Context intelligence for your data and AI agents at scale
TL;DR
AWS used its New York Summit to announce AWS Context, a coming service that maps relationships across data lakes, warehouses, databases, streams, and internal knowledge into a managed knowledge graph. Agents are meant to query that graph at runtime through agentic search and MCP. Access is tied to IAM and Lake Formation permissions, so queries can be governed and audited.
Nauti's Take
The framing is heavy on AWS launch language, but the underlying point is real: agents usually fail less because of the model and more because context, permissions, and business rules are missing. AWS is pulling the context layer close to S3, Glue, Lake Formation, and Bedrock.
That may be useful for AWS-heavy teams. For everyone else, it is a signal that context infrastructure is becoming as important as the model itself.
Briefingshow
AWS is moving agents from prompt demos into the messy reality of enterprise data: who can see what, which table means what, and which join is actually valid. If the context layer works, agents should guess less and lean more on governed business logic. The catch: much of this sits deep inside the AWS stack, and the key piece, AWS Context, is not available yet.
Sources
