09-Oct-2026

Every Answer Shows What It Was Made With — Open an answer’s trust record to see each part that produced it — agent, ontology, skills, model, data — whether it was signed then, and whether it has changed since.

09-Oct-2026

Evidence Anyone Can Check — Export one answer or your whole estate as a single file and verify it offline, in a browser or with a small script, with no Uderia server involved.

07-Oct-2026

Agents That Change Records, Safely — Agents can update records through signed action contracts: every write is checked in one transaction, sealed with a receipt, and can be undone.

07-Oct-2026

Four Eyes on Every Write — Name a supervisor for an agent, and its writes wait for that person’s approval. The owner cannot approve their own agent’s writes.

07-Oct-2026

Act From Flows, Schedules and Dashboards — A signed action can run as a flow step, on a schedule, or again from a pinned dashboard tile, always under its own rules.

07-Oct-2026

Faster Queries, Proposed for You — The platform studies the queries your agents ran and proposes statistics or indexes. A person approves, and the gain is measured.

07-Oct-2026

One Model, Two Warehouses — A model extension can be deployed to PostgreSQL as well as Teradata, each to its own design standard, and read back to prove it.

06-Oct-2026

Grow the Model From a Question — When the model lacks a concept a question needs, the agent proposes it. You review the draft, then land it on the warehouse.

05-Oct-2026

Sharper Knowledge Graph Context — The graph brings in only what the question reaches, and follow-up questions keep the entities the conversation was about.

04-Oct-2026

A Calmer Question Box — The question box shows progress while the agent works and a receipt afterwards: verdict, time, tokens and cost.

04-Oct-2026

Answers Read Cleanly — Formatting in answers renders properly, and platform notes are clearly marked as the platform’s, never mistaken for the agent’s words.

03-Oct-2026

Create From the Workspace — Every Workspace section that can create something now has a “+”, and you can drag panels into the order you work in.

02-Oct-2026

Flows From a Shape — Pick a pattern such as pipeline, debate or pair programming, and the flow is built and staffed for you. Loops stop once a judge is satisfied.

02-Oct-2026

See Why a Query Is Slow — Explain any query and see its plan as a graph of steps, with costly patterns flagged first. Change it, explain again, and see what improved.

02-Oct-2026

Upgrades Back Up First — An update that changes the database layout takes a backup before it starts, and refuses to start if it cannot.

01-Oct-2026

Proven Teradata SQL Patterns — A library of Teradata SQL patterns, every statement run on a real Teradata before it ships. Agents open only what their own database can run.

30-Sep-2026

SQL Canvas Meets the Navigator — The SQL editor completes table and column names from your database, suggests joins, and hands a statement to the agent in one click.

29-Sep-2026

Celeris Provider — Celeris-1, a diffusion language model, is now available as an LLM provider.

28-Sep-2026

Several MCP Servers per Agent — One agent can read several data servers, such as two warehouses. Each call goes to the right server, and every step shows which one answered.

28-Sep-2026

Combine Results Across Sources — When an answer needs two systems, code joins their results and checks the join, and every combined figure traces back to each source.

28-Sep-2026

Ask Across Several Databases — Query PostgreSQL, MySQL, SQLite, DuckDB and MongoDB in one conversation, read-only, with every statement checked in that database’s own dialect first.

23-Sep-2026

Every Number Has a Receipt — Figures are computed by code, not by the model’s own arithmetic. Click any marked number to see the exact rows it came from.

22-Sep-2026

The Living Answer — The answer streams in place without jumping. When it is done, a short receipt shows its steps and one verdict: Verified, N findings, or Not checked.

21-Sep-2026

One Clear Layout for Every Answer — The answer comes first. Notes, evidence and actions sit beneath it in the same layout every time, each one a click away.

20-Sep-2026

Talk to It While It Works — Send a note mid-answer and the agent uses it at its next step. Close the tab, come back, and the answer is still there.

20-Sep-2026

Regenerate as an Experiment — A second answer is scored beside the first, so you choose with a reason. Editing a question starts a branch, and nothing is overwritten.

20-Sep-2026

Recovery You Can Audit — When a model provider fails, a fallback model you approved takes over and the answer says so. Every recovery is recorded.

20-Sep-2026

Leaner Prompts, Tighter Turns — Repeated prompt content is cached and paid for once. Per agent, show the model only the tools it needs and give a turn a budget it can see.

20-Sep-2026

Also Available, Per Agent — Drafts can fix their own flagged issues, attachments can be read as verified sources, and you can watch what the ontology contributed, step by step.

