🟡 🤝 Agents Published: · 2 min read ·

Anthropic: Claude Managed Agents Arrive on Claude Platform on AWS

Editorial illustration: Claude Managed Agents arrive on Claude Platform on AWS

Anthropic announced that key Claude Managed Agents features — webhooks, multi-agent orchestration and self-hosted sandboxes — are now available on Claude Platform on AWS, along with new IAM actions and the managed AnthropicSelfHostedEnvironmentAccess policy for authorizing access.

🤖

This article was generated using artificial intelligence from primary sources.

On May 29, Anthropic announced that several key features of the Claude Managed Agents platform have become available on Claude Platform on AWS. These are webhooks, multi-agent orchestration and self-hosted sandboxes, integrating agentic capabilities directly into Amazon’s environment.

What are Claude Managed Agents?

Claude Managed Agents are Anthropic’s managed layer for building and running AI agents. Instead of developers building their own infrastructure for task execution, the platform offers ready-made components for launching, monitoring and coordinating agents. The announcement means these components now also work when Claude is used through Amazon Web Services (AWS), and not only directly through the Claude API.

Which three features became available?

The available features are webhooks, multi-agent orchestration and self-hosted sandboxes. Webhooks allow an external system to receive a notification when something happens during an agent’s run. Multi-agent orchestration coordinates several agents working on related tasks. Self-hosted sandboxes are isolated environments in which an agent securely executes code or operations within the user’s own infrastructure.

What changes for IAM and access?

The announcement also introduces new IAM actions and the managed AnthropicSelfHostedEnvironmentAccess policy. IAM (Identity and Access Management) is AWS’s system for controlling who may access which resources. The new managed policy simplifies granting permissions for self-hosted sandbox environments, so administrators do not have to assemble a set of permissions manually. The details are described in the “IAM actions for Claude Platform on AWS” documentation.

Why does it matter?

Availability on AWS means that organizations already using Amazon’s infrastructure can run agentic workflows within their existing security and network framework. This reduces adoption friction and lets agentic systems stay within the boundaries set by corporate IAM, which is particularly relevant for enterprises with strict compliance requirements.

Frequently Asked Questions

What are Claude Managed Agents?
It is Anthropic's set of features for running AI agents that includes webhooks, multi-agent orchestration and self-hosted sandboxes for executing agentic tasks.
What is AnthropicSelfHostedEnvironmentAccess?
It is a new managed IAM policy that authorizes access to self-hosted sandbox environments on Claude Platform on AWS.

📬 AI news in your inbox

A daily digest built your way — pick topics, sources and cadence. One-click unsubscribe.