Version 1.17.0

What’s 🆕 ?

Guardrails

We're excited to introduce a new capability in our platform stack: Guardrails for our application. These mechanisms are designed to ensure that Large Language Models (LLMs) generate safe, accurate, and appropriate content. We are starting with three rules and plan to expand further, so stay tuned! You're welcome to read more here (link).

New Context Types in the "AI Assistant Retrieval" App:

We've added new context types to the application: VectorDB and Knowledge.

Copy Tenant ID:

A small but significant enhancement! We've added the ability to copy the tenant ID directly from the settings component in the UI.

Improvements 🤯

Improved Splitting Mechanism in Knowledge Indexing:

We've optimized the splitting mechanism for knowledge indexing in the application chat, resulting in more efficient document retrieval. This improvement ensures that the knowledge functionality performs even better now.

Bug Fixes🐛

Resolved Issue in Advanced Agent App:

We've fixed a problem concerning the use of two tools of the same type within the app