Is there any built-in chatbot or conversational assistant available in Discourse?

Hi everyone,

We’re exploring ways to improve the user support experience in our Discourse instance and wanted to understand whether Discourse provides any built-in chatbot or conversational assistant functionality.

Specifically, we’re looking for something that can:

  • Answer users’ questions about forum features and navigation.
  • Guide users to the appropriate categories or topics.
  • Help users understand how to use different Discourse functionalities.
  • Provide contextual assistance directly within the forum.

We are aware of Discobot, but from our understanding it is primarily intended for onboarding and interactive tutorials. We’re wondering if there is any built-in chatbot functionality beyond Discobot.

If anyone has implemented something similar, we’d appreciate hearing about your approach and any recommendations.

Thanks in advance!

See Discourse AI - AI bot which uses personal messages to communicate.

For actual chat:

  • There’s @merefield 's excellent Chatbot plugin

  • Also, we’ve built an external service that connects to the Discourse API and handles chats, then at the backend it uses an LLM and can do RAG on content in selected private or public categories. You can see it in action on for instance https://swapd.co/

All of these features are available through configuring Discourse AI, which is included by default on new Discourse installs…

You can also enable a dedicated page for AI conversations using the plugin, which is how we power ask.discourse.com — there’s more documentation available in Discourse AI and in Topics tagged ai

We have our own self hosted API key of claude and GPT LLMs.. How can we configure this LLMs in our discourse AI. And also in the screenshot what we should add in the provider options if it is self hosted.

can you help me with this??

Good news! I kindly suggest you to search and read, all what you are looking for is already here :slight_smile:

If you’re self-hosting, the configuration depends entirely on how you’re doing it and, specifically, on what you’re hosting; setting up Litellm is different from Ollama, Qwen or HuggingFace direct interferences.

If your endpoint is Open API compatible just choose Open AI and set your credentials / endpoint. You can saw the default LLMs created for a guide and/or check the specific documentation for the models and your proxy/provider to avoid guessing your settings.

You have a paid service level agreement?

If you become a business sponsor on GitHub I’ll be happy to assist you with Chatbot.

If you need urgent professional help, also consider Marketplace

This is our end point: https://genai-nexus.api.corpinter.net/openai/deployments/gpt-5.5/chat/completions?api-version=2024-10-21

and this is our model : gpt-5.5

Can you help in configuring it?