This post explores the integration of Informatica-hosted Model Context Protocol (MCP) servers with AWS Agentic AI services, providing joint enterprise customers with programmatic access to the core data management capabilities of the Informatica Intelligent Data Management Cloud (IDMC) platform. By leveraging these MCP servers, organizations can ground AI agents in Amazon Bedrock and Amazon Quick with relevant metadata and high-fidelity, governed enterprise data. This architectural approach ensures that agentic workflows are powered by trusted data and reliable enterprise context, significantly enhancing the trustworthiness of AI outcomes while mitigating the risk of LLM hallucinations.

In this section we will provide an architecture overview of the integration Informatica-hosted MCP server with Amazon Bedrock AgentCore.

Amazon Bedrock AgentCore is a fully managed, framework-agnostic platform designed to build, deploy and operate autonomous AI agents securely and at scale. It highlights how the platform simplifies agent development by providing modular components to handle infrastructure complexities. Key architectural components described include:

• Runtime – a serverless environment for hosting the agentic workloads.
• Gateway – a central interface connecting the agents to Informatica hosted MCPs.
• Identity – which manages secure authentication and fine-grained authorization to enable secure the outbound connection between the Gateway and Informatica hosted MCPs using Service-to-Service OAuth 2.0.

Learn more about Amazon Bedrock AgentCore via these Amazon Bedrock AgentCore Tutorials.

At Informatica World 2026, as part of Informatica’s Headless Data Management initiative, the general availability of five Informatica MCP servers was announced. Their integration with the AWS Agent Registry (in private preview) enables seamless discovery and provisioning of these MCPs directly via the Amazon Bedrock AgentCore Gateway. Getting Started with Informatica MCP Servers.

To create a new gateway and configure the Gateway to add Informatica MCPs as target, refer the following documentation:

• Create an Amazon Bedrock AgentCore gateway
• Set up outbound authorization for your gateway
• Add targets to an existing AgentCore gateway

The Informatica-hosted MCPs can also integrate with Amazon Quick, an AI assistant and enterprise AI platform form AWS designed to bring generative AI-powered search, workflow automation and business intelligence together. Learn more: Amazon Quick – AI Assistant.

In the section, we’ll explore the scenario of a data professional tasked with calculating real-world fulfillment windows after a purchase request is submitted. We will demonstrate how integrating Amazon Quick with Informatica hosted MCPs empowers the data analyst via an agentic workflow to:

1. Discover essential data assets with speed and precision.
2. Investigate schemas and data profiles via automated metadata inspection.
3. Evaluate and compare against diverse datasets to determine the optimal fit for the specific business case.
4. Procure the data through formalized access requests and governed provisioning workflows.
5. Seamlessly deliver the data assets to approved environments for analytics.
6. Bring reference data from Informatica Master Data Management evaluate the impact of supplier performance on order delivery timelines.

The Data Discovery & Procurement Agent is integrated with the following Informatica hosted MCPs

• Informatica CDGC Metadata Search MCP
• Informatica Data Provisioning MCP
• Informatica Job Management MCP
• Informatica Master Data Management MCP

[Step-1] Data Discovery: The data analyst initiates data discovery using natural language, specifying the analytics use case and desired insight. The agentic workflow then leverages the Informatica CDGC Metadata Search MCP to instantly surface information about relevant governed assets and enterprise context from the Informatica Cloud Data Governance & Catalog Service

[Step-2] Metadata Exploration: For comprehensive technical analysis via the Informatica CDGC Metadata Search MCP, the agent provides granular schema details and column-level profiles, surfacing attributes including table properties, column definitions from the business glossary, volume metrics, certification status, and domain context. This enables the analyst to fully understand and trust the data before gaining access.

[Step-3] Metadata-Driven Data Selection: Via the Informatica CDGC Metadata Search MCP, the agent enables the analyst to conduct a rigorous comparison of alternative data assets and their column-level profiles across multiple metrics. This eliminates ambiguity, ensuring the confident selection of the most suitable data to power advanced analytics and AI initiatives.

[Step-4] Data Procurement: With a comprehensive understanding and trust in the dataset for the analytics use case, via the agent and Informatica Data Provisioning MCP, the data analyst initiates a request-approval to the data owner for access through the Informatica Cloud Data Marketplace service. This process ensures self-service, governed data access with end-to-end auditability for compliance, allowing the analyst to select the target analytics platform from approved options, such as Amazon Redshift, Snowflake or Cloud Data Access Management (CDAM) Dynamic Access. The agent responds with the data procurement request details.

[Step-5] Data Delivery: Upon approval from the data owner, the agent executes a Informatica Data Integration task via the Informatica Job Management MCP, provisioning the requested data to the analyst's chosen platform – in this case, Amazon Redshift.

[Step-6] Lead Time Impact Analysis: Equipped with supplier lead time insights, the analyst leverages high-quality, high-fidelity master data from Informatica MDM. By synthesizing analytics with critical cross-domain intelligence, the agent accurately quantifies the impact of supplier performance on order delivery timelines.

By enabling seamless discovery of the MCPs within AWS Agent Registry, this integration establishes programmatic access to the core data management capabilities of the Informatica IDMC platform. By bringing governed metadata, high-fidelity master records and trusted data with relevant enterprise context directly to AI agents, organizations can advance their agentic initiatives to drive meaningful outcomes.

Ready to ground your AI agents in trusted, governed data? Explore Informatica's AI Agent Engineering Platform and see how MCP servers can power your agentic workflows on AWS.