AWS MCP Servers (Part-1): Your AI Assistant's Gateway to Real-Time AWS Knowledge

Bridge the gap between AI model knowledge cutoffs and current AWS services using MCP servers. This comprehensive guide walks you through setting up AWS Knowledge, Documentation, API, and Core MCP servers to enhance your AI coding assistant with accurate, real-time AWS information.

Ajaykumar K V

AI Engineer | AWS Community Builder

Published Jan 20, 2026

Last modified Jan 20, 2026

Welcome to the first installment of our comprehensive blog series on AWS MCP Servers! In this series, we'll explore how AWS has built an extensive collection of Model Context Protocol (MCP) servers that transform how developers interact with AWS services through AI-powered coding assistants.

Whether you're using Amazon Kiro, Amazon Q Developer, Cursor, Cline, Claude Code,Github Copilot or any other MCP-compatible assistant, these servers unlock powerful capabilities that make cloud development faster, more accurate, and more efficient.

What is the Model Context Protocol (MCP)?

The Model Context Protocol (MCP) is an open protocol developed by Anthropic that enables seamless integration between Large Language Model (LLM) applications and external data sources and tools. Think of it as a universal adapter that allows AI assistants to:

Access real-time documentation and knowledge bases
Execute commands and interact with external services
Retrieve contextual information to improve response quality
Perform specialized tasks through standardized tool interfaces

How MCP Works (High Level Architecture Diagram) :

MCP servers are lightweight programs that expose specific capabilities through the standardized protocol. Your AI coding assistant acts as the MCP client, maintaining connections to one or more MCP servers that provide additional context and tools.

Why AWS MCP Servers?

AWS MCP Servers solve several critical challenges when working with AI assistants for cloud development:

1. Improved Output Quality

By providing relevant, up-to-date information directly in the model's context, MCP servers significantly improve responses for AWS-related tasks:

Reduces hallucinations - No more made-up API parameters or non-existent services
Accurate technical details - Correct syntax, proper configurations
Best practices alignment - Recommendations follow current AWS guidelines

2. Access to Latest Information

AI models have knowledge cutoff dates, but AWS releases new services and features constantly. MCP servers bridge this gap by providing:

Latest API documentation
New service announcements
Updated SDK references
Recent best practices

3. Workflow Automation

MCP servers convert complex AWS workflows into tools that AI assistants can use directly:

Infrastructure as Code generation
Resource management
Cost analysis
Security assessments

Getting Started Category Servers

Let's dive deep into the four essential MCP servers that form the foundation of our AWS AI-assisted development experience.

1. AWS MCP Server (Managed)

The all-in-one solution for AWS interactions

The AWS MCP Server is a remote, fully-managed MCP server hosted by AWS that combines comprehensive AWS API support with access to the latest documentation. This is the recommended starting point for most users.

Key Features

Feature	Description
Comprehensive AWS API Support	Access to all AWS services through natural language
Command Validation	Ensures safety by validating commands before execution
Security Controls	Built-in security mechanisms to prevent harmful operations
Agent SOPs	Standard Operating Procedures help agents complete tasks accurately
Latest Documentation	Real-time access to current AWS documentation

Configuration (Amazon Kiro CLI)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
{

  "mcpServers": {

    "aws-mcp": {

      "command": "uvx",

      "timeout": 100000,

      "transport": "stdio",

      "args": [

        "mcp-proxy-for-aws@latest",

        "https://aws-mcp.us-east-1.api.aws/mcp",

        "--metadata", "AWS_REGION=us-west-2"

      ]

    }

  }

}

Use Cases

Infrastructure Management: "List all my EC2 instances in us-west-2"
Resource Exploration: "Show me the configuration of my production RDS database"
Service Discovery: "What S3 buckets do I have and their storage classes?"
Troubleshooting: "Why is my Lambda function timing out?"

