The bespoke work of stitching together a context layer between enterprise data stores and AI agents has long been a pain point with no standard service to automate or maintain the evolving graphs. Amazon is making a direct play to change that with the release of AWS Context, a new knowledge graph service designed to get smarter through agent usage over time. Announced at the AWS Summit NYC, the service is the centerpiece of a broader context intelligence stack that includes the general availability of Amazon S3 Annotations and a preview of skill assets for the AWS Glue Data Catalog.
How AWS Context Builds a Self-Learning Knowledge Graph
AWS Context automatically constructs a knowledge graph from existing enterprise data without requiring manual curation. The service maps relationships across disparate data sources, inferring what tables exist, what columns mean, how sources relate, and which sources are authoritative. It combines semantic search with graph-level reasoning, making this structured context available to AI agents at runtime.
“This service automatically builds a knowledge graph from all your existing data,” said Swami Sivasubramanian, vice president of Agentic AI at AWS. “This service infers relationships across your data sets, business rules, and domain knowledge, and makes all of it available to your agents and your organization at runtime.”
The graph improves itself over time by learning which sources produce correct results and which parts of the graph see the most usage, eliminating the need for human re-curation. Data stewards manage the graph through the AWS Management Console, where they can review inferred relationships, promote them 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 existing enterprise controls.
Why the Context Layer Is Now a Contested Architectural Category
The context layer has become a critical architectural category with significant competition. Snowflake recently announced its own approach with Horizon Context and Cortex Sense services. Microsoft provides contextual capabilities through its Fabric IQ platform, which offers a semantic ontology for data. Redis has developed a context platform that optimizes data for retrieval, while vector database vendor Pinecone offers its Nexus context service that compiles enterprise data into task-specific artifacts before agents query them.
AWS’s structural argument centers on zero-integration friction for enterprises already running S3, Glue, and Lake Formation. The service extends the existing identity model with no data movement required, allowing context from systems outside AWS to be pulled into the same graph through third-party catalog connections. All metadata is published in Apache Iceberg format to Amazon S3 Tables, queryable via Athena, Redshift, Spark, or any Iceberg-compatible engine without proprietary APIs.
AWS’s Three-Layer Context Intelligence Stack
AWS is layering multiple services to help enterprises build context across the data stack. Amazon S3 Annotations enables users to attach rich business context directly to individual S3 objects at the storage layer. AWS Glue Data Catalog skill assets attach domain knowledge at the catalog layer, linking runbooks, query patterns, and usage rules to data assets across the estate. AWS Context then synthesizes both into the knowledge graph that agents query at runtime, combining semantic search with graph-level reasoning across structured and unstructured sources, with each layer feeding the next.
What Makes AWS Context Different From Other Context Offerings
The key architectural premise for AWS Context is that the graph should learn from how agents use it automatically, without human re-curation. “Your agents now get smarter without you having to rebuild anything from scratch,” Sivasubramanian said. The service provides agentic search APIs and MCP tools that work across Bedrock AgentCore, EKS, or any MCP-compatible framework.
“Context makes agents more powerful and as the whole world is building agents, every agentic platform vendor needs a context capability,” said Holger Mueller, VP and Principal analyst at Constellation Research. “The concern — as with all context offerings — is going to be performance, especially for transactional data.”
Who Should Evaluate AWS Context Now
For enterprises already invested in the AWS data ecosystem, particularly those using S3, Glue, and Lake Formation, AWS Context presents an opportunity to set up a context layer with minimal upfront integration work. Organizations building AI agents that need to query across diverse data sources with consistent permissions and auditing should explore the service through the AWS Management Console. The key factor to evaluate is whether the self-learning graph approach meets performance requirements for your specific use cases, particularly for transactional data workloads where latency and accuracy are critical.
/i.s3.glbimg.com/v1/AUTH_08fbf48bc0524877943fe86e43087e7a/internal_photos/bs/2026/B/N/5EQ9QeQAa2XoA9mRtS0w/marvels-wolverine-state-of-play.jpg "Marvel’s Wolverine Settings: What Actually Works in Combat")
