What's the development planning of the AI Chatbot assistant trained on workspace data?

I am wondering what the Fibery team has planned about the AI Chatbot.

The AI Chatbot will be the most relevant feature in Fibery, if we look at the current trends in technology development and usage. I am already using LLM APIs in my Fibery workspace through scripts, but the ability to work with local Fibery workspace data is currently only possible through the current (experimental) AI Chatbot.

What do you want to do with chatbot? Any specific use cases, questions?

I think this topic of the functions and evolution of AI, especially local data-aware AI in Fibery is very important, and I will take some time to answer that question properly. In the meantime, anyone else is welcome to join the conversation with ideas.

I have a lot of entities in different DB for the video games I develop. Being able to ask questions about the current project like how many ennemies have melee attacks and can have the stunted status or how many time a Character is mentioned in the dialogues would be helpful.


The chatbot answering questions about structured data is a huge improvement which likely will accelerate developent as well as usage of Fibery. Thank you for prioritizing that.

Chatbot chat and message entities

While using the chatbot, the most practical feature needed, is the ability to automatically store the complete chat conversations as entities, or better, that chatbot messages are already implemented as Fibery entities so that conversion is not needed.

Unified system of AI chatbot chats and Comment threads

I see a good opportunity to unify or at leas align the AI chatbot chats with the experimental comment thread feature. For that to be useful the chatbot chat messages as well as thread comments would need to be exposed as entities that can be worked with like any other entity in Fibery.

Some use cases:

  1. Knowledge Management: Organize, search, and categorize valuable insights from AI chats for future reference.
  2. Task Conversion: Easily turn chat discussions into tasks or link them to projects.
  3. Collaboration Enhancement: Share, assign, or tag chatbot chat and chatbot comment entities to team members.
  4. Workflow Automation: Use chatbot interactions to trigger automated workflows or notifications.
  5. Customer Support Improvement: Track and manage customer chatbot interactions as entities for support and feedback analysis.

Another need which is pretty important in my daily use of the chatbot as well as the AI search, is the ability to filter the knowledge context / search context based on particular field values.
Basing a lot on databases only is very limiting, because as far as it concerns me and what I see from others, is that databases are used in dynamic ways. For example, a lot of databases have worflows or states, which determine whether an entity is draft or published, or minor version or major version, etc.
These are very very important distinctions.
The Chatbot context base as well as the AI search base are currently able to exclude or include databases, not dynamically. For effective results and optimal use of the AI, we need to feed it what matters.
Would you agree, and what is your plan to address this?