Filtering Modes

Filtering Mode: Determines What to Scan For

1. Auto-Smart Mode

• The Recommended Mode: Hand crafted filters designed to catch every category of scammer/spammer I become aware of
• Designed to have an extremely low false positive rate. Any time I come across even a single false positive, I attempt to make adjustments to avoid that condition.
• Identifies wide range of scammer types, such as WhatsApp scammers and 'sex bot' scammers
• Detects impersonators
• Detects scam links (when new scam domains are found being used in spam comments, I add them to a list file on my 'YT-Spam-Domains-List' repository, so the app can automatically fetch the latest scam domains)
• Also uses multi-part scoring algorithm to detect many elusive spammer types that don't have unique characteristics

2. Sensitive-Smart Mode

• Uses the same checks as Auto-Smart Mode, but certain filtering criteria are more sensitive
• Good for detecting more elusive spammers, but will probably result in more false positives (anywhere from only a few more, to sometimes WAY more)

3. Specific Channel ID / Link

4. Within Username

• Specific Special Characters: Emojis, Unicode Symbols, etc
• Strings: Words, Sentences, Phrases, etc
• Custom Regex Pattern

5. Within Comment Text

• Specific Special Characters: Emojis, Unicode Symbols, etc
• Strings: Words, Sentences, Phrases, etc
• Custom Regex Pattern

6. Within Username and Comment Text

• Specific Special Characters: Emojis, Unicode Symbols, etc
• Strings: Words, Sentences, Phrases, etc
• Custom Regex Pattern

7. ASCII Mode (Within Usernames)

1. Allow Standard + Extended ASCII
2. Allow Standard ASCII Only
3. NUKE MODE: Allow ONLY Numbers, Letters, and Spaces