the conversation of 40 registered members.

The most 10 active members are:

1. username_1 (x messages)
2. username_2 (x messages)
3. username_3 (x messages)
4. username_4 (x messages)
5. username_5 (x messages)
6. username_6 (x messages)
7. username_7 (x messages)
8. username_8 (x messages)
9. username_9 (x messages)
10. username_10 (x messages)

The most 10 inactive members are:

1. @userid_1 (x messages)
2. @userid_2 (x messages)
3. @userid_3 (x messages)
4. @userid_4 (x messages)
5. @userid_5 (x messages)
6. @userid_6 (x messages)
7. @userid_7 (x messages)
8. @userid_8 (x messages)
9. @userid_9 (x messages)
10. @userid_10 (x messages)

Any members who send less than 5 messages within last 30 days will be kicked out from this group. This group forbids silent readers. You have been warned!

Configuration:
/stat_period = 30 (days)
/stat_active = 10 (members)
/stat_inactive = 10 (members)
/stat_statement = "Any members who send less than 5 messages within last $period days will be kicked out from this group. This group forbids silent readers. You have been warned!"

Please note that the inactive members list does mention the members so they get the notification. If you like, you could set this as config as well, such as /notify_inactive.

I hope this request will be implemented. Thank you.

Most of the spam I receive is from Bots linking to other groups:

If there was a command I could activate to delete automatically any message that contains t.me/* it would reduce the amount of spam by 90%

Scenario: bot is coming to the chat, trying to send spam, cannot do that, goes away.
Shieldy is finding the bot, trying to delete it, having problems because bot is deleted by itself

Basically, Strict mode is on, Captcha is used as authentication.

• User A joins.
• A sends some link to a scam site
• Bot Does not delete message until the user is kicked

Expected Behaviour:

• Bot deletes any messags that does not have to do with the captcha until they solved the captcha.

Can you integrate the Option for kick or bann?
Can get the Option for no time Limit?

When I add html tags or surround with ** ** or __ __ the text is not changed.

It would be awesome if a web link could be associated with the confirmation button so that when it is clicked to confirm a user isn't a bot that also are taken to the web link that has been configured and attached to the button.

Would it be possible to add an option to define custom capture questions and specific questions? That way you could only let people in the group chat that know a fact only known to that specific group of people.
Thank you.

I appreciate the inclusion of the command to delete the entry message of beings that have been booted due to failing the captcha. I would like to be able to also auto-delete the removal messages (separate from the delete entry message option), as my channel is getting so inundated that the messages are disrupting the log. It's better with the entry messages removed, but I would like to be able to make it a non-event if you didn't notice while it was happening. Some people might like to leave them, which is why I think it should be a separate option.

Thank you very much for making Shieldy!

I didn't comment this on #52 to avoid polluting the issue with something possibly unrelated.

Even with /deleteEntryOnKick we are still seeing both messages (joined and removed) in our channels:

We've already tried to set the bot's permissions with as much permission as possible, but without success:

Is there anything else that we need to set or try, please?

Using $fullname variable in a custom message won't evoke a mention. Is there a entity that doesn't care if the new user hasn't a username and mention him/her always? Some people that didn't solve the captcha said they didn't notice my group has this condition, probably cause they weren't mentioned. By the way, can you make available more variables to help further customization this message, like something related to the time that user joined the group?

Hello,

I just discovered that if the bot kicks out a user, the user will be unban (or unlimited) and can rejoin the group without limitations.
To simulate, it's very simple:

• Set up the bot to kick if user doesn't solve the CAPTCHA.

• Lets someone joins the group and manually (or with another bot) ban him from the group

• after the time runs out wait few seconds and the user should be able to join the group again

I recreated this bug multiple times and it can cause conflict with other bot (like antistormbot) that ban immediately Arab users or userbot.

Thanks.

Hi @backmeupplz backmeupplz !

I'm using shieldy for my community where Japanese people are majority.
Is there anyway, you can add Japanese translation in shieldy?

I am happy to do any translation but I am non-dev and have no idea where and how to make use of your open sourecode.

Thanks!

HI
HOW CAN I TRANSLATE THIS BOT ?

Setting a greeting message with /greeting, like this one

Will result in this version being sent out

Missing the links and the markdown code formatting

Please add /challengeme command to test current options and captcha.

Bot is set to picture captcha mode and additionally /restrict and /strict.

1 min after user successfully passes captcha his text message with hashtags are removed.

That way we could also set the period we think it's best for the new user only use text in his/her message. For example, it would be nice if we could extent to a week, enough time to analyze if he/she went to spam the group or seek real help and so on.

Hi!

Because people tend to join chats just for reading, the greeting messages usually flood the chat. It would be awesome if Shieldy could remove all its greetings after a set period of time (e.g. 3 min, which is enough time for an average person to understand he/she is welcome).
Shieldy is nice, but its messages are not usually in line with actual chat topic and other discussions in the chat. It would be nicer if it could do its job under the hood as much as possible. The best technology is invisible.

