Introducing AWS MCP Servers for code assistants (Half 1)

We’re excited to announce the open supply launch of AWS MCP Servers for code assistants — a collection of specialised Mannequin Context Protocol (MCP) servers that deliver Amazon Net Companies (AWS) finest practices on to your improvement workflow. Our specialised AWS MCP servers mix deep AWS information with agentic AI capabilities to speed up improvement throughout key areas. Every AWS MCP Server focuses on a selected area of AWS finest practices, working collectively to offer complete steerage all through your improvement journey.

This publish is the primary in a collection overlaying AWS MCP Servers. On this publish, we stroll by way of how these specialised MCP servers can dramatically cut back your improvement time whereas incorporating safety controls, price optimizations, and AWS Effectively-Architected finest practices into your code. Whether or not you’re an skilled AWS developer or simply getting began with cloud improvement, you’ll uncover how one can use AI-powered coding assistants to sort out frequent challenges corresponding to advanced service configurations, infrastructure as code (IaC) implementation, and information base integration. By the top of this publish, you’ll perceive how one can begin utilizing AWS MCP Servers to remodel your improvement workflow and ship higher options, sooner.

If you wish to get began immediately, skip forward to the part “From Idea to working code in minutes.”

AI is reworking how we construct software program, creating alternatives to dramatically speed up improvement whereas enhancing code high quality and consistency. In the present day’s AI assistants can perceive advanced necessities, generate production-ready code, and assist builders navigate technical challenges in actual time. This AI-driven method is especially useful in cloud improvement, the place builders have to orchestrate a number of companies whereas sustaining safety, scalability, and cost-efficiency.

Builders want code assistants that perceive the nuances of AWS companies and finest practices. Specialised AI brokers can tackle these wants by:

• Offering contextual steerage on AWS service choice and configuration
• Optimizing compliance with safety finest practices and regulatory necessities
• Selling essentially the most environment friendly utilization and cost-effective options
• Automating repetitive implementation duties with AWS particular patterns

This method means builders can deal with innovation whereas AI assistants deal with the undifferentiated heavy lifting of coding. Whether or not you’re utilizing Amazon Q, Amazon Bedrock, or different AI instruments in your workflow, AWS MCP Servers complement and improve these capabilities with deep AWS particular information that will help you construct higher options sooner.

Mannequin Context Protocol (MCP) is a standardized open protocol that allows seamless interplay between giant language fashions (LLMs), information sources, and instruments. This protocol permits AI assistants to make use of specialised tooling and to entry domain-specific information by extending the mannequin’s capabilities past its built-in information—all whereas holding delicate information native. By means of MCP, general-purpose LLMs can now seamlessly entry related information past preliminary coaching information and be successfully steered in the direction of desired outputs by incorporating particular context and finest practices.

Speed up constructing on AWS

What in case your AI assistant may immediately entry deep AWS information, understanding each AWS service, finest apply, and architectural sample? With MCP, we will rework general-purpose LLMs into AWS specialists by connecting them to specialised information servers. This opens up thrilling new prospects for accelerating cloud improvement whereas sustaining safety and following finest practices.

Construct on AWS in a fraction of the time, with finest practices routinely utilized from the primary line of code. Skip hours of documentation analysis and instantly entry ready-to-use patterns for advanced companies corresponding to Amazon Bedrock Information Bases. Our MCP Servers will assist you write well-architected code from the beginning, implement AWS companies appropriately the primary time, and deploy options which might be safe, observable, and cost-optimized by design. Rework the way you construct on AWS at the moment.

• Implement AWS finest practices routinely – Write well-architected code from the beginning with built-in safety controls, correct observability, and optimized useful resource configurations
• Reduce analysis time dramatically – Cease spending hours studying documentation. Our MCP Servers present contextually related steerage for implementing AWS companies appropriately, addressing frequent pitfalls routinely
• Entry ready-to-use patterns immediately – Use pre-built AWS CDK constructs, Amazon Bedrock Brokers schema turbines, and Amazon Bedrock Information Bases integration templates that observe AWS finest practices from the beginning
• Optimize price proactively – Stop over-provisioning as you design your answer by getting cost-optimization suggestions and producing a complete price report to investigate your AWS spending earlier than deployment

To show this imaginative and prescient into actuality and make AWS improvement sooner, safer, and extra environment friendly, we’ve created AWS MCP Servers—a collection of specialised AWS MCP Servers that deliver AWS finest practices on to your improvement workflow. Our specialised AWS MCP Servers mix deep AWS information with AI capabilities to speed up improvement throughout key areas. Every AWS MCP Server focuses on a selected area of AWS finest practices, working collectively to offer complete steerage all through your improvement journey.

