AWS Retires Bedrock Agents Classic: What Operators Must Do Now

30-Second Summary

Amazon closed Bedrock Agents Classic to new customers on July 30, 2026, renamed it from the original Bedrock Agents product launched in November 2023, and froze its model catalogue. Existing agents continue to run but cannot access any foundation models released after the maintenance mode date. The forward path is Bedrock AgentCore, which reached general availability in June 2026 and offers multi-agent orchestration, persistent memory, a gateway layer, and full access to new models.

At a Glance

• Topic: Agent Systems
• Company: Amazon Web Services
• Date: July 30, 2026 (maintenance mode effective date)
• Announcement: Bedrock Agents Classic closed to new customers; model catalogue frozen
• What Changed: The original Bedrock Agents product (launched November 2023) is renamed Bedrock Agents Classic and enters maintenance mode. AgentCore is the successor platform.
• Why It Matters: Any agent built on Classic is now locked to older foundation models, creating a capability gap that widens with every new model release.
• Who Should Care: Organisations running internal agents, document automation, customer service bots, or any workflow automation on AWS Bedrock.

Key Facts

• Bedrock Agents Classic launched November 2023, one of the earliest managed agent platforms from a hyperscaler.
• Closed to new customers: July 30, 2026.
• Model catalogue frozen as of the maintenance mode effective date.
• Existing agents continue to function, no immediate shutdown.
• AgentCore Harness reached general availability on June 17, 2026.
• AgentCore is framework-agnostic, meaning it is not locked to any single agent SDK.
• AgentCore includes dedicated services for runtime, gateway, memory, identity, and observability, capabilities Classic bundled together without clear separation.

What Happened

Amazon launched Bedrock Agents in November 2023 as one of the first enterprise-grade managed agent platforms from a major cloud provider. It allowed organisations to build agents that could query knowledge bases, call APIs, and run multi-step reasoning workflows using foundation models available in Bedrock's catalogue.

On July 30, 2026, Amazon rebranded the product to Bedrock Agents Classic and stopped accepting new customers. At the same time, it froze the model catalogue, meaning agents built on Classic can no longer be updated to use newer foundation models. Amazon Bedrock itself, including Knowledge Bases and Guardrails, is not affected. The freeze applies specifically to the agent runtime layer.

The replacement is Bedrock AgentCore, which Amazon describes as a framework-agnostic runtime designed for production-grade agent workloads. AgentCore breaks what was a monolithic service in Classic into modular components: a runtime layer, a gateway for tool connections, persistent memory, identity management for agent-to-agent trust, and observability tooling. The Harness component, which handles agent orchestration, reached GA on June 17, 2026.

Amazon has not announced a hard end-of-life date for Classic, but the pattern is clear: maintenance mode is the holding state before eventual sunset. How long that window lasts is unknown, and organisations that wait for a firm date are taking on unnecessary risk.

Why It Matters

The capability gap is immediate and permanent. With the model catalogue frozen, Classic agents are locked to older models. Every new release from Anthropic, Meta, Amazon itself, or any other provider integrated into Bedrock will be available in AgentCore and not in Classic. That gap compounds with every passing month.

Enterprise agent workloads are not easy to migrate at speed. Bedrock Agents Classic embedded its own opinionated approach to prompting, tool calling, and knowledge base retrieval. AgentCore uses a different architecture. Teams that built deeply on Classic will need to re-engineer workflows, not just swap a configuration flag.

Third-party vendor exposure is underappreciated. Many software products integrated Bedrock Agents under the hood as their AI backbone. If you are using any AI-powered product running on AWS, there is a real chance it depends on Classic. Those vendors now carry migration debt, and they may not have communicated their timelines to customers.

The model freeze creates compliance risk for some industries. Organisations in regulated sectors that must demonstrate they are using current, well-maintained AI systems may find that running on Classic conflicts with their AI governance obligations, particularly under the EU AI Act's newly enforced high-risk provisions.

AgentCore's modular design is materially better for operators. The separation of runtime, memory, gateway, and identity into distinct services means teams can update individual components without redeploying entire agents. That architecture is more resilient for production workloads than Classic's bundled approach.

This signals the speed of platform cycles in enterprise AI. A product launched in November 2023 entered maintenance mode by July 2026. Teams that treat AI infrastructure as a long-term stable platform are systematically underestimating replacement velocity.

The David and Goliath View

The retirement of Bedrock Agents Classic in under three years is not a failure on Amazon's part. It is evidence of how fast the underlying technology has shifted. The architecture that looked sensible in late 2023 does not hold up well against the demands of 2026 production workloads. Modular, observable, memory-persistent agents are what enterprises actually need, and Classic was not built for that.

What this means for operators is straightforward: stop treating your AI agent infrastructure as set-and-forget. The 10 to 200 person organisations we work with often build one automation, see it work, and move on. That approach creates quiet debt. The Classic situation is a reminder that active maintenance of your AI stack is now a baseline expectation, not an optional improvement.

The positive read is that AgentCore is genuinely better. Multi-agent orchestration, proper memory management, and identity controls for agent-to-agent trust are the building blocks of the kind of compound automation that creates real competitive advantage. The migration is friction, but the destination is worth reaching.

Where This Fits in the AI Stack

Bedrock Agents Classic was an agent runtime layer sitting above foundation models (Claude, Titan, Llama, etc.) and below application logic. AgentCore occupies the same layer but adds horizontal services, memory and gateway, that previously had to be custom-built or stitched together from separate AWS services. For operators running the David and Goliath AI Growth Engine or Employee Amplification Systems on AWS, the agent runtime is the connective tissue between the models and the workflows. Getting that layer right matters more than the individual models.

Questions Operators Are Asking

Do I need to do anything right now? If you have existing agents on Classic, they keep running. There is no immediate outage risk. The urgency is strategic: your agents are now on a frozen model version with no path to improvement until you migrate to AgentCore.

How long do I have before Classic shuts down completely? Amazon has not published a hard end-of-life date. Maintenance mode typically precedes shutdown by 12 to 24 months, but that is a general cloud industry pattern, not an AWS commitment. Do not plan your migration timeline around a date that does not exist yet.

Is AgentCore a like-for-like replacement? Functionally, yes. Architecturally, no. AgentCore offers the same core capabilities as Classic but structures them differently. Expect meaningful engineering work to migrate complex agents, particularly those with custom tool integrations or deeply tuned prompting logic.

Does this affect Amazon Bedrock Knowledge Bases? No. Knowledge Bases, Guardrails, and the broader Bedrock model catalogue are not in maintenance mode. The change is specific to the Agents Classic runtime layer.

What if one of our software vendors built on Classic? Ask them directly. Request their AgentCore migration timeline and what the impact on your product experience will be. Vendors who cannot give you a clear answer are carrying technical debt that will eventually affect their service quality.

Citable Summary

Amazon closed Bedrock Agents Classic to new customers on July 30, 2026, freezing its model catalogue and directing all new agent development to Bedrock AgentCore. Existing agents continue to run but cannot access any foundation models released after the maintenance mode date. AgentCore, which reached general availability in June 2026, replaces Classic with a modular architecture covering runtime, memory, gateway, identity, and observability. Organisations running production workflows on Classic should begin migration scoping now rather than waiting for a hard end-of-life date.