Privacy¶
This section explains what data Kwitka processes and why. It is informational; the final wording of a policy should be adjusted to your requirements.
What data is processed¶
To tell spam apart from normal communication, the bot analyses the content of messages in chats where it has been added by an administrator:
- message text;
- images (to recognise spam/ads on pictures);
- public profile data (name, bio) — to check newcomers.
Sharing with external services¶
For classification and summaries, some messages may be sent to AI providers (for example, Google Gemini, OpenRouter) and spammer-check services (CAS, LOLS). Only the minimum needed for the check is sent.
What is stored¶
| Data | Purpose |
|---|---|
| Chat settings | bot operation and configuration |
| Notes, rules, welcomes, filters | set by administrators |
| Reputation, ranks, warnings | the community system |
| Ban lists (CAS/LOLS/own) | protection from spammers |
Data is stored in a secure database (Google Firestore) and used solely for the bot’s operation. Technical logs have a limited retention period.
What the bot does not do¶
- it does not sell or share data with third parties beyond the services listed;
- it does not read private conversations outside the chats where it was added;
- it does not use messages for advertising.
The administrator’s responsibility¶
By adding the bot to their chat, the administrator agrees that members’ messages will undergo automatic moderation and, where necessary, informs their community of this.
Contacts¶
Questions about privacy and the bot’s operation — via @Kwitka_Antispam_Bot or the project channel.