2. AWS Knowledge MCP Server

Your gateway to comprehensive AWS knowledge

The AWS Knowledge MCP Server is a fully-managed remote MCP server that provides structured access to the complete AWS knowledge ecosystem - far beyond just documentation.

Knowledge Sources

The server indexes and provides access to:

Source	Description
AWS Documentation	Complete service documentation
API References	Detailed API specifications
What's New Posts	Latest service announcements
Getting Started Guides	Quick-start tutorials
Builder Center	Developer resources and patterns
Blog Posts	Technical deep-dives and tutorials
Architectural References	Reference architectures
Well-Architected Guidance	Best practices across pillars
Troubleshooting Guides	Error solutions and debugging
AWS Amplify Documentation	Full-stack development guidance
CDK/CloudFormation Docs	Infrastructure as Code resources

Available Tools

1. `search_documentation`

Search across all AWS documentation with optional topic-based filtering.

1
Example: "Search for S3 bucket lifecycle policies"

2. `read_documentation`

Retrieve and convert AWS documentation pages to markdown format.

1
Example: "Read the documentation for Lambda concurrency limits"

3. `recommend`

Get content recommendations for AWS documentation pages.

1
Example: "Recommend related content for EC2 Auto Scaling"

4. `list_regions`

Retrieve a list of all AWS regions with identifiers and names.

1
Example: "List all available AWS regions"

5. `get_regional_availability`

Check regional availability for services, features, SDK APIs, and CloudFormation resources.

1
Example: "Is Amazon Bedrock available in eu-west-1?"

Configuration

1
2
3
4
5
6
7
8
9
{

  "mcpServers": {

    "aws-knowledge-mcp-server": {

      "url": "https://knowledge-mcp.global.api.aws",

      "type": "http",

      "disabled": false

    }

  }

}

For clients that don't support HTTP transport, use the fastmcp proxy:

1
2
3
4
5
6
7
8
{

  "mcpServers": {

    "aws-knowledge-mcp-server": {

      "command": "uvx",

      "args": ["fastmcp", "run", "https://knowledge-mcp.global.api.aws"]

    }

  }

}

Key Benefits

No AWS Account Required: Get started immediately without credentials
No Authentication: Subject to rate limits but freely accessible
Always Current: Continuously updated with latest AWS information
Comprehensive Coverage: Goes beyond docs to include blogs, architectures, and more

Examples:

3. AWS Documentation MCP Server

Local server for AWS documentation access

The AWS Documentation MCP Server is a locally-hosted MCP server that provides direct access to AWS documentation. It's ideal for users who prefer local execution or need offline-capable documentation access.

Features

Feature	Description
Read Documentation	Fetch and convert AWS docs to markdown
Search Documentation	Search using official AWS Search API
Recommendations	Get related content suggestions
China Region Support	Access AWS China documentation
Corporate Network Support	Custom User-Agent for proxy environments

Available Tools

1. `read_documentation`

Fetches an AWS documentation page and converts it to markdown format.

1
read_documentation(url: str) -> str

Example Usage:

1
"Look up documentation on S3 bucket naming rules. Cite your sources."

2. `search_documentation`

Searches AWS documentation using the official AWS Documentation Search API.

1
2
3
4
5
6
7
search_documentation(

    ctx: Context,

    search_phrase: str,

    limit: int,

    product_types: Optional[List[str]],

    guide_types: Optional[List[str]]

) -> SearchResponse

Example Usage:

1
"Search for Lambda cold start optimization techniques"

3. `recommend`

Gets content recommendations for an AWS documentation page.

1
recommend(url: str) -> list[dict]

Example Usage:

1
"Recommend content for page https://docs.aws.amazon.com/AmazonS3/latest/userguide/bucketnamingrules.html"

4. `get_available_services` (China only)

Gets a list of available AWS services in China regions.

