jev_antispam_bot

What it is

A minimal grammY Telegram bot for group administrators that asks TypeSafe's Jev model whether each group message is spam and deletes only high-confidence matches, targeting common Telegram spam patterns.

How it uses Jev

For each new or edited text message or media caption from a non-admin group member or external channel identity, the bot sends one Jev request with twenty-one spam questions plus one linkage question per recent message. If any spam signal reaches SPAM_THRESHOLD (default 0.90), Jev separately identifies the contiguous recent messages belonging to the same campaign, and the bot deletes only that linked suffix plus the current message.

Primitives:choicescore

Technique worth stealing

Use Jev's linkage question to identify a spam campaign's contiguous message suffix, then delete only that linked suffix plus the current message.

Try it

bun install && bun run check && bun start

View on GitHub