Overview of domain-specific MCP Servers for AWS improvement

Our specialised MCP Servers are designed to cowl distinct points of AWS improvement, every bringing deep information to particular domains whereas working in live performance to ship complete options:

• Core – The inspiration server that gives AI processing pipeline capabilities and serves as a central coordinator. It helps present clear plans for constructing AWS options and may federate to different MCP servers as wanted.
• AWS Cloud Growth Package (AWS CDK) – Delivers AWS CDK information with instruments for implementing finest practices, safety configurations with cdk-nag, Powertools for AWS Lambda integration, and specialised constructs for generative AI companies. It makes certain infrastructure as code (IaC) follows AWS Effectively-Architected ideas from the beginning.
• Amazon Bedrock Information Bases – Allows seamless entry to Amazon Bedrock Information Bases so builders can question enterprise information with pure language, filter outcomes by information supply, and use reranking for improved relevance.
• Amazon Nova Canvas – Gives picture technology capabilities utilizing Amazon Nova Canvas by way of Amazon Bedrock, enabling the creation of visuals from textual content prompts and coloration palettes—good for mockups, diagrams, and UI design ideas.
• Value – Analyzes AWS service prices and generates complete price reviews, serving to builders perceive the monetary implications of their architectural choices and optimize for cost-efficiency.

Stipulations

To finish the answer, you’ll want to have the next stipulations in place:

• uv bundle supervisor
• Set up Python utilizing uv python set up 3.13
• AWS credentials with acceptable permissions
• An MCP-compatible LLM consumer (corresponding to Anthropic’s Claude for Desktop, Cline, Amazon Q CLI, or Cursor)

From idea to working code in minutes

You possibly can obtain the AWS MCP Servers on GitHub or by way of the PyPI bundle supervisor. Right here’s how one can get began utilizing your favourite code assistant with MCP help.

To put in MCP Servers, enter the next code:

# Set up and setup the MCP servers
{
  "mcpServers": {
    "awslabs.core-mcp-server": {
      "command": "uvx",
      "args": [
        "awslabs.core-mcp-server@latest"
      ],
      "env": {
        "FASTMCP_LOG_LEVEL": "ERROR",
        "MCP_SETTINGS_PATH": "path to your mcp server settings"
      },
      "autoApprove": [],
      "disabled": false
    },
    "awslabs.bedrock-kb-retrieval-mcp-server": {
      "command": "uvx",
      "args": ["awslabs.bedrock-kb-retrieval-mcp-server@latest"],
      "env": {
        "AWS_PROFILE": "your-aws-profile",
        "AWS_REGION": "us-east-1"
      }
    },
    "awslabs.cdk-mcp-server": {
      "command": "uvx",
      "args": ["awslabs.cdk-mcp-server@latest"],
      "env": {
        "FASTMCP_LOG_LEVEL": "ERROR"
      }
    },
    "awslabs.cost-analysis-mcp-server": {
      "command": "uvx",
      "args": ["awslabs.cost-analysis-mcp-server@latest"],
      "env": {
        "AWS_PROFILE": "your-aws-profile",
        "FASTMCP_LOG_LEVEL": "ERROR"
      }
    },
    "awslabs.nova-canvas-mcp-server": {
      "command": "uvx",
      "args": ["awslabs.nova-canvas-mcp-server@latest"],
      "env": {
        "AWS_PROFILE": "your-aws-profile",
        "AWS_REGION": "us-east-1"
      }
    }
  }
}

AWS MCP Servers in motion

Right here’s how AWS MCP servers rework the event expertise:

Developer: “I have to construct an AI-powered chatbot utilizing Amazon Bedrock that may reply questions from our firm’s information base. I additionally wish to add a instrument for the chatbot to name our inner API.”

Core: “I’ll assist you construct an Amazon Bedrock Information Bases chatbot with API integration. Let’s create an structure that makes use of Amazon Bedrock Brokers with a customized motion group to name your inner API.”

Core generates a complete structure diagram displaying the information base integration, Amazon Bedrock Brokers configuration with motion teams, API connectivity, and information move between elements.

AWS CDK: “Right here’s the infrastructure code to your chatbot with the Amazon Bedrock Brokers motion group. I’ve included correct IAM roles, safety controls, and Lambda Powertools for observability.”

// CDK code with Effectively-Architected finest practices
 const knowledgeBase = new BedrockKnowledgeBase(this, 'CompanyKB', {
   embeddingModel: BedrockFoundationModel.TITAN_EMBED_TEXT_V1,
   vectorStore: new OpenSearchServerlessVectorStore(this, 'VectorStore', {
     encryption: OpenSearchEncryption.KMS,
     ebs: OpenSearchEbsOptions.provisioned(100, OpenSearchVolumeType.GP3)
   })
 });

 // Motion Group for API integration
 const actionGroupSchema = new BedrockAgentSchema(this, 'ApiActionSchema', {
   schemaPath: 'schemas/api-action-schema.json',
   // Schema will likely be auto-generated out of your Lambda code
   generateFromLambda: {
     lambdaPath: 'lambda/api-actions.py',
     outputPath: 'schemas/api-action-schema.json'
   }
 });

 const apiActionGroup = new BedrockAgentActionGroup(this, 'ApiActionGroup', {
   actionGroupName: 'InternalApiActions',
   description: 'Actions to work together with inner API',
   actionGroupExecutor: new BedrockAgentLambdaExecutor(apiActionsLambda),
   actionGroupSchema: actionGroupSchema
 });

The CDK MCP Server generates full AWS CDK code to deploy all the answer. It routinely runs cdk-nag to establish potential safety points and offers remediation steps for every discovering, ensuring that the infrastructure follows AWS Effectively-Architected finest practices.