Configuration

Standard Configuration (Linux/macOS):

1
2
3
4
5
6
7
8
9
10
11
12
13
14
{

  "mcpServers": {

    "awslabs.aws-documentation-mcp-server": {

      "command": "uvx",

      "args": ["awslabs.aws-documentation-mcp-server@latest"],

      "env": {

        "FASTMCP_LOG_LEVEL": "ERROR",

        "AWS_DOCUMENTATION_PARTITION": "aws"

      },

      "disabled": false,

      "autoApprove": []

    }

  }

}

Windows Configuration:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
{

  "mcpServers": {

    "awslabs.aws-documentation-mcp-server": {

      "disabled": false,

      "timeout": 60,

      "type": "stdio",

      "command": "uv",

      "args": [

        "tool",

        "run",

        "--from",

        "awslabs.aws-documentation-mcp-server@latest",

        "awslabs.aws-documentation-mcp-server.exe"

      ],

      "env": {

        "FASTMCP_LOG_LEVEL": "ERROR",

        "AWS_DOCUMENTATION_PARTITION": "aws"

      }

    }

  }

}

Environment Variables

Variable	Description	Default
`FASTMCP_LOG_LEVEL`	Logging level (DEBUG, INFO, WARNING, ERROR, CRITICAL)	WARNING
`AWS_DOCUMENTATION_PARTITION`	AWS partition (`aws` or `aws-cn` for China)	aws
`MCP_USER_AGENT`	Custom User-Agent for corporate proxies	Chrome-based default

When to Use Local vs Remote

Use Case	Recommended Server
Quick start, no setup	AWS Knowledge MCP Server (Remote)
Corporate network restrictions	AWS Documentation MCP Server (Local)
China region documentation	AWS Documentation MCP Server (Local)
Offline capability needed	AWS Documentation MCP Server (Local)
Comprehensive knowledge access	AWS Knowledge MCP Server (Remote)

Examples:

4. Core MCP Server

Intelligent orchestration and planning for AWS development

The Core MCP Server is a meta-server that provides intelligent planning capabilities and can dynamically proxy other MCP servers based on role-based configurations. It's the command center for complex AWS development workflows.

Key Features

Planning and Orchestration

Provides tools for prompt understanding and translation to AWS services, helping AI assistants create better execution plans.

Dynamic Proxy Server Strategy

The Core MCP Server can automatically import and proxy other MCP servers based on role-based environment variables, creating tailored configurations for specific use cases.

Role-Based Server Configuration

Enable specific roles by setting environment variables. Each role corresponds to a logical grouping of MCP servers:

Role	Description	Included Servers
`aws-foundation`	AWS knowledge and API servers	aws-knowledge-server, aws-api-server
`dev-tools`	Development tools	git-repo-research-server, code-doc-gen-server, aws-knowledge-server
`ci-cd-devops`	CI/CD and DevOps	cdk-server, cfn-server
`container-orchestration`	Container management	eks-server, ecs-server, finch-server
`serverless-architecture`	Serverless development	serverless-server, lambda-tool-server, stepfunctions-tool-server, sns-sqs-server
`analytics-warehouse`	Data analytics	redshift-server, timestream-for-influxdb-server, dataprocessing-server, syntheticdata-server
`data-platform-eng`	Data platform engineering	dynamodb-server, s3-tables-server, dataprocessing-server
`frontend-dev`	Frontend development	frontend-server, nova-canvas-server
`solutions-architect`	Solution architecture	diagram-server, pricing-server, cost-explorer-server, syntheticdata-server, aws-knowledge-server
`finops`	Financial operations	cost-explorer-server, pricing-server, cloudwatch-server, billing-cost-management-server
`monitoring-observability`	Monitoring	cloudwatch-server, cloudwatch-appsignals-server, prometheus-server, cloudtrail-server
`caching-performance`	Caching	elasticache-server, memcached-server
`security-identity`	Security and identity	iam-server, support-server, well-architected-security-server
`sql-db-specialist`	SQL databases	postgres-server, mysql-server, aurora-dsql-server, redshift-server
`nosql-db-specialist`	NoSQL databases	dynamodb-server, documentdb-server, keyspaces-server, neptune-server
`timeseries-db-specialist`	Time series databases	timestream-for-influxdb-server, prometheus-server, cloudwatch-server
`messaging-events`	Messaging	sns-sqs-server, mq-server
`healthcare-lifesci`	Healthcare and life sciences	healthomics-server

Available Tools

`prompt_understanding`

Helps provide guidance and planning support when building AWS Solutions for a given prompt.

