Skip to content
AI Radar by Wagner Solution
CS
Menu
worth noting AI agents

Tutorial connects a Strands agent to enterprise data through Strategy Mosaic and Amazon Bedrock AgentCore

only one source so far

A tutorial on the AWS blog shows how a Strands agent in Amazon Bedrock AgentCore accesses Strategy Mosaic through MCP. The example combines customer and purchase data from Amazon Aurora and Amazon Redshift.

A tutorial published on the AWS blog and prepared with representatives from Strategy connects a Strands agent to the Strategy Mosaic semantic layer. The agent runs in the Amazon Bedrock AgentCore runtime and queries data from Amazon Aurora and Amazon Redshift through an MCP interface. It uses Claude Sonnet 4.6 through Amazon Bedrock to interpret requests and plan queries.

The example keeps customer details and loyalty program membership in an Amazon Aurora database separate from purchase data in an Amazon Redshift data warehouse. Strategy Mosaic sends a separate SQL query to each source and combines the intermediate results using shared attributes. Metric definitions, join rules and caller-based row and column filters are stored in Strategy Mosaic. One example is a query asking which of the 20 customers with the highest total spending are not yet loyalty program members.

According to the article, access to Strategy Mosaic through an MCP server has been available as standard since the March 2026 release of Strategy One. Details can be found in the source article.

What changed

Why it matters

The tutorial provides a concrete architecture for questions that require both operational and analytical data to answer. It shows where to handle database integration and business definitions so the agent does not need to receive them in the instructions for every query.

Two audiences, two different impacts

What this means

01

For individuals

Developers have a template for connecting an agent to data from two database systems through a single MCP interface, leaving the semantic layer to combine the data.

What to do When testing a similar integration, use the sample query asking which of the 20 customers with the highest lifetime spending are not yet loyalty program members.
More practical updates →
02

For a business

For enterprise customer analytics, the example centralizes definitions of spending, membership and access filters in Strategy Mosaic. These rules remain separate from the instructions individual users give the agent.

Development
What to decide Before deployment, verify that the metric definitions and caller-based filters comply with company policies for customer data.
More business impacts →
Amazon Aurora Amazon Bedrock AgentCore Amazon Redshift MCP Strands Strategy Mosaic

Check the original

Event sources

only one source so far · 1 publisher, 0 independent. We count feeds from the same owner only once.

1
AWS Machine Learning Blog primary source · first detected Build a semantic layer for agentic AI on AWS with Strategy Mosaic and Amazon Bedrock AgentCore