[Digital Today reporter Chi-gyu Hwang (황치규)] Amazon Web Services (AWS) announced three new services it bills as a context intelligence stack for AI agents.
The core is 'AWS Context', which automatically builds knowledge graphs from existing data and learns agent usage patterns to improve itself.
According to a VentureBeat report, AWS Context combines semantic search and graph-level reasoning to automatically identify relationships between tables, the meaning of columns and the reliability of data sources. This lets agents immediately pull the information they need when executing tasks.
All queries can be tracked by identity through access history.
Amazon S3 Annotation supports adding business context directly to S3 objects. AWS Glue Data Catalog Skill Assets links runbooks, query patterns and usage rules to data assets. AWS Context integrates the two layers to build a knowledge graph.
Competition is already intense in the context layer market. Snowflake has Horizon Context and Cortex Sense, Microsoft has Fabric IQ, and Redis and Pinecone have each released context platforms.
AWS said companies already using AWS S3, Glue and Lake Formation can extend their existing permission systems without moving data, positioning that as a differentiator.