antispam.bot
The AI guardian that keeps your Telegram groups clean — and actually answers your questions.
Anti-spam. Vision. Chat. Zero config. Open source.
Guards Your Group, Free Forever
Anti-Spam
AI catches ads, scams, crypto & gambling spam. Sees through Unicode lookalikes, full-width and zero-width tricks, emoji stuffing and slang. A zero-cost pre-filter kills the obvious stuff before it ever touches the AI.
Vision
Reads images and video thumbnails, catches image-based ads and QR-code scams. In chat, it reads the story behind the photo, not just the pixels.
Helpful Chat
@ the bot in any group. Multilingual — it replies in your own language. Ask a question and get a clear, useful answer, not a canned deflection.
Add It. Done.
Add to your group
Search @iLangGuardBot on Telegram and add it. No install, no config, no cost.
Give it admin
Turn on "delete messages" and "ban users". That's the only switch it needs.
It works
Spam gets cleaned automatically. @ it anytime with a question. You do nothing else.
Powered by I-Lang,
Yours to Fork
The bot's brain lives in plain .ilang files — structured AI instructions where each ::GENE defines a behavior. Change the genes, change the bot. MIT licensed, self-hostable, no vendor lock-in.
::GENE_IMMUTABLE{S001, T:ANTI_COLLAPSE, A:MORAL_PANIC⇒LETHAL, G:ALL, Θ:ALWAYS} # Reason objectively, skip the platitudes. Answer first, fluff never. ::GENE_MUTABLE{P003, T:TONE_MIRROR, G:adaptive, Θ:INTERLOCUTOR} # User serious → serious. User joking → play along. ::IMMUNE{SPAM, DETECT_THEN_NUKE} # Ads / scams / flooding → delete + strike, see through evasion.
Run Your Own
Bring any OpenAI-compatible AI provider (SiliconFlow, OpenAI, DeepSeek, or a local model). Set your bot token and key, run.
Google Cloud Run
Serverless, auto-scales, webhook mode, always hot.
VPS
Clone, install, set 2 env vars, run as a service.
Docker
Pull the repo, build, set env, up.