AWS enters the context layer race with a graph that learns from agents, not manual configuration
Building a context layer between business data stores and AI agents is an innovative task, without a common service to automate or maintain graphs over time. Amazon is making a direct play to change that.
Amazon on Wednesday stepped into the space, announcing a series of three products it positions as a core intelligence stack for AI agents. At its core is AWS Context, a new information graph service that gets smarter with agent usage over time. AWS also announced the general availability of Amazon S3 Annotations and a preview of capabilities in the AWS Glue data catalog.
The context layer is now a contested architectural category and there is no shortage of options from different vendors. AWS is entering that market with a different architectural premise: that the graph should learn how agents use it automatically, without human reconfiguration.
"Your agents are now intelligent without having to rebuild anything from scratch," said Swami Sivasubramanian, vice president of Agentic AI at AWS, during his AWS Summit NYC keynote.
"This service automatically builds an information graph from all your existing data," he said. "This service integrates relationships across all your data sets, business rules, and domain information, and makes it all available to your agents and your organization at runtime."
AWS Context builds a self-learning knowledge graph from existing data
It’s a problem AWS says it’s seen repeatedly in customer deployments.
AWS Context sets relationships to existing data automatically: what tables exist, what the columns mean, how the sources are related and which sources have authority. It combines semantic search with graph-level reasoning and applies relationships across datasets, business rules and domain knowledge, making it all available to agents at runtime.
"The knowledge graph evolves over time as it learns which sources are producing the right results and which components are being used," Sivasubramanian said.
Data managers manage the graph through the AWS Management Console, review relationships created, promoted to production and attach business definitions and usage rules. Every query inherits the calling user’s IAM and Lake Formation permissions, making agent data access auditable through controls that businesses already rely on.
All metadata is published in Apache Iceberg format to Amazon S3 Tables, which can be queried with Athena, Redshift, Spark or any Iceberg compatible engine, with no proprietary APIs. Third-party catalog connections are supported, so context from systems outside of AWS can be pulled into the same graph. Agents query the agent search APIs and MCP tools across Bedrock AgentCore, EKS and any MCP-compatible framework.
Context is more than just a single service
Context is a complex environment and AWS layers multiple services to help businesses create context across the data stack.
Amazon S3 annotations. This service allows users to attach rich business context to the storage layer, directly to individual S3 objects.
AWS data catalog capabilities assets. The glue capability assets attach domain information to the catalog layer, linking runbooks, query patterns and usage rules to data assets across the legacy.
AWS Context then integrates both into the graph of information queried by agents at runtime, combining semantic search and graph-level reasoning across structured and unstructured sources. Each layer feeds the next.
AWS is entering a very competitive content space
Snowflake announced its approach to content earlier this month with its Horizon Context and Cortex Sense services. Microsoft provides context through its Fabric IQ platform that provides a semantic ontology for data. Redis has developed a context platform that optimizes data for retrieval. Vector database vendor Pinecone has its own Nexus core offering that bundles business data into task-specific artifacts before agents even query them.
The argument for AWS architecture is straightforward: for businesses that already use S3, Glue and Lake Formation, AWS Context extends the existing ownership model without the necessary data movement. The pitch is zero-integration friction – not just cost integration.
"Context makes agents powerful and as the world works to build, every agent platform marketer needs content competence," Holger Mueller, VP and Senior Analyst at Constellation Research, told VentureBeat.
Mueller noted that AWS is no different. "The concern – like all spices offered – will be the performance, especially the transaction data, we will see," he said.
