AWS unveils new AI services—Continuum and enhanced security—empowering intelligent agents with contextual data and robust pro

Editorial illustration for AWS launches Continuum and another service to add context, security to AI agents

AWS launches Continuum and another service to add...

AWS launches Continuum and another service to add context, security to AI agents

By AI Daily Post Edited by Brian Petersen, Editor-in-Chief

June 21, 2026 • 2 min read

At the AWS Summit in New York, Amazon unveiled two services aimed at shoring up the weaknesses of AI agents in real‑world deployments. The first, AWS Continuum, promises to automate the detection, prioritization and remediation of code vulnerabilities that emerge when agents generate software. The second, AWS Context, supplies agents with a shared knowledge graph that encodes business‑specific information, giving them the context needed to make sensible decisions.

While the DevOps Agent now includes verification steps that test AI‑generated code in production‑like environments before it goes live, the new offerings extend that safety net across the entire lifecycle. AWS also rolled out its coding agent Kiro as an iOS app and added more data connectors and security filters to the Bedrock AgentCore platform. The company says early pilots will determine how well the services keep up with the speed of AI‑driven development, a pace that traditional security tools struggle to match.

In short, Amazon is trying to make AI agents production‑ready by filling the gaps in context and security.

AWS says AI agents lack business context and security, launches two services to patch the gaps Key Points - Amazon is launching two new AWS services: AWS Continuum, which automates the fixing of code vulnerabilities, and AWS Context, which feeds AI agents business knowledge through a knowledge graph to improve their decision-making. - The AWS DevOps Agent now includes verification capabilities that check AI-generated code before it goes live, automatically testing it in production-like environments to catch potential system failures early. - AWS is also releasing its coding agent Kiro as an iOS app for on-the-go control, while expanding the Bedrock AgentCore platform with additional data connectors and security filters.

— AWS, AWS says AI agents lack business context and security, launches two services to patch the gaps - THE DECODER

Why this matters

We see AWS addressing two persistent gaps in AI‑driven workflows: code security and contextual awareness. Continuum promises to automate vulnerability remediation, while Context injects business knowledge through a knowledge graph, aiming to make agents’ decisions more grounded. For developers, the added verification step in the DevOps Agent could reduce the manual review burden, yet it remains unclear whether automated testing in production‑like environments will catch edge‑case exploits.

Founders may welcome a turnkey solution that claims to tighten security, but the cost and integration effort could temper enthusiasm. Researchers will note the shift toward embedding structured knowledge into agents, a move that aligns with recent academic calls for richer context, though the practical depth of the graph’s data has not been disclosed. Ultimately, the services could streamline deployment pipelines, but their real‑world impact will depend on how effectively they balance automation with the nuanced judgement that security and business logic often require.

We’ll watch how early adopters evaluate the trade‑offs between convenience and the lingering uncertainty around fully automated safeguards.