Beyond the external CAS list, Telm runs its own cross-group intelligence: the same spam surfacing across many protected groups is recognised as a coordinated campaign, and a confirmed spammer can be banned in every group they are in at once. It is deliberately conservative, fully logged and reversible. Available on Basic and above; on per group by default.
Want to set this up? Do it right in your Telm account.
Open in your dashboard1Our own cross-group shield
The external CAS blocklist catches spammers already known elsewhere. The global spam database is Telm's own layer: it learns from every group we protect and reacts to threats CAS has not seen yet.
It has two parts working together — a cross-group campaign detector and federated bans.
2Coordinated-campaign detection
Spammers rarely hit just one group. The same or nearly identical text often appears across many communities at once, or from several fresh accounts inside a single group in a short window.
Telm watches for exactly this. When the same content surfaces across a number of protected groups within a day, it is treated as a coordinated blast and removed on sight. Inside one group, the same content from several different accounts in a short window is itself grounds to ban the authors and clean up copies that already landed.
3Federated bans — ban once, gone everywhere
When the same account posts the same confirmed spam in more than one of your groups within a short window, Telm treats it as a federated spammer and bans it across every group where it is present — not just the one where it was caught.
Guards at the entrance and on each message make sure a federated spammer cannot re-enter or post in any of your protected groups. Everything is recorded, and an admin can revoke a federated ban (which unbans the account everywhere) at any time.
- Triggered by the same user posting the same spam across multiple groups in a short window.
- Bans the account everywhere it is present, not just where it was first caught.
- Entry and message guards block re-entry and re-posting.
- Fully logged and reversible by an admin from the dashboard.
4How much proof it needs
Not every spam action is treated the same. The clearest, most obvious abuse spreads fastest, while anything borderline has to show up in more groups first — so the strongest global action only fires on the most obvious spam.
- Clear, high-confidence spam that was actioned in several groups spreads the fastest.
- Borderline cases must repeat across more groups before they spread, which avoids over-reaction.
- Only automatic spam decisions qualify; your custom rules, trigger commands and manual actions never spread to other groups.
5Safeguards
Federated action is deliberately conservative so it only fires on genuinely coordinated abuse.
- Only confident, automated spam verdicts qualify — trigger commands, custom rules and simple CAS hits do not spread.
- It requires the same spam across multiple groups within a short window, not a single sighting.
- Every federated ban is logged and can be revoked by an admin, which lifts it everywhere at once.
- You can turn the feature off per group at any time.
6Turning it on
The global spam database is enabled per group and is on by default on eligible plans, right next to the CAS toggle. Only strong, automated spam verdicts feed federated bans — your custom rules and manual actions do not, to avoid spreading a local decision globally.
7How it complements CAS
CAS and the global spam database are not competitors — they cover different moments in a spammer's life, and running both is what closes the gap.
- CAS knows spammers already reported across all of Telegram; it stops them at the door on any plan.
- The global spam database reacts to fresh campaigns hitting your protected groups before they ever reach an external list.
- Together they cover both the already-known abuser and the brand-new coordinated blast.