Add function to identify known Twitter bots

src/config.ts

@@ -33,14 +33,6 @@ export const tweetIgnoreList = new Set<string>([
   // empty for now
 ])
 
-// ignore known bots; we don't want them endlessly replying to each other
-export const twitterUsersIgnoreList = new Set<string>([
-  '1598922281434103808', // ChatGPTBot
-  '1506967793409065000', // ReplyGPT
-  '1607692579243687936', // ChatSonicAI
-  '2308326482' // dustyplaylist
-])
-
 // Used by the author(s) for faster testing and feedback
 export const priorityUsersList = new Set<string>([
   '327034465', // transitive_bs

src/mentions.ts

@@ -4,6 +4,7 @@ import * as config from './config.js'
 import * as db from './db.js'
 import * as twitter from './twitter.js'
 import type * as types from './types.js'
+import { isKnownTwitterBot } from './twitter-known-bots.js'
 import { getTwitterUserIdMentions } from './twitter-mentions.js'
 import {
   getTweetUrl,
@@ -309,7 +310,8 @@ export async function isValidMention(
     return false
   }
 
-  if (config.twitterUsersIgnoreList.has(mention.author_id!)) {
+  // Ignore known bots; we don't want them endlessly replying to each other
+  if (isKnownTwitterBot(mention.author_id!)) {
     return false
   }
 

src/twitter-known-bots.ts

@@ -0,0 +1,107 @@
+// These are stored as case-insensitive, lowercase strings
+export const twitterKnownBots = new Set<string>(
+  [
+    'threadreaderapp',
+    'SaveToNotion',
+    'ChatGPTBot',
+    'AskDexa',
+    'PingThread',
+    'readwiseio',
+    'threader',
+    'unrollthread',
+    'ReplyGPT',
+    'ChatSonicAI',
+    'dustyplaylist',
+    'pikaso_me',
+    'RemindMe_OfThis',
+    'SaveMyVideo',
+    'QuotedReplies',
+    'poet_this',
+    'MakeItAQuote',
+    'colorize_bot',
+    'DearAssistant',
+    'WhatTheFare',
+    'wayback_exe',
+    'TayTweets',
+    'deepquestionbot',
+    'MoMARobot',
+    'phasechase',
+    'poem_exe',
+    'desires_exe',
+    'HundredZeros',
+    'dscovr_epic',
+    'MagicRealismBot',
+    'MuseumBot',
+    'TwoHeadlines',
+    'pentametron',
+    'earthquakebot',
+    '_grammar_',
+    'netflix_bot',
+    'redbox_bot',
+    'nicetipsbot',
+    'the_ephemerides',
+    'year_progress',
+    'IFindPlanets',
+    'emojimashupbot',
+    'translatorbot',
+    'MetaculusAlert',
+    'hashtagify',
+    'GooogleFactss',
+    'Timer',
+    'DownloaderBot',
+    'QuakesToday',
+    'Savevidbot',
+    'Growthoid',
+    'greatartbot',
+    'Stupidcounter',
+    'everyword',
+    'fuckeveryword',
+    'big_ben_clock',
+    'LetKanyeFinish',
+    'RedScareBot',
+    'EnjoyTheFilm',
+    'DBZNappa',
+    'Exosaurs',
+    'exoslash',
+    'BloombrgNewsish',
+    'AutoCharts',
+    'HottestStartups',
+    'metaphorminute',
+    'unchartedatlas',
+    'YesYoureRacist',
+    'YesYoureSexist',
+    'accidental575',
+    'EarthRoverBot',
+    'happened_today',
+    'anagramatron',
+    'pentametron',
+    'stealthmountain',
+    'SortingBot',
+    'flycolony',
+    'chernobylstatus',
+    'blitz_bot_test',
+    'unescobot',
+    'wahlumfrageBot',
+    'NeonaziWallets',
+    'LatencyAt',
+    'teololstoy',
+    'trumpretruth',
+    'UnrollHelper',
+    'bot4thread'
+  ].map((s) => s.toLowerCase())
+)
+
+export function isKnownTwitterBot(username: string) {
+  return twitterKnownBots.has(username.toLowerCase())
+}
+
+export function isLikelyTwitterBot(username: string) {
+  username = username.toLowerCase()
+
+  if (isKnownTwitterBot(username)) return true
+  if (/bot$/.test(username)) return true
+  if (/gpt$/.test(username)) return true
+  if (/status$/.test(username)) return true
+
+  return false
+}