AWS Launches Managed MCP Server for Secure AI Agent Access to Cloud Services
Breaking: AWS MCP Server Now Generally Available
AWS today announced the general availability of the AWS MCP Server, a managed remote server that gives AI agents and coding assistants secure, authenticated access to all AWS services through a fixed set of tools. The server is part of the Agent Toolkit for AWS and addresses a key challenge: allowing agents to interact with AWS without granting excessive permissions.
“This solves a critical pain point for developers building AI agents that need to access cloud services safely,” said Sarah Chen, AWS Director of AI Developer Tools. “Without such a server, agents often rely on outdated training data or generate overly broad IAM policies.”
The AWS MCP Server uses the Model Context Protocol (MCP) and executes over 15,000 AWS API operations using existing IAM credentials. It also provides real-time documentation retrieval, so agents always work with current information.
Key Features and Enhancements
With general availability, AWS introduced several new capabilities. The server now supports IAM context keys, eliminating the need for a separate IAM permission to use the server. Documentation retrieval no longer requires authentication, reducing friction.
The run_script tool allows agents to execute short Python scripts in a sandboxed environment that inherits IAM permissions but has no network access. This enables data processing without exposing local file systems or shells.
“The run_script tool is a game-changer for multi-step workflows,” added Chen. “It chains API calls and computes results in a single round-trip, significantly reducing token consumption and latency.”
Another major update replaces Agent SOPs with Skills, which provide curated guidance and best practices for specific tasks. Skills are designed to help coding agents build more effectively on AWS.
Background
AI coding agents have become increasingly useful for automating cloud infrastructure tasks, but they struggle with AWS due to outdated training data and a tendency to use AWS CLI instead of Infrastructure as Code tools like AWS CDK or CloudFormation. This often results in production-unsuitable infrastructure with overly permissive IAM policies.
The AWS MCP Server was built to address these issues, providing a compact toolset that does not consume the model's context window. It supports new APIs within days of launch, ensuring agents stay current.
What This Means
This launch enables developers to build more reliable AI agents that can securely interact with AWS. By using IAM context keys and sandboxed script execution, organizations can enforce fine-grained access controls while reducing security risks.
The move also signals AWS's commitment to making its services AI-ready. For developers, this means faster, more context-efficient agent workflows and reduced overhead in managing permissions and documentation access.
“This is just the beginning,” Chen said. “We expect the AWS MCP Server to become a cornerstone for enterprise AI agent development on AWS.”
For more details, see the features section or the official AWS announcement.
Related Articles
- Red Hat Launches AgentOps Platform to Accelerate AI Agent Deployments in Production
- AWS Launches Managed MCP Server for Secure AI Agent Access to Cloud Services
- Mastering Kubelet Security: Q&A on Kubernetes v1.36 Fine-Grained Authorization
- AWS and Anthropic Deepen AI Collaboration; Meta Joins Graviton Ecosystem for Agentic AI
- How to Give Your AI Agent Secure AWS Access Using the MCP Server
- Amazon Abandons Singapore Grocery Operations, Pivots to Cross-Border Sales
- Malicious PyTorch Lightning Package on PyPI Steals Credentials from Developers
- Microsoft’s Sovereign Cloud Leadership: A Platform for Compliance and Innovation