AWS pricing checks are effortless with Pricing MCP Server

Cloud bills are hard to predict. Developers spin up instances, test workloads, and move fast, but often only discover the true cost after the invoice arrives. Pricing tools like the AWS Pricing Calculator or Cost Explorer can help, but require leaving the workflow and context-switching. The result is a lack of cost awareness at the very moment when architecture choices are being made.

The Pricing MCP Server resolves this issue by bringing real-time pricing awareness in AWS directly into the development process. It ensures cost visibility when decisions are made without breaking flow or context.

This newsletter will walk you through everything you need to know about the Pricing MCP Server, including:

• What the AWS Pricing MCP Server actually is

• Who can benefit from using it

• How you can set it up step-by-step in a local environment

By the end, you’ll understand the value of the Pricing MCP Server and have the skills to integrate it into your workflow. This means you can make architecture and design decisions with real-time AWS pricing insights at your fingertips.

What is the AWS Pricing MCP Server?#

The AWS Pricing MCP Server is an open-source service that plugs directly into AI assistants that support the Model Context Protocol (MCP)MCP is a protocol that allows LLMs to connect with external data sources and APIs in a consistent way.. For example, assistants like Amazon Q Developer CLI, Claude Desktop, Cursor, or Windsurf can all use MCP servers to fetch real-time data.

So, by using AWS Pricing MCP Server, instead of navigating through AWS consoles or calculators, you can simply ask your assistant, “What’s the hourly price of an m7g.large instance in us-east-1?” or “What would 200 GB of gp3 storage cost per month in ap-south-1?”

How does the MCP server work?#

After integrating the AWS Pricing MCP Server with your AI assistants, your request is routed to the Pricing MCP Server when you interact with AI assistants and ask a pricing question. The server then calls the AWS Price List API, which exposes API operations such as DescribeServices, GetAttributeValues, and GetProducts. These APIs return the official AWS prices for your specific services and regions.

The server formats the results and sends them back to your assistant, making pricing checks feel as natural as asking a colleague.

The results are always current and accurate because the server is tied directly to AWS APIs.

Who should use this tool?#

The Pricing MCP Server is useful for a wide range of users. Developers and reliability engineers can validate design choices without switching context. Solution architects can quickly compare Regions and pricing while planning architectures with clients. FinOps teams can use it for quick what-if scenarios and data exports. Even engineering managers can use it to understand the financial trade-offs of design decisions during roadmap planning.

AWS Pricing MCP Server setup#

The Pricing MCP Server can be connected to any MCP-compatible client, including assistants such as Amazon Q Developer CLI, Claude Desktop, Cursor, and Windsurf. Each tool can use the server to fetch real-time AWS pricing data.

This tutorial will focus on Amazon Q Developer CLI, an AWS command-line tool that integrates directly into your terminal. You can query AWS pricing information directly from your terminal by the tutorial’s end.

Let’s learn how to set up the AWS Pricing MCP Server with Amazon Q Developer CLI on your local machine step by step.

Step 1: Set up the Amazon Q Developer CLI#

The first step is to install the Amazon Q app.

• On macOS and Linux, you can install it with Homebrew using the following command:

• Once the app has been installed, launch it. Click the “Get started” button.

• You’ll be asked to enable the shell integrations. This will allow you to run Amazon Q from the shell.

• Click the “Install” button to integrate Amazon Q with your local shell.

• Next, you’ll be asked to enable accessibility for Amazon Q to add text on your behalf. You can simply skip it by clicking “Skip.”

• You'll next be asked to sign in to start using Amazon Q. Click the “Sign in” button.

• You’ll be redirected to AWS Skill Builder page. Authenticate yourself using your AWS Skill Builder ID to start using Amazon Q.

• Once that is done open a new terminal and verify that Q is working by running the command given below:

If the installation was successful, this command prints the version of the Q Developer CLI installed on your machine.

• If the installation was successful, use the following command in terminal to start chatting with Amazon Q.