Amazon Bedrock Information Bases retrieval: “I’ve configured the optimum settings to your information base queries, together with correct reranking for improved relevance.”

Amazon Bedrock Information Bases MCP Server demonstrates how one can construction queries to the information base for max relevance, offers pattern code for filtering by information supply, and exhibits how one can combine the information base responses with the chatbot interface.

Amazon Nova Canvas: “To reinforce your chatbot’s capabilities, I’ve created visualizations that may be generated on demand when customers request information explanations.”

Amazon Nova Canvas MCP server generates pattern pictures displaying how Amazon Nova Canvas can create charts, diagrams, and visible explanations primarily based on information base content material, making advanced info extra accessible to customers.

Value Evaluation: “Primarily based in your anticipated utilization patterns, right here’s the estimated month-to-month price breakdown and optimization suggestions.”

The Value Evaluation MCP Server generates an in depth price evaluation report displaying projected bills for every AWS service, identifies price optimization alternatives corresponding to reserved capability for Amazon Bedrock, and offers particular suggestions to scale back prices with out impacting efficiency.

With AWS MCP Servers, what would sometimes take days of analysis and implementation is accomplished in minutes, with higher high quality, safety, and cost-efficiency than handbook improvement in that very same time.

Greatest practices for MCP-assisted improvement

To maximise the advantages of MCP assisted improvement whereas sustaining safety and code high quality, builders ought to observe these important pointers:

• All the time overview generated code for safety implications earlier than deployment
• Use MCP Servers as accelerators, not replacements for developer judgment and experience
• Hold MCP Servers up to date with the newest AWS safety finest practices
• Observe the precept of least privilege when configuring AWS credentials
• Run safety scanning instruments on generated infrastructure code

Arising within the collection

This publish launched the foundations of AWS MCP Servers and the way they speed up AWS improvement by way of specialised, AWS particular MCP Servers. In upcoming posts, we’ll dive deeper into:

• Detailed walkthroughs of every MCP server’s capabilities
• Superior patterns for integrating AWS MCP Servers into your improvement workflow
• Actual-world case research displaying AWS MCP Servers’ impression on improvement velocity
• Easy methods to lengthen AWS MCP Servers with your individual customized MCP servers

Keep tuned to find out how AWS MCP Servers can rework your particular AWS improvement situations and assist you construct higher options sooner. Go to our GitHub repository or Pypi bundle supervisor to discover instance implementations and get began at the moment.

In regards to the Authors

Jimin Kim is a Prototyping Architect on the AWS Prototyping and Cloud Engineering (PACE) staff, primarily based in Los Angeles. With specialties in Generative AI and SaaS, she loves serving to her clients succeed of their enterprise. Exterior of labor, she cherishes moments along with her spouse and three cute calico cats.

Pranjali Bhandari is a part of the Prototyping and Cloud Engineering (PACE) staff at AWS, primarily based within the San Francisco Bay Space. She focuses on Generative AI, distributed programs, and cloud computing. Exterior of labor, she loves exploring numerous mountaineering trails, biking, and having fun with high quality household time along with her husband and son.

Laith Al-Saadoon is a Principal Prototyping Architect on the Prototyping and Cloud Engineering (PACE) staff. He builds prototypes and options utilizing generative AI, machine studying, information analytics, IoT & edge computing, and full-stack improvement to resolve real-world buyer challenges. In his private time, Laith enjoys the outside–fishing, images, drone flights, and mountaineering.

Paul Vincent is a Principal Prototyping Architect on the AWS Prototyping and Cloud Engineering (PACE) staff. He works with AWS clients to deliver their revolutionary concepts to life. Exterior of labor, he loves taking part in drums and piano, speaking with others by way of Ham radio, all issues residence automation, and film nights with the household.

Justin Lewis leads the Rising Expertise Accelerator at AWS. Justin and his staff assist clients construct with rising applied sciences like generative AI by offering open supply software program examples to encourage their very own innovation. He lives within the San Francisco Bay Space along with his spouse and son.

Anita Lewis is a Technical Program Supervisor on the AWS Rising Expertise Accelerator staff, primarily based in Denver, CO. She focuses on serving to clients speed up their innovation journey with generative AI and rising applied sciences. Exterior of labor, she enjoys aggressive pickleball matches, perfecting her golf sport, and discovering new journey locations.