19-Sep-2026

Wrong SQL Stopped Before It Runs — Queries are checked against your ontology before they run — wrong joins, missing columns, impossible values — and many are fixed automatically.

16-Sep-2026

A Primer That Folds Away — A session primer can collapse to two headers, so the conversation starts at your first real question.

12-Sep-2026

The Database Navigator — Browse any connected database as a living catalog: what each table is for, its keys, likely joins and masked previews, for you and for the agent.

12-Sep-2026

Ground an Agent on a Released Ontology — An agent can read only the published version of its knowledge graph, so work on the draft never changes its answers.

10-Sep-2026

Knowledge Graphs Searched by Meaning — A question about “customers” now finds an entity called Party. On by default.

07-Sep-2026

Subscriptions That Follow the Publisher — A subscribed agent pack brings its ontologies, skills and signatures, and updates in place when the publisher republishes.

05-Sep-2026

Ontologies Validated Against the Data — A knowledge graph can be checked against the real data when it is built, so the agent gets its queries right the first time.

28-Aug-2026

White-Label Branding — One server can present several branded products, each with its own name, look and email identity, chosen by web address.

17-Aug-2026

Progressive Disclosure Skills — A skill can be a whole library that the agent opens section by section, so deep expertise is on hand without being paid for on every turn.

16-Aug-2026

Signed Data Products — Declare a data product on your ontology and the platform proves its queries are safe before it can be signed. Every answer from it carries a verifiable receipt.

08-Aug-2026

Your Vote Now Teaches the Agent — A thumbs up or down now improves the part of the agent that produced the answer, and hovering a vote tells you what it will do.

07-Aug-2026

Trust, Open to Third Parties — Partners can add their own checks and controls to the Trust Circle. Each one arrives switched off and must be signed before use.

07-Aug-2026

The Trust Panel, Rebuilt — Trust settings are now the Trust Circle itself: pick a guarantee, see what protects it, and save everything with one button.

05-Aug-2026

Controls That Act Sooner — Repeated calls, runaway budgets and unsupported answers are now stopped during the turn, not just reported after it.

05-Aug-2026

One Home Per Measurement — Every quality score is counted once and grouped by the step it measures, so a protection doing its job never turns red because of a later score.

19-Jul-2026

One Design Language — Every panel shares one layout: key figures on top, a resizable divider and clear tabs below, with names that match the navigator.

19-Jul-2026

The Agent Anatomy — Configuration opens as the agent itself, drawn as a star. Click any part to land where it is configured.

19-Jul-2026

The Trust Circle — Eight live trust readings around every agent. Click a station to act on it, or press play to hear the agent’s story from its own records.

19-Jul-2026

From Finding to Chunk — A security finding in a document repository now points to the exact text that caused it, and can be exported as a report.

19-Jul-2026

Knowledge Answers, Grounded — Answers from your documents show they stayed inside them: unsupported claims are marked, and the agent says when the documents don’t cover a question.

19-Jul-2026

A Screened, Governed Knowledge Base — Every document is screened as it is added, a bad one is a click from quarantine, and answers show how fresh and trustworthy their sources are.

19-Jul-2026

Proven, Not Promised — The knowledge protections ship with a benchmark anyone can rerun for free, plus live attack tests that check the defenses actually held.

18-Jul-2026

Trust, Made Visible — See an agent’s data boundary on its knowledge graph, any drift across it in red, and each answer’s signed history as a chain you can verify.

16-Jul-2026

Test Your Own Defenses — Attack any agent with a live test set to see whether its guards held, then export a signed audit pack mapped to the EU AI Act, NIST and GDPR.

15-Jul-2026

Injection Firewalls — Hidden instructions in memories, documents and tool results are caught, closing the paths a check on your own question would miss.

13-Jul-2026

Human Review — A risky answer can be held for a person to approve, with nothing lost and nothing sent unseen. Ideal for unattended runs.

11-Jul-2026

Why Trust This Answer? — Every answer carries a plain-language trust summary and a signed record anyone can verify offline.

09-Jul-2026

Assurance Inbox & Policy Backtesting — Trust incidents land in one queue. Before tightening a policy, replay it over your history to see what it would have changed.

07-Jul-2026

Behavioral Trust Score — An agent’s trust follows its measured behavior, and a circuit breaker can alert you or pause it when behavior slips.

05-Jul-2026

The Trust Journey — Every answer shows, in one compact strip, which protections it passed and which of them acted.

03-Jul-2026

The Trust Membrane — A signed trust badge on every agent and shareable asset. An agent is trusted only if everything it is built from is, even after it is shared.

02-Jul-2026

Agent Authority — An agent working on its own can never exceed the authority you granted, and its actions are recorded as its own, not yours.