• Now, ask Amazon Q the following question:

Amazon Q will provide you with an answer, but this answer is based on the data it was trained on and may not reflect the latest AWS pricing as pricing changes frequently. Now, we’ll integrate the AWS Pricing MCP Server to enable Amazon Q to access the latest AWS pricing.

Step 2: Install prerequisites#

The Pricing MCP Server runs as a local process. To start it easily, we’ll use uvx, a lightweight tool runner.

• Start with executing the following command to exit the Amazon Q CLI:

• Now, execute the command given below to install the uvx package.

• The AWS Pricing MCP Server is built in Python, meaning it runs as a Python application. To launch it, you need a Python runtime installed locally. Execute the command given below to install Python:

Step 3: Configure AWS credentials#

The server needs permission to call the AWS Pricing API. You can provide that through an AWS profile.

• To configure the AWS credentials, execute the following command:

You will be prompted to enter an access key ID, secret access key, default region, and output format. For region, you can safely enter us-east-1, as that is one of the endpoints where the Pricing API is available.

The IAM user or role whose credentials you configure must have at least these permissions:

This policy grants read-only access to the Price List API. You can attach it to your user, role, or group in IAM.

Step 4: Connect the server to Amazon Q Developer#

Amazon Q Developer CLI uses a configuration file called mcp.json to know which MCP servers to connect to. Amazon Q fetches these configurations automatically from ~/.aws/amazonq/mcp.json.

• Start by creating the ~/.aws/amazonq directory using this command:

• Next, create the mcp.json file using the command given below:

This command will create the file and open it for editing.

• Paste the configurations given below in your terminal and save the changes to the mcp.json file:

Here’s what this does:

• command tells Amazon Q Developer to launch the server using uvx.

• args specifies the MCP server package to install and run.

• env passes environment variables to the server. AWS_PROFILE selects the profile we configured earlier. AWS_REGION sets the API endpoint for pricing queries. FASTMCP_LOG_LEVEL reduces noise in the logs.

• disabled being false means the server is active.

Step 5: Verify the connection#

Before you start using Amazon Q integrated with Pricing MCP Server, verify that the integration was setup correctly.

• To do that, type the following command in the terminal:

You should see pricing tools listed from the aws-pricing server.

• If you don’t see the aws-pricing server listed as an MCP server, double-check the JSON syntax in your configuration file and make sure your credentials are valid by running:

If this returns your account information, your profile is set up correctly.

Step 6: Ask your pricing question#

Now you are ready to use the server.

• Open the Q Developer CLI, by running the following command:

• In the Q Developer CLI, ask the same question as earlier:

You’ll see something like this as the initial response:

This response indicates that Q forwards the request to the MCP server, which will fetch the data from the Pricing API and respond with the official price. You can continue by asking more complex questions, like comparing instance prices across regions or checking which regions support a particular storage class.

If you encounter slow responses, try setting the AWS_REGION environment variable in your configuration to the Pricing API endpoint closest to you (us-east-1, eu-central-1, or ap-south-1).

Integrating pricing into your workflows#

Once you have the Pricing MCP Server running, the real value comes from weaving it into the tools you already use every day. If you are working with Amazon Q, it means you can ask natural questions about AWS service costs directly from your coding environment and get precise answers without leaving your flow. In a CI/CD pipeline, the server can be queried to validate that new resources or architectures stay within budget before deployment. For developers building custom dashboards, the server can be integrated with your monitoring stack so that pricing data sits right alongside usage metrics, giving you a live picture of both consumption and cost.

This shift means you no longer need to toggle between the AWS Pricing Calculator, spreadsheets, and your IDE. Instead, pricing awareness becomes part of the development process itself, guiding design choices as naturally as a unit test or linting rule.

Wrapping up#

The AWS Pricing MCP Server shifts cost awareness from a late-stage concern to an early-stage design tool. By integrating directly with Amazon Q Developer CLI, it gives developers real-time pricing data without breaking their workflow. That means fewer surprises