Hi there,

I recently got this from the bot:

Hi there! It's Nikita, the creator of Shieldy. Over 65 000 chats added Shieldy! It is free and I sell no ads so I decided to tell you about my new project  Todorant.

Todorant is a smart todo list that uses cognitive psychology to make your brain feel pleasure by achieving goals. I created Todorant for myself — and I use it every day — but my friends asked me to give them access to Todorant. And today anyone can use Todorant:

As a website
On iOS
On Android

I outlined the main productivity rules in my post "How I launched 7 products in 1 year" and in the video-summary "How to use Todorant" (just 2 minutes long). Briefly, Todorant makes you add an exact date or month to every task, you only see one task at a time to focus on the main screen, unpleasant tasks are called frogs and should be done first thing in the morning.

I hacked my productivity with the simple rules and made the app so that you could do it as well! Todorant is completely free for the first 30 days, and you can even use mobile apps without login! No strings attached — just try Todorant as a website, on iOS or Android today. This is how you can support free Voicy that transcribes over 90 000 voice messages a day! Thank you!

— @borodutch

Why the heck would we start getting spam from the bot?

Pretty much what it says on the tin. I used the /customCaptchaMessage command, and got the prompt to respond with my new message, but then nothing happens once I reply, and the bot continues to use the default message. If I use the command again, it says it is going back to the default. I have tried messages with and without using the variables, and of three different lengths (kept making it shorter in case that was the problem).

Hi, I just installed Shieldy Bot, and I am happy with it overall- there is one command that I don't think is documented well, and might be able to use some modification, depending on how it actually works.
Does /deleteEntryMessages delete all entry messages, or only ones from the users who fail the captcha?

To avoid entering 10 or more commands and causing 20+ annoying notifications in a group, the user could set up Sheildy with 1 command. Using a proposed new config string, generated from an existing group or direct chat with Sheildy.

1. Start a chat directly with Shieldy
2. Enter all the commands as required
3. Enter the new command like /viewConfigString
4. Sheildy outputs a string of your config
5. Go to your group and enter the string as a command
6. Or if that is not directly possible, enter a new command like: /inputConfigString and wait for the response. Then enter the string when prompted.
7. After entering the the config string by method step 5 or 7, Shieldly confirms the new config is accepted.

Also, the same command could be used to take the config from one group, and use it in another. This also saves remembering each setting that was previously used. Or for sharing with others who like your settings but are not familiar with the settings.

With /restrict option, it is said that user can send only plain text. But it's OK to send stickers, and shieldy even do not ban them, and do not delete entry message (we also set this parameter and usually it works great).

It would be nice if we can send something like /info_config and see all the options set at that moment. Cause when we try to see what is set, regularly we accidentally turn on/off/change the settings we don't want to change.

Suppose we have a spammer that joins channel and immediately leaves. If /deleteEntryOnKick is set bot shows the captcha and upon timeout removes captcha and entry message, but leaving message is still on the channel.

Example of bot usernames:

Proposal:
a) first read all pending messages in the group;
b) if /deleteEntryOnKick is set and user already left then remove both entry and part messages;
d) show captcha;
e) if user leaves without solving the captcha remove captcha immediately (and entry/part messages according to /deleteEntryOnKick).

Thanks in advance.

Probably similar to #76

Now there's no wrong highlighting, just not shown

Shieldy is not answering

I have some bots that ban people before and after they enter the group. In the last case, Shieldy still presents a welcome for them, waiting for the time I set to expire. It would be nice if Shieldy detect those bans or kicks and delete its one's messages for them.

I have some bots that ban people before and after they enter the group. In the last case, Shieldy still presents a welcome for them, waiting for the time I set to expire. It would be nice if Shieldy detect those bans or kicks and delete its one's messages for them.

I had made a few tests and noticed that when I select "Digits" Captcha and then set a customCaptchaMessage, even when it's on and showing me it will use my custom message, it doesn't. It uses the default message. I have a guess on that: all the other captcha types are compound, with two messages. "Digits" is not, placing the arithmetic operation in the same message as the captcha message. Maybe forcing the operation on another message would fix that. To feel less uncomfortable for so few characters in a message, I suggest forcing them to be as big/tall as they can. By the way, this is the only option blind-friendly that also defeat bots and userbots that only click a "I'm not a bot" button, killing two birds with one stone hahaha

how can we see people who are ban and kicked or we unkick / unban

In most large groups, it's common to see lots of inactive or deleted accounts and inactive users who haven't been logged in to Telegram for a long time. These people are junks to the group and create false data of real members of the group. So, could you provide a command to remove deleted account and members who last seen is long time ago? I propose /remove_deleted_account and /remove_inactive_account accordingly.

What do you think? Is it possible to be done by a bot? Thank you.

I think the first time the user should be kicked since they might not have understood about the captcha, but the second time it'll be better to ban them