Threatstream: The private channel for mods to categorize local GrAEs

[DecentralizedDan]

There will be a private channel populated by Bouncer that contains the following:

• pURLs we detect
• oURLs we detect
• oURLs reported by users
• oURLs "translated" by users as per #106
• pURLs and oURLs sent from other related servers
  • The servers they get threat data from MUST be related!
  • This ensures that they are mutually protecting each other from related threats
  • It prevents the feeling of "We're doing unpaid labor for Interlock"
  • Instead, the vibe is "We're protecting our community together"
• It will also eventually take events from the Telegram Bouncer bot, like in #115
• ...and more???

Here's what will be posted in the Threatstream:

• message author ID
• channel name
• flagged by
• message
  • Link to the message, too, so mods can see the context it was posted in
  • If the message is > 200 N characters, show 1/2 N characters before and after the GrAE that triggered it

Mods will vote up or down on whether these items are safe or not

• Vote them safe and protect them, as per #123
• Voting them unsafe will delete them
  • This action will be logged and posted to the logging channel, naturally

Items voted safe and unsafe will be sent BE, along with user ID and mod ID, to protect other servers where these come up

• This is related to #100 and #102

[DecentralizedDan]

Flagging Spam for Threatstream Process

flowchart TD
START([OP posts]) --> A(User flags post)
A --> AA{Is message\nprotected?}
AA --> |Yes|AAA([Ignored])
AA --> |No|B(Bouncer prompts user to\ncut-and-paste the URL)
B --> |Escape|BAILS([User bails])
B --> C{Was the string\nin the message?}
C --> |Yes|E{Is the content a GrAE?\nURL, wallet ID, email...}
E --> |No|D("Bouncer prompts for #quot;translation#quot;")
D --> |Escape|BAILS
C --> |No|B
D --> E{Is the string a URL?}
E --> |Yes|THREAT(Data sent to Threatstream:\nOP, Channel, Reporter, URL, Message)
THREAT --> EE>Logged]
EE --> H{Mod agrees?}
H --> |Yes|BE(Data sent BE)
BE --> DELETED(Message deleted)
DELETED --> DM([OP is DM'ed])
DELETED --> G>Logged]
H --> |No|I(Bouncer posts that\nmessage is permitted)
I --> G
I --> END([Message is protected])