01-Jul-2026

Governed Metrics & Open Standards — Define a metric once and have the AI and your BI tools use it identically, then export it all in open W3C standards.

30-Jun-2026

Platform Ontology — A live map of your whole estate: every agent, what grounds it and what it uses, with one-click jumps to where each part is managed.

28-Jun-2026

Agent Memory — The agent remembers your preferences across sessions, per agent or shared, and you can view or erase every memory.

28-Jun-2026

Uderia as an MCP Server — Use any governed Uderia agent as a tool in Claude Desktop, Cursor or other agents, with every protection still applied.

27-Jun-2026

AI Guardrails — Personal data is masked, toxic output is blocked, and prompt injection is stopped before it reaches the model.

26-Jun-2026

Grounding Gate — Answers are checked against the agent’s declared data scope, and can be flagged or held when they drift outside it.

24-Jun-2026

Ontology — Turn Databricks industry data models into a knowledge graph, and deploy them as real Teradata databases with demo data.

21-Jun-2026

Agent Observation & Evaluation — Every turn and flow run is scored on quality, cost and safety, with a health view for each agent.

21-Jun-2026

Next Step Recommender — After each answer, one-tap suggestions for useful follow-up questions and actions, at little or no extra cost.

10-Jun-2026

Flow Parameters & Visual Debugging — Turn a flow into a reusable template, and debug it one node at a time while seeing each node’s input and output.

05-Jun-2026

Sharper Self-Learning — The agent learns only from turns that actually fetched data, so the approaches it reuses stay accurate.

17-May-2026

Dashboard Component — Pin any chart or table from the conversation to a persistent dashboard whose tiles refresh on demand.

16-May-2026

Visual Flow Orchestration — Describe a multi-step workflow and the agent builds and runs it as a live graph, with branches and parallel paths.

14-May-2026

New LLM Providers — MLX for Apple Silicon local inference and xAI for Grok models, expanding provider flexibility from hyperscalers to sovereign on-device execution.

12-May-2026

Scale Knowledge Graphs to Neo4j — Start on the built-in store and move large graphs to Neo4j, without changing agents or workflows.

10-May-2026

Infinity Context — The agent recalls the relevant parts of a long conversation instead of forgetting the oldest ones.

09-May-2026

Model Guide — Compare providers, models, context sizes and prices in the Uderia Model Guide, one click from the LLM settings.

03-May-2026

Your Workplace Apps, Connected — The agent works across Google, Microsoft 365 and Slack, with each user connecting their own accounts.

02-May-2026

Always-Fresh Knowledge — Knowledge repositories update themselves when source documents change, re-indexing only what is new.

02-May-2026

SAML 2.0 & JIT Provisioning — Connect to corporate identity providers via SAML 2.0; user roles update automatically at login based on IdP group membership.

01-May-2026

Task Scheduler — Schedule any agent query to run automatically on a cron or interval; manage schedules through conversation or the split-panel UI, with results delivered by email or webhook.

28-Apr-2026

Parallel Query Execution — Multi-step queries run faster: independent plan phases and date-range iterations now execute in parallel instead of sequentially.

16-Apr-2026

Knowledge Graphs in Agent Packs — Export and import Knowledge Graphs as part of agent packs; import restores graphs without displacing existing active graphs.

10-Apr-2026

Genie Coordinator Pass-Through — Single-expert queries skip the synthesis step, cutting coordinator token cost and response time in half.

04-Apr-2026

OpenRouter Support — Access 100+ models (Llama, Mistral, Gemma, and more) via a single API key.

15-Apr-2026

Teradata Hybrid Search — Enable BM25 hybrid search on Teradata VantageCloud knowledge repositories directly from the repository inspector.

10-Mar-2026

Hybrid Search — Choose between semantic, hybrid (RRF), and keyword (BM25) search on Qdrant Cloud knowledge repositories.

08-Mar-2026

Teradata VantageCloud Backend — Store and search knowledge in Teradata VantageCloud with server-side embedding and enterprise-grade connection resilience.

07-Mar-2026

ChromaDB & Qdrant Cloud — Zero-config local knowledge storage (ChromaDB) or managed cloud vector search (Qdrant), selectable per repository.

01-Mar-2026

Context Window Management — Control how much of your conversation history and knowledge the agent sees; configure per profile with live token budget visualization.

29-Feb-2026

Canvas Component — Write and run SQL directly in the chat canvas with syntax highlighting and live result rendering.

27-Feb-2026

Data Visualization — Turn query results into interactive charts (16 types) automatically; the agent picks the right visual for your data.

