Connect to Any LLM Securely and Easily

Fully realize the power of AI in Liferay DXP with a native MCP server.

MCP: Your Backbone for Scalable, Enterprise-Ready AI

Connect LLMs with minimal configuration.

Liferay’s out-of-the-box Model Context Protocol (MCP) server gives you a standardized, secure, and extensible way to deliver contextual data from your Liferay applications to any LLM. Instead of building and maintaining custom data pipelines for each use case, you can easily connect LLMs to Liferay's MCP server. With the MCP handling data exchange, your LLMs stay consistently informed and easily scalable across your DXP.

Develop AI features faster.

Make it easy for developers to create custom AI features that tap directly into your organization's data and business logic. With LLMs interacting natively with Liferay DXP, the MCP server provides a reliable framework for AI development, making teams faster and reducing integration hurdles. Using MCP support to manage data access and context, developers can build AI capabilities that:

  • Surface real-time product recommendations based on customer data from Liferay's native analytics.

  • Create personalized marketing content instantly for specific segments made within the Liferay platform.

  • Automate complex processes like creating support tickets or routing approval requests based on company policies.

Ensure enterprise-grade AI security and data privacy.

All data exposed via the MCP server respects Liferay's robust permissions and access control model, si data access through the LLM is restricted to only authorized users. This guarantees that MCP-powered AI interactions remain fully aligned with your compliance and data privacy requirements.

How the MCP Server Works with Liferay DXP

MCP Server + AI Hub

The MCP server acts as the bridge that enables connections to the LLMs powering AI agents within Liferay's AI Hub. By routing contextual data through the MCP, these AI services become available as actors you can orchestrate directly within your automations. With MCP support built in, AI Hub workflows can leverage LLM intelligence without custom integrations, making it faster to design, refine, and scale intelligent automations across your DXP.

Learn more about Liferay's AI Hub here.

MCP Server + Security

The MCP automatically respects all Liferay permissions, so that an agent run by a user can only see the same content and sensitive data that the user is allowed to see to begin with.

MCP Server + Low-Code

Users can create and configure new MCP tools and prompt templates through a low-code interface.

Configuring MCP

Using Liferay as an MCP Server

The Model Context Protocol (MCP) is an open source standard for connecting to AI applications, such as GitHub Copilot or Cursor. Learn how Liferay DXP functions as an MCP server that AI applications can use.

What Is an MCP Server (Model Context Protocol)?

MCP is the open-source standardized way to connect AI systems and applications to external tools. Using MCP, AI applications like Claude or ChatGPT can connect to external data sources, tools, and workflows, enabling seamless integration to access key information and perform tasks. So you're able to interact with the MCP server through the AI application's UI.

How do MCPs work?

MCP servers provide functionality to build interactions between AI assistants and applications through three key building blocks:

  • Tools: Functions that your LLM can actively call and decide when to use based on user requests. Tools can write to databases, call external APIs, modify files, or trigger other logic.

  • Resources: Passive data sources that provide read-only access to information for context, such as file contents, database schemas, or API documentation.

  • Prompts: Pre-built instruction templates that tell the model to work with specific tools and resources.

Ultimately, MCP servers help reduce development time and complexity when building or integrating with an AI application or agent. With Liferay's MCP Server, you're able to connect your preferred LLMs directly and securely, with minimal configuration - fully realizing the value of AI within the platform.

Frequently-Asked Questions

MCP servers are programs that expose specific capabilities to AI-powered tools through standardized protocol interfaces. Common examples include:
- File system servers that allow AI to read and navigate documents.
- Database servers for data queries.
- GitHub servers for code management.
- Calendar servers for scheduling.

Liferay's MCP Server provides out-of-the-box tools and low-code interfaces to create prompts. By connecting LLMs to Liferay's MCP server, you can leverage those AI services to configure agents within Liferay DXP with the AI Hub. This way, the MCP ensures that every interaction remains consistent, governed, and extensible as your AI needs grow.

Think of Liferay's MCP server as LLM-agnostic. You can use it to connect to your preferred LLM, like OpenAI, Gemini, Azure AI, and more.

Liferay's MCP server includes the core MCP server, business user tool and prompt management, out-of-the-box context providers, permissions, and a clear API and developer framework.

Through the MCP, LLMs can securely access the full context of your DXP application and analytics. The MCP automatically respects all Liferay permissions, so that an agent run by a user can only see the same content that the user is allowed to see to begin with.

Make Liferay DXP AI Ready