Read-only Mode (Early Preview)

Launch Oxy from anywhere without requiring an existing project folder:

• Zero-Setup Deployment - Start fresh with oxy serve --readonly from any directory
• GitHub Repository Integration - Connect to your GitHub repository during initial setup to bootstrap your project
• Streamlined Onboarding - Guided workflow to get started without manual project configuration

Note: This is an early preview feature and may encounter some edge cases during initial use. Starting the server without --readonly will not show any changes in behavior for now

Secret Management System and Github Settings

Secure handling of sensitive configuration and credentials:

• Encryption Master Key Support - Provide your own encryption key or let Oxy generate one automatically
• Secure Storage - User secrets like Github tokens or OPENAI API KEYs are encrypted and stored safely within the deployment
• GitHub Settings Management - Configure GitHub repository sync and manage connection settings for seamless project integration

Enhanced Streaming Controls

Improved control over AI conversations:

• Stream Cancellation - Stop ongoing AI responses mid-stream with dedicated cancel functionality
• Responsive Interface - Immediate feedback when canceling operations for better user control

Bug Fixes & Improvements

Fixed minor type conversion issues in the unified semantic layer and updated dependencies for improved stability

Entity-First Unified Semantic Layer

Introducing semantics.yml - a comprehensive approach to data modeling that unifies all entity definitions across your entire Oxy deployment:

• Automatic Semantic Collection - Run oxy sync to bootstrap unified semantic definitions from all your models, then customize as needed
• Agent Integration - Reference semantic types directly within agents or workflows using Jinja syntax with dimensions object, e.g., {{ dimensions.month }}
• Workflow Variables with JSON Schema - Implement semantic type referencing as JSON Schema within workflow variables for structured validation
• Centralized Entity Management - Entire semantic layer consolidated in semantics.yml for consistency across your deployment

Note: This is an early release feature and may encounter some edge cases during initial use.

Example of semantics.yml, which you can find in our public source code

User Management & Admin Controls

Complete user administration system with role-based access control:

• Admin Configuration - Define administrators directly in config.yml with simple email list
• User Management Interface - Access full user management at /users endpoint
• User Lifecycle Management - Admins can delete and restore user accounts
• Role-Based Authorization - Clear separation between admin and standard user capabilities

Example config.yml, which you can find in our public source code 

Users can then be managed at <your-deployment-url>/users

Revamped Management Interface

Database management and API Keys management UIs are slightly reworked to follow a unified design language with consistent styling and interaction patterns

Enhanced Chat Streaming Experience

Conversations with agents now persist seamlessly across page navigation and reloads, with improved first message flow for smoother user experience

Document update

Added more endpoints to the API reference. Visit <your-deployment-url>/apidoc for a closer look at our latest Openapi documentation

Other Improvements

Updated to Rust 1.88 (latest version), resolved ClickHouse stability issues, and improved artifact panel scrolling behavior. Enhanced dependency management for better overall platform reliability

New Authentication System

We've implemented authentication with multiple new options to fit your workflow:

• Built-in Authentication - Self-contained email/password and Google OAuth with zero external dependencies - perfect for quick deployments
• AWS Cognito Integration - Enterprise authentication with Application Load Balancer integration
• Google Identity-Aware Proxy (IAP) - Zero-trust security with Google Load Balancer integration
• Serverless IAP for Cloud Run - Optimized authentication for Google Cloud serverless environments

For more information, please visit our Authentication document 

API doc & API Keys Support

Oxy now allows programmatic access for integrations and automation using API keys - perfect for headless operations and third-party integrations

An API doc is available at your-deployment-url/apidoc while the API key is available under your-deployment-url/api-keys

Enhanced Agent Capabilities

Our AI agents are now more powerful and reliable:

• Routing Agents - Smart dispatchers that automatically direct tasks to the right specialized agents, workflows, or SQL queries using semantic similarity matching. Features built-in result synthesis for coherent responses across multi-agent systems.
• Artifacts - Better handling of AI-generated content by highlighting code blocks, queries or workflows that are utilized as source of inference for answer
• Multiple messages in one thread, with memory - Evolved from one shot simple ask-and-answer interactions to true conversations that remember context and build on previous exchanges

To set up routing agents, please visit our routing agents documentation

Below is a simple example of a routing agent 

Advanced Database Sync & Build System

Streamline your data workflows with powerful new sync capabilities:

• Database Sync/Build API - Previously CLI-only operations now available through API for programmatic automation and integrations
• Oxy Sync & Oxy Build Frontend Integration - Run sync and build operations directly from the UI instead of switching to command line
• Column and Table Description Sync - Enhanced sync operations now include column descriptions and table documentation alongside schema data
• Skipped and Overwritten Files Tracking - Clear visibility into which files were updated, skipped, or replaced during sync operations

From the UI, you can start viewing databases and perform operation like sync and build similarly to the CLI commands at <your-deployment-url>/databases 

Improved Developer Experience

Enhanced IDE and file management capabilities:

• Complete File Management in IDE - Create, edit, rename, and remove files directly in the IDE
• Enhanced Workflow Diagrams - Better visualization of your data workflows
• Query Result CSV Download - Export query results for external analysis

Better Thread and Chat Management

Significantly improved conversation handling:

• Paginated Thread view- Handle large numbers of conversations efficiently
• Unified Thread System - All tasks like running workflows & building data app can now use a consistent thread-based approach
• Efficient Thread Management - Easily review and remove threads 
• Enhanced Data Visualization - Improved charting capabilities, better complex data structure display, and smoother loading states for a superior data exploration experience
• Token Counter - Added token usage tracking to help monitor and optimize AI model usage.

Other Performance & Platform Improvements

Enhanced platform stability with improved data fetching, strengthened database connections, and better chat streaming architecture. We also implemented comprehensive logging and error handling improvements behind the scene to gracefully handle edge cases and large data loading scenarios

Latest long term support version of Node and Rust is used with the most sensibly updated sets of dependencies to ensure backward compatibility and security

There is also a new Docker Release for easy access to the latest container images.

Documentation Updates

• Reference Architecture Guide - Comprehensive best practices for production deployments on AWS
• Step-by-Step Deployment Guide - Complete instructions to get Oxy running in your environment
• Hands-On Tutorial - Interactive guide to learn Oxy through practical examples

This release adds native support for data applications. It includes an open, declarative framework for building data applications, and an IDE for Oxy to code these applications with the given framework. We also added dark mode to make our product experience easier on the eyes. 

This release includes native visualization for the Oxy GUI experience. Further, agents can now natively execute workflows. Google BigQuery support has been added as a native data source.

This release includes a built-in MCP-compliant server, enabling seamless context management and integration with clients that support the MCP specification. Ideal for LLM orchestration and centralized context management.MCP server supports two transport options: server sent events and stdio 
Please visit our documentation for detailed instruction: https://docs.oxy.tech/learn-about-oxy/mcp-usage#setting-up-the-mcp-server

We’ve launched oxy serve to make it easier for you to interact with your agents. Now you can:

1. Seamlessly select specific agents for your queries
2. Access your conversation history in the ‘Threads’ section

Try this new feature by running oxy serve from your command line.