AWS MCP Server

Official, AWS-operated remote MCP server that gives coding agents authenticated access to 300+ AWS services, sandboxed code execution, and AWS documentation.

Description

The AWS MCP Server is an AWS-managed remote MCP server and the core component of the Agent Toolkit for AWS. It does not run locally; it runs as a hosted endpoint at AWS and is used through a regular AWS account.

Per the documentation, an agent can call any of 300+ AWS services through a single authenticated endpoint, run Python code against AWS services in an isolated sandbox with no access to the local filesystem, search and retrieve current AWS documentation, and pull in project-specific instructions via loadable "agent skills". Documentation search and skill retrieval work without AWS credentials; actual AWS API calls require authentication through your own AWS permissions (IAM).

AWS offers two authentication paths. With OAuth, a human authenticates in the browser, or an automated agent authenticates by requesting a token; this works with web clients (such as Claude.ai) and most IDE, terminal, or desktop clients without installing anything locally. With SigV4, you first sign in via the AWS CLI (aws login) and then use the separate MCP Proxy for AWS to sign requests with your own AWS credentials — per the documentation, this is the recommended path for terminal/IDE coding agents such as Claude Code, Kiro, or Codex, and the only way to switch between multiple AWS accounts in one session or to hide write-capable tools entirely via read-only mode.

The server is currently available in two regions: US East (N. Virginia) at https://aws-mcp.us-east-1.api.aws/mcp and Europe (Frankfurt) at https://aws-mcp.eu-central-1.api.aws/mcp.

Registering via OAuth in Claude Code:

Code
claude mcp add aws-mcp https://aws-mcp.us-east-1.api.aws/mcp --transport http

Registering via SigV4 (e.g. Cursor, Claude Desktop) through the MCP Proxy for AWS:

json
{
  "mcpServers": {
    "aws-mcp": {
      "command": "uvx",
      "args": [
        "mcp-proxy-for-aws-cli@latest",
        "https://aws-mcp.us-east-1.api.aws/mcp",
        "--metadata", "AWS_REGION=us-west-2"
      ]
    }
  }
}

AWS explicitly recommends removing older, separate servers such as aws-api-mcp-server and aws-knowledge-mcp-server from client configuration to avoid tool conflicts, since the AWS MCP Server merges these capabilities and, per the documentation, adds extra IAM security controls.

Why one endpoint instead of many separate servers

Before the AWS MCP Server, teams wanting to reach several AWS services often had to run and configure multiple separate MCP servers in parallel. The central endpoint solves this by bundling over 300 services behind a single authenticated connection. For users, this means less configuration overhead and a single place for permission management through IAM, instead of maintaining access rights across several independent servers.

Access and Logging

All capabilities are available through a single endpoint. Agents can search the documentation and retrieve service information without signing in. For AWS API calls, Python scripts in a sandbox (an isolated area) and curated skills, agents authenticate with your existing IAM credentials. Access can be controlled through IAM, metrics go to CloudWatch, and CloudTrail logs all API calls so they stay traceable.

Who the server fits

The server is particularly useful for development teams that already actively use AWS for infrastructure, databases, or AI services and want to handle recurring cloud operations directly from their coding agent — for example debugging failed Lambda invocations, looking up current service limits, or running small analysis scripts in the isolated sandbox. For organizations with strict compliance requirements, the SigV4 path via the MCP Proxy for AWS is relevant because it fits cleanly into existing IAM roles and CloudTrail logging.

Requirements

An AWS account. For OAuth, an MCP client with OAuth support is enough, and its IAM role/user needs the AWSMCPSignInOAuthAccessPolicy managed policy attached. For SigV4, you also need the AWS CLI (version 2.32.0 or later) and uv/uvx for the MCP Proxy for AWS.

Installation instructions

The AWS MCP Server runs as an AWS-managed remote service — unlike many MCP servers, there is nothing to launch locally via npx or Docker. Setup is mainly about choosing the right authentication path.