Configuration

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
{

  "mcpServers": {

    "awslabs.core-mcp-server": {

      "command": "uvx",

      "args": ["awslabs.core-mcp-server@latest"],

      "env": {

        "FASTMCP_LOG_LEVEL": "ERROR",

        "aws-foundation": "true",

        "solutions-architect": "true"

      },

      "autoApprove": [],

      "disabled": false

    }

  }

}

Windows Configuration:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
{

  "mcpServers": {

    "awslabs.core-mcp-server": {

      "disabled": false,

      "timeout": 60,

      "type": "stdio",

      "command": "uv",

      "args": [

        "tool",

        "run",

        "--from",

        "awslabs.core-mcp-server@latest",

        "awslabs.core-mcp-server.exe"

      ],

      "env": {

        "FASTMCP_LOG_LEVEL": "ERROR",

        "AWS_PROFILE": "your-aws-profile",

        "AWS_REGION": "us-east-1",

        "aws-foundation": "true",

        "solutions-architect": "true"

      }

    }

  }

}

Benefits of the Proxy Server Strategy

Simplified Configuration: Enable multiple servers with a single environment variable
Reduced Duplication: Servers are imported only once, even if needed by multiple roles
Tailored Experience: Create custom server configurations for specific use cases
Flexible Deployment: Easily switch between different server configurations

5. AWS API MCP Server

Direct AWS CLI access through your AI assistant

The AWS API MCP Server enables AI assistants to interact with AWS services through AWS CLI commands, providing programmatic access to manage your AWS infrastructure.

Key Features

Feature	Description
Comprehensive AWS CLI Support	All commands from the latest AWS CLI version
Command Selection Help	Assists in selecting appropriate CLI commands
Command Validation	Validates commands before execution
Hallucination Protection	Only executes valid AWS CLI commands
Security-First Design	Multiple layers of protection
Read-Only Mode	Optional restriction to read-only operations

Available Tools

1. `call_aws`

Executes AWS CLI commands with validation and proper error handling.

1
Example: "aws ec2 describe-instances --region us-west-2"

2. `suggest_aws_commands`

Suggests AWS CLI commands based on natural language queries. Provides the 5 most likely CLI commands with complete parameter descriptions.

1
Example: "How do I list all S3 buckets with their sizes?"

3. `get_execution_plan` (Experimental)

Provides structured, step-by-step guidance for complex AWS tasks through agent scripts.

Configuration

Linux/macOS:

1
2
3
4
5
6
7
8
9
10
11
12
13
{

  "mcpServers": {

    "awslabs.aws-api-mcp-server": {

      "command": "uvx",

      "args": ["awslabs.aws-api-mcp-server@latest"],

      "env": {

        "AWS_REGION": "us-east-1"

      },

      "disabled": false,

      "autoApprove": []

    }

  }

}

Windows:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
{

  "mcpServers": {

    "awslabs.aws-api-mcp-server": {

      "command": "uvx",

      "args": [

        "--from",

        "awslabs.aws-api-mcp-server@latest",

        "awslabs.aws-api-mcp-server.exe"

      ],

      "env": {

        "AWS_REGION": "us-east-1"

      },

      "disabled": false,

      "autoApprove": []

    }

  }

}

Environment Variables

Variable	Description	Default
`AWS_REGION`	Default AWS region	us-east-1
`AWS_API_MCP_PROFILE_NAME`	AWS Profile for credentials	default
`READ_OPERATIONS_ONLY`	Restrict to read-only operations	false
`REQUIRE_MUTATION_CONSENT`	Ask consent before write operations	false
`AWS_API_MCP_WORKING_DIR`	Working directory for operations	Platform temp dir

