We are thrilled to introduce Oracle AI Database Private Agent Factory, a no-code agentic platform to build, deploy, run, and manage AI agents with Oracle’s AI-native database. Agentic AI promises to deliver intelligent autonomy to enterprise applications harnessing the data convergence built with Oracle AI Database. Agent Factory helps build and deploy trusted agents by harnessing all private enterprise data with security, safety, and privacy, enabling you to run dynamic agentic business workflows at scale.
With the no-code Agent Factory, Oracle AI Database 26ai reinforces its position as an AI-native platform for production workloads, enabling you to provide sophisticated answers, make decisions autonomously with human-in-the-loop, and take automated actions that combine private database data with AI models trained on public information. This offers a huge productivity boost for enterprise workflows today.
Agent Factory includes pre-built, enterprise-ready agents and workflows, including the Knowledge Agent for retrieval-augmented generation (RAG) from unstructured enterprise data, the Data Analysis Agent for exploring structured datasets. Additional prebuilt agents are planned to be included in subsequent releases.

Agent Factory Container
The Agent Builder is a simple no-code canvas to build a variety of agentic patterns – connect MCP servers with data inputs from documents, databases, users (chat inputs), prompts, a choice of LLMs (that are configurable to be private or on cloud), and in-database AI execution with Oracle Select AI. Data sources can be set up to ingest data from files, Oracle databases, SharePoint, object store, and through REST APIs. Once an agent is created, tested and published, it is accessible through secure REST APIs, so applications can easily include these agents in their business logic. The embedding models and the LLMs run in a supplied container privately or call public endpoints for LLMs, either with OCI Gen AI services or directly to OpenAI, Gemini, and other models. The Oracle AI Database 26ai is configured to use with the Agent Factory to keep the agent runtime data and for the use of the vector store and hybrid search.

Portability is a unique feature in the Agent Factory, as agents are constructed using the open-source Open Agent Specification. A Wayflow runtime environment is available to import and export agents with frameworks like LangGraph, CrewAI, and other frameworks that expose their agent specs.

Agent Factory Key Differentiators:

  • Native Oracle Database Integration – Seamless connectors to Oracle AI Database and AI Vector Search, and OCI Generative AI LLM services
  • Air-Gapped Deployment – Runs on Cloud IaaS or in secure, fully isolated environments, a critical requirement for regulated industries.
  • Enterprise Security – Built-in authorization, auditing, and monitoring using Oracle AI Database’s advanced security features.
  • Prebuilt and agent templates– Ready-to-use agents and workflows for finance, HR, law enforcement, and more.
  • Cross-Framework Compatibility – Agents can be exported in Open Agent Specification for use in frameworks such as LangGraph, CrewAI, and other frameworks.

With full lifecycle capabilities—including connectors to LLMs and embeddings, multi-agent orchestration, and tool integration via REST and Model Context Protocol (MCP)— Agent Factory gives enterprises a powerful “start here” platform to safely adopt and scale AI. The no-code simple user interface provides entry points to your agent builder journey.

  1. Pre-built agents
  2. Agents from templates
  3. Agent Builder Canvas
  4. Data, User, LLM configuration

Agent Factory landing page

Agent Factory is 100% focused on AI inferencing and providing the best business outcomes using Gen AI innovations, and the advances provides by the frontier models. These frontier models are trained in AI Factories, powered by OCI and Nvidia, where power is the input and intelligence (tokens) are the outputs, as you see in the illustration below.

Security and data governance are enforced in-database. Built-in data privacy protection will enable row-, column-, and cell-level controls plus dynamic data masking, so both users and AI agents only see authorized data. SQL Firewall adds scalable, in-database protection against unauthorized SQL and injection attacks without adding a mid-tier server that adds extra network latency and management costs.

Mission-critical operations are covered from end to end. Oracle AI Database Zero Data Loss Cloud Protect provides on-premises databases with real-time protection by continuously shipping changed data to the cloud — helping defend against data loss and ransomware.

The Agent Factory helps keep enterprise data private by operating close to where the data is stored – either in sovereign environments, privately (on-premises) and/or in air-gapped secure environments. Oracle AI Database allows you to build a context graph close to the data, which in turn helps provide the right business context to Gen AI, and workflows to focus on the right business outcomes. Applications call these agents through REST APIs, Chatbot web interfaces, or Physical AI SDKs. Agents are interoperable in a marketplace, with the portability supported by the Open Agent Specification. The context graph is uniquely powered by Oracle AI Database using AI built into the data layer.

Bottom line: Agent Factory in the Oracle AI Database uniquely builds the context graph for Agentic AI, allowing you to easily build trustworthy, high performance AI solutions with the databases that already run your business. It enables delivery of trusted AI-powered insights, innovations, and productivity for all data, across the cloud, multicloud, and on-premises. Agent Factory is available everywhere you can run Oracle AI Database, for no additional charge​.

Agent Factory is available now at no additional cost to Oracle AI Database customers, downloadable from the Oracle Cloud Marketplace, and Oracle Downloads. It works with the Oracle AI Database however you choose to install it.

For on-premises users, Agent Factory is available with Oracle AI Database on the following platforms:

  • Exadata Cloud@Customer
  • Compute Cloud@Customer
  • Oracle Exadata Database Machine
  • Oracle Database Appliance
  • Oracle Private Cloud Appliance
  • Linux x86-64 platforms (Enterprise Edition, Free Edition)

Additionally, Oracle has adopted a cloud-first, developer-first strategy. Pursuant to that strategy Agent Factory is also available with the Oracle AI Database on the following cloud platforms:

  • Oracle Autonomous AI Database
  • Oracle Exadata Database Service
  • Oracle Base Database Service
  • Oracle Database@Azure
  • Oracle Database@Google Cloud
  • Oracle Database@AWS
  • Oracle AI Database installed in Oracle Cloud Infrastructure Compute VMs

Try the hands-on-lab for Oracle AI Database Private Agent Factory today to get started on your journey of building and deploying enterprise Agentic AI to automate turning data into trusted actions.