Choosing OAuth or SigV4

OAuth suits you if you want no local installation, your client only supports remote MCP servers, or you use a single AWS account. SigV4 is the right choice for terminal/IDE coding agents, for switching between multiple AWS accounts in one session, for enforcing read-only mode, or if your organization restricts browser-based OAuth sign-in.

Step 1: Setting up the AWS MCP Server via OAuth

First attach the managed policy to your IAM role:

bash
aws iam attach-role-policy \
  --role-name MyRole \
  --policy-arn arn:aws:iam::aws:policy/AWSMCPSignInOAuthAccessPolicy

Then configure the endpoint URL in your client, for example in Claude Code:

bash
claude mcp add aws-mcp https://aws-mcp.us-east-1.api.aws/mcp --transport http

If your client only supports OAuth with an explicit trigger (such as Claude Desktop or Cursor), append ?oauth=initialize to the URL. On the first tool call, a browser window opens for AWS sign-in.

Step 2: Setting up the AWS MCP Server via SigV4

First install the AWS CLI (2.32.0 or later) and sign in:

bash
aws login
aws sts get-caller-identity

Then install uv and configure the MCP Proxy for AWS in your client configuration:

json
{
  "mcpServers": {
    "aws-mcp": {
      "command": "uvx",
      "args": [
        "mcp-proxy-for-aws-cli@latest",
        "https://aws-mcp.us-east-1.api.aws/mcp",
        "--metadata", "AWS_REGION=us-west-2"
      ]
    }
  }
}

Replacing older AWS MCP servers

If you already have aws-api-mcp-server or aws-knowledge-mcp-server configured, remove those entries from your client configuration to avoid tool conflicts — the AWS MCP Server covers both areas of functionality.

Step 3: Scoping permissions and verifying setup

Before setup, prepare an IAM role as narrow as possible, allowing only the services actually needed. After starting, test with a harmless question like "What AWS Regions are available?" to confirm tools loaded, and check CloudTrail to see which actions the agent actually performs.

Source: docs.aws.amazon.com/agent-toolkit, checked on 2026-09-05.

claude mcp add aws-mcp https://aws-mcp.us-east-1.api.aws/mcp --transport http

Authentication

Two paths: OAuth (browser-based login, or token requests for automated agents, no local process required) or SigV4 via the AWS CLI plus the MCP Proxy for AWS, which signs requests with your own AWS credentials. Only SigV4 supports switching between multiple AWS accounts in one session.

Required access permissions

Effective permissions exactly match those of the IAM role or user, or the OAuth session, in use — the server itself grants no additional rights. Depending on the assigned IAM permissions, read and write access to any of the 300+ supported AWS services is possible. Per AWS, access can additionally be restricted via IAM condition keys specific to the MCP server.

Transmitted or stored data

Requests and tool parameters are sent over HTTPS to the AWS-operated endpoint in the selected region. Responses can include resource data from any AWS services used, and are passed to the MCP client and its language model. Per the documentation, Python code runs in an isolated AWS sandbox with no access to the local filesystem. AWS logs usage via CloudWatch and CloudTrail.

Security risks

Because the server can call any of the 300+ AWS services, an overly broad IAM role or a compromised OAuth token enables far-reaching actions in the AWS account, up to changes on production resources. Grant only minimal, task-scoped IAM permissions, use IAM condition keys for the MCP server where possible, enable read-only mode for analysis-only work with SigV4, and monitor CloudTrail logs for unusual activity.

License and costs

License
Verwalteter AWS-Dienst ohne eigenes Open-Source-Lizenzmodell; Begleit-Tools (MCP Proxy for AWS, agent-toolkit-for-aws) unter Apache-2.0.
Cost
free

Per AWS, the AWS MCP Server itself carries no additional charge — only the AWS resources and API calls actually used through the server are billed at regular AWS rates.

Alternatives

Not recorded yet.

At a glance

Status
Official server
Deployment
Remote
Current version
Not recorded yet.
GitHub stars
3,497
Last reviewed
07.09.2026

Repository and documentation

Categories

Supported clients