Security Considerations

Single User Only: Not designed for multi-tenant environments
IAM-Based Access Control: Use appropriate IAM roles (ReadOnlyAccess for exploration, AdministratorAccess for full control)
Read-Only Mode: Set READ_OPERATIONS_ONLY=true for safe exploration
Mutation Consent: Set REQUIRE_MUTATION_CONSENT=true for explicit approval of write operations

Examples :

Practical Example: Building a Cost-Optimized Architecture

Let's walk through a real-world example of how these MCP servers work together to help you design and implement a cost-optimized AWS architecture.

Scenario

You're tasked with designing a serverless API backend for a new application. You want to ensure it follows AWS best practices and is cost-effective.

Step 1: Research Best Practices (AWS Knowledge MCP Server)

Prompt:

1
2
"What are the best practices for building serverless APIs on AWS? 

Include information about API Gateway, Lambda, and DynamoDB."

The AWS Knowledge MCP Server searches across documentation, blogs, and Well-Architected guidance to provide comprehensive recommendations.

Step 2: Check Regional Availability (AWS Knowledge MCP Server)

Prompt:

1
"Check if Lambda SnapStart and API Gateway HTTP APIs are available in eu-central-1"

The server uses get_regional_availability to confirm service availability in your target region.

Step 3: Get Specific Documentation (AWS Documentation MCP Server)

Prompt:

1
"Read the documentation for Lambda provisioned concurrency and when to use it vs SnapStart"

The local documentation server fetches and formats the specific documentation you need.

Step 4: Plan the Architecture (Core MCP Server)

Prompt:

1
2
"Help me plan a serverless architecture with API Gateway, Lambda, and DynamoDB 

for a REST API handling 1000 requests per second"

The Core MCP Server's prompt_understanding tool helps translate your requirements into specific AWS services and configurations.

Step 5: Explore Existing Resources (AWS API MCP Server)

Prompt:

1
"List my existing Lambda functions and their configurations in eu-central-1"

The API server executes the appropriate AWS CLI commands to show your current infrastructure.

Step 6: Estimate Costs (Using Cost MCP Servers - covered in Part 5)

Prompt:

1
"Estimate the monthly cost for this architecture with 1000 RPS average load"

Installation Quick Reference

Prerequisites

Install uv (Python package manager):### macOS/Linux

1
curl -LsSf https://astral.sh/uv/install.sh | sh

Windows (PowerShell)

1
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

Install Python 3.10+:

1
uv python install 3.10

Recommended Starter Configuration

For most users, start with this configuration combining the essential servers:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
{

  "mcpServers": {

    "aws-knowledge-mcp-server": {

      "url": "https://knowledge-mcp.global.api.aws",

      "type": "http",

      "disabled": false

    },

    "awslabs.aws-api-mcp-server": {

      "command": "uvx",

      "args": ["awslabs.aws-api-mcp-server@latest"],

      "env": {

        "AWS_REGION": "us-east-1",

        "READ_OPERATIONS_ONLY": "true"

      },

      "disabled": false

    },

    "awslabs.aws-documentation-mcp-server": {

      "command": "uvx",

      "args": ["awslabs.aws-documentation-mcp-server@latest"],

      "env": {

        "FASTMCP_LOG_LEVEL": "ERROR"

      },

      "disabled": false

    }

  }

}

Configuration File Locations

IDE/Tool	Configuration Path
Kiro (Global)	`~/.kiro/settings/mcp.json`
Kiro (Project)	`.kiro/settings/mcp.json`
Cursor	`.cursor/mcp.json`
VS Code	`.vscode/mcp.json`
Claude Code	`.mcp.json`
Windsurf	`~/.codeium/windsurf/mcp_config.json`

Summary

In this first part of our AWS MCP Servers series, we've covered the foundational servers that every AWS developer should know:

Server	Type	Best For
AWS MCP Server	Remote/Managed	All-in-one AWS interactions
AWS Knowledge MCP Server	Remote/Managed	Comprehensive AWS knowledge access
AWS Documentation MCP Server	Local	Direct documentation access, corporate networks
Core MCP Server	Local	Orchestration and role-based server management
AWS API MCP Server	Local	Direct AWS CLI command execution

These servers form the foundation for AI-assisted AWS development, providing your coding assistant with the knowledge and capabilities it needs to help you build better cloud applications.

What's Next?

In Part 2: Infrastructure & Deployment, we'll explore the MCP servers that help you build and manage AWS infrastructure:

AWS IaC MCP Server
AWS CloudFormation MCP Server
AWS Terraform MCP Server
Amazon EKS MCP Server
Amazon ECS MCP Server
AWS Serverless MCP Server
And more!

Resources

GitHub Repository: github.com/awslabs/mcp
Documentation Site: awslabs.github.io/mcp
Model Context Protocol: modelcontextprotocol.io
AWS Blog: aws.amazon.com/blogs

Thank you for reading Part 1 of the AWS MCP Servers Blog Series! I hope this guide helps you get started with enhancing your AI coding assistant with real-time AWS knowledge and capabilities.

If you found this helpful, please share it with your team and fellow developers. I'd love to hear about your experiences using AWS MCP Servers—feel free to reach out with questions, feedback, or success stories.

Stay tuned for Part 2, where we'll dive into Infrastructure & Deployment servers!

# ai # ai-agents # ai-engineering # mcp # kiro

Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.

Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article

Builder Center

AWS MCP Servers (Part-1): Your AI Assistant's Gateway to Real-Time AWS Knowledge

Ajaykumar K V

What is the Model Context Protocol (MCP)?

How MCP Works (High Level Architecture Diagram) :

Why AWS MCP Servers?

1. Improved Output Quality

2. Access to Latest Information

3. Workflow Automation

Getting Started Category Servers

1. AWS MCP Server (Managed)

Key Features

Configuration (Amazon Kiro CLI)

Use Cases

2. AWS Knowledge MCP Server

Knowledge Sources

Available Tools

1. search_documentation

2. read_documentation

3. recommend

4. list_regions

5. get_regional_availability

Configuration

Key Benefits

3. AWS Documentation MCP Server

Features

Available Tools

1. read_documentation

2. search_documentation

3. recommend

4. get_available_services (China only)

Configuration

Environment Variables

When to Use Local vs Remote

4. Core MCP Server

Key Features

Planning and Orchestration

Dynamic Proxy Server Strategy

Role-Based Server Configuration

Available Tools

prompt_understanding

Configuration

Benefits of the Proxy Server Strategy

5. AWS API MCP Server

Key Features

Available Tools

1. call_aws

2. suggest_aws_commands

3. get_execution_plan (Experimental)

Configuration

Environment Variables

Security Considerations

Practical Example: Building a Cost-Optimized Architecture

Scenario

Step 1: Research Best Practices (AWS Knowledge MCP Server)

Step 2: Check Regional Availability (AWS Knowledge MCP Server)

Step 3: Get Specific Documentation (AWS Documentation MCP Server)

Step 4: Plan the Architecture (Core MCP Server)

Step 5: Explore Existing Resources (AWS API MCP Server)

Step 6: Estimate Costs (Using Cost MCP Servers - covered in Part 5)

Installation Quick Reference

Prerequisites

Windows (PowerShell)

Recommended Starter Configuration

Configuration File Locations

Summary

What's Next?

Resources

1. `search_documentation`

2. `read_documentation`

3. `recommend`

4. `list_regions`

5. `get_regional_availability`

1. `read_documentation`

2. `search_documentation`

3. `recommend`

4. `get_available_services` (China only)

`prompt_understanding`

1. `call_aws`

2. `suggest_aws_commands`

3. `get_execution_plan` (Experimental)