Thank you a lot for taking the time to write this code, and submitting this PR! :D
I am, however, unsure of the benefit of adding this filter, for these 2 reasons:

• Doesn't provide which keyword triggered the filter to the handler function, which means you need to check it in your handler function afterwards
• Usually one would just use a RegexHandler for this sort of thing. This has the added benefit of actually passing to the handler function which keyword matched.

Any of the devs feel differently?

The purpose of this filter is different. It doesn't check if a keyword exists in message, but if the message is an exact match with one of the strings in the list.

For example if you have a keyboard markup generated from a list:

buttons = ['Start', 'Settings', 'Back']
markup = ReplyKeyboardMarkup([[x] for x in buttons])
bot.send_message(chat_id, text, reply_markup=markup)

You can use MessageHandler(Filters.in_list(buttons), callback) in one of the states of ConversationHandler to filter out any replies that don't match one of the buttons and direct them to fallback handler. And you can get the string that triggered that filter from update.message.text, as it is an exact match to the one in the list.

Alright, you're right. I agree that we need much a filter. I think we should simply name it in instead of in_list since it should work on everything that the python keyword in does.

@jsmnbom It doesn't specify what attribute of an update you're actually comparing though. Shouldn't it be text_in(), caption_in(), callback_query_in() perhaps?

Yes, in is a reserved keyword. Maybe msg_in?

It might be a good idea to go with @JosXa 's suggestion then: text_in ?

I'm definitely not a fan of having list in there, since it works with more things than lists (other sequences and sets and stuff too).

@icecom-dg I'd like msg_in with parameters to control whether to check message text or caption

Review requested. Probably not the best docstring.