Japella's architecture splits out the connections to different protocols (eg Telegram, Discord) into "adaptor" services, and all bot functionality is implemented in a separate service, meaning that the chatbot functionality can easily work with new protocols, by just creating a new adaptor, or, can relay chat messages across protocols, for example.
Japella is only distributed as a x86_64 Linux container, and requires a RabbitMQ server as the message queue.
There is ready-to-go docker-compose.yml file here: https://jamesread.github.io/Japella/installation/docker-compose/
- All code and assets are Open Source (AGPL).
- No company is paying for development, there is no paid-for support from the developers.
- No separate core and premium version, no plus/pro version or paid-for extra features.
- No SaaS service or "special cloud version".
- No "anonymous data collection", usage tracking, user tracking, telemetry or email address collection.
- No requests for reviews in any "app store" or feedback surveys.
- No prompts to "upgrade to the latest version".
- No internet-connection required for any functionality.
Documentation for how to install and use Faridoon is available at https://jamesread.github.io/Japella/