🛡️ Security & anti-fraud

S1 →Uses Jev to assess hate speech and spam in messages; the bot executes deletion, review, and timeout rules.

S1 →Breaks messages into multiple spam and contextual-relevance questions; the application combines thresholds, exemptions, and deterministic rules to decide what to delete.

S1 →Separately assesses whether messages are spam and whether they involve scams or phishing; configuration determines whether the program records the result or moves them to a review folder.

S1 →Jev assesses message risk and multiple moderation criteria; the bot acts according to configured policies.

S1 →Optionally returns a spam probability. Code combines it with other detection results using thresholds and a veto mode to decide whether to act on a message.