The anti-ban system has multiple layers that work together to make the bot look like a real human user:
Session Warmup (7 days)
New sessions are limited to 15 messages per day. Each day, the limit increases until it reaches 200 messages/day after 7 days. This prevents WhatsApp from flagging a sudden spike in activity from a new linked device.
Human-Like Timing
Read-But-Skip
15% of the time in group chats, the bot reads a message but does not respond. Real people do not reply to every message in a group.
Message Variation
Each command has 50-100 different response templates. Variables like {name}, {time}, and {date} are injected dynamically. Occasional typos and casual phrasing are added. Zero-width characters ensure every message has a unique byte fingerprint.
Media Fingerprint Jittering
Random bytes are appended to stickers and images so each file has a unique hash. WhatsApp cannot detect them as bot-generated copies.
Activity Hours
The bot is quieter between 12am-6am (local time). Responses are shorter and slower during these hours, mimicking real sleep patterns.
Daily Cap
Hard limit of 200 messages per day per session. Combined with warmup, this keeps activity within normal human usage patterns.