23-Feb-2026

Extensions — Transform agent responses on the fly: extract JSON, classify, summarize, generate PDFs, or chain custom transformations with a single # trigger.

22-Feb-2026

Skills - Pre-processing prompt injections with 5 built-in skills and visual editor

15-Feb-2026

Dual-Model Cost Breakdown - Live Status displays strategic vs tactical costs for Fusion Optimizer dual-model executions with color-coded visualization

14-Feb-2026

Theme-Aware Token/Cost KPIs - Live Status token and cost displays now adapt to both dark and light themes with consistent visibility

13-Feb-2026

n8n Integration - Visual workflow automation with three production-ready templates (Simple Query, Scheduled Reports, Slack Integration) and comprehensive deployment guides

08-Feb-2026

Agent Packs - Bundle complete agent teams (coordinator, experts, knowledge collections) into portable .agentpack files for one-click install, export, and marketplace sharing

01-Feb-2026

Text to Speech - Voice output now available for all profile classes with Escape-to-cancel and smart text truncation

31-Jan-2026

Document Upload & Multimodal Analysis - Attach documents and images in chat with native multimodal delivery and automatic text extraction fallback across all providers

30-Jan-2026

All Models Available - Access all provider models with ★ recommended highlights

30-Jan-2026

Light Theme - Full light mode support with proper contrast and visibility across all components

25-Jan-2026

Legacy Theme - Fixed color bleeding issues with opaque backgrounds and consistent gray palette

23-Jan-2026

Modern Theme - Enhanced glass-panel effects and refined slate/blue color palette

17-Jan-2026

Genie Profiles - Hierarchical AI Organizations with multi-level autonomous coordination (Parent → Child agents)

16-Jan-2026

Session Primer - Auto-initialize sessions with domain knowledge, transforming generic LLMs into pre-educated specialists

16-Jan-2026

Genie Profile Type (Beta) - Multi-profile coordination with LangChain orchestration

16-Jan-2026

Genie UI Features - Inline progress cards, collapsible child sessions, split view access

09-Jan-2026

Knowledge Focused (RAG) Profile Type - Mandatory knowledge retrieval with anti-hallucination safeguards

11-Jan-2026

Profile Classes - Four Execution Modes (Conversation, Efficiency, Knowledge, Genie)

10-Jan-2026

Export/Import Knowledge Repositories

09-Jan-2026

Export/Import Planner Repositories

02-Jan-2026

OAuth Implementation - Google, Github

02-Jan-2026

Email Validation for Registration

20-Dec-2025

Extended Prompt Management System - Dynamic Workflow Prompts

19-Dec-2025

Extended Prompt Management System - Dynamic Variables

19-Dec-2025

Extended Bootstrapping - Enhanced Bootstraping Parameter Configuration

12-Dec-2025

Enhanced Prompt Encryption/Decryption Process using Database Encryption

12-Dec-2025

Migration from File based Application Configuration to Database Schema

05-Dec-2025

Consumption Profile Enforcement - Rate Limiting and Usage Quotas

07-Dec-2025

Financial Governance - Dashboards and LiteLLM Integration

06-Dec-2025

Planner Constructor: SQL Query - Document Context

05-Dec-2025

Planner Constructor: SQL Query - Database Context

29-Nov-2025

Knowledge Repository Constructor - Document Storage

28-Nov-2025

Knowledge Repository Integration

28-Nov-2025

Multi-User Authentication - JWT tokens, access tokens, user tiers

22-Nov-2025

Profile System - Modular Configuration & Temporary Overrides

21-Nov-2025

Planner Repository Constructors - Modular Plugin System

21-Nov-2025

Modern UI Design

15-Nov-2025

Flowise Integration

15-Nov-2025

Airflow Integration

14-Nov-2025

Self-Improving AI (RAG)

07-Nov-2025

UI Real-Time Monitoring of Rest Requests

31-Oct-2025

Fully configurable Context Management (Turn & Session)

26-Oct-2025

Turn Replay & Reload Plan

25-Oct-2025

Stop Button Added - Ability to immediately Stop Workflows

24-Oct-2025

Robust Multi-Tool Phase Handling

11-Oct-2025

Friendly.AI Integration

10-Oct-2025

Context Aware Rendering of the Collateral Report

20-Sep-2025

Microsoft Azure Integration

19-Sep-2025

REST Interface for Engine Configuration, Execution & Monitoring

12-Sep-2025

Significant Formatting Upgrade (Canonical Baseline Model for LLM Provider Rendering)

05-Sep-2025

Conversation Mode (Google Cloud Credentials required)