AWS Knowledge MCP Server

What is AWS Knowledge MCP Server?

AWS Knowledge MCP Server is part of the AWS MCP Servers suite, operating as a remote, fully managed MCP server hosted by AWS. It exposes an intelligent interface for AI assistants and developer tools to retrieve, search, and interact with the latest AWS official documentation, including API references, blog posts, architectural patterns, and prescriptive guidance. By centralizing and continuously updating this knowledge base, it ensures your AI agents or workspace always leverage the most accurate and relevant AWS resources, far beyond what static foundation models can provide.

How to Configure AWS Knowledge MCP Server

AWS Knowledge MCP Server is provided as a fully managed remote service, so there is no installation or server setup required on your side. To connect, simply add its entry to your MCP client configuration. Here's a sample configuration block for various environments (replace or include with your own mcp.json settings):

{
  "mcpServers": {
    "awslabs.aws-knowledge-mcp-server": {
      "type": "remote",
      "host": "api.aws.amazon.com",
      "port": 443,
      "protocol": "https",
      "env": {
        "AWS_PROFILE": "your-aws-profile",
        "AWS_REGION": "us-east-1"
      }
    }
  }
}

For clients such as Amazon Q CLI, Cline, Cursor, Windsurf, or VS Code, you can add this block to your respective configuration file (e.g., ~/.aws/amazonq/mcp.json, .cursor/mcp.json, or .vscode/mcp.json). No local Python installation or Docker container is needed. Just ensure your client’s AWS credentials and permissions are configured.

How to Use AWS Knowledge MCP Server

Once the AWS Knowledge MCP Server is added to your MCP configuration, your AI assistant or IDE can start using it automatically. You can:

• Prompt your AI agent with queries like “Search the official AWS docs for Lambda concurrency settings” or “What’s new with S3 APIs?”
• Request best practices or architectural guidance, e.g., “List Well-Architected recommendations for serverless applications.”
• Use conversational interfaces (Amazon Q, Cursor, Cline, etc.) and prompt them to consult the AWS Knowledge MCP Server for relevant documentation, blog posts, or reference material.
• For programmatic interaction, MCP clients use standardized endpoints such as /resources/search, /resources/read, or /tools/call to access documentation and tooling functions exposed by the server. Approval flows in your client may prompt you to confirm tool usage. By default, the server handles all infrastructure and content updates—no additional action is required from your side.

Key Features

• Real-Time Access to AWS Knowledge: Instantly retrieve the most current AWS documentation, API references, and new features.
• Comprehensive Content Coverage: Includes Docs, API references, What's New posts, Getting Started guides, Builder Center content, official blog posts, architectural frameworks, and Well-Architected best practices.
• Best Practice Guidance: Surfaces prescriptive, up-to-date recommendations tailored to your use case.
• Zero Setup: No installation—connect and use as a managed remote service.
• Integration with Major AI Assistant Clients: Works seamlessly with Amazon Q CLI, Cline, Cursor, Windsurf, Claude Desktop, and other tools supporting the MCP standard.
• Secure and Scalable: Hosted by AWS with enterprise-grade security and automatic updates.

Use Cases

• LLM-Powered Developer Assistants: AI coding assistants can access AWS documentation to generate accurate, context-aware code for any AWS service or SDK.
• Interactive Documentation Retrieval: Instantly answer developer queries about AWS services, APIs, and best practices during coding, troubleshooting, or architectural design.
• Up-to-Date Knowledge Search: Stay current on new AWS announcements, avoiding limitations of models trained on outdated datasets.
• Integrated Well-Architected Reviews: Guide teams on compliance, reliability, performance, and security standards using the latest AWS Well-Architected content.
• Conversational Q&A Bots: Build chatbots or knowledge agents that answer AWS-related questions using the freshest official materials.

FAQ

Q: Do I need to install or host my own AWS Knowledge MCP Server instance?
A: No. AWS Knowledge MCP Server is a remote, fully managed service provided by AWS. You only need to configure your MCP client to connect—no server installation or infrastructure deployment is needed.

Q: Is the AWS Knowledge MCP Server kept up to date with the latest AWS documentation?
A: Yes. Content on the server is continuously synchronized with AWS, ensuring it always serves the most recent documentation, API changes, and best practice updates.

Q: What permissions or AWS credentials are required to use the server?
A: Reading public documentation requires only basic AWS credentials (e.g., read access). Accessing private or account-specific knowledge (if enabled) might require additional permissions as documented per client.

Q: Can I use the AWS Knowledge MCP Server offline or locally?
A: No. As a managed remote service, you must have internet access to use the AWS Knowledge MCP Server. For local/offline access, you may run other local MCP servers, but knowledge content will not update in real-time.

Q: With what clients can I use AWS Knowledge MCP Server?
A: It works with any MCP-compatible client, including Amazon Q CLI, Cline, Cursor, Windsurf, Claude Desktop, and VS Code.

Q: What types of AWS content can I search or retrieve through the server?
A: You can access official documentation, API references, Well-Architected guidance, blog articles, architectural patterns, and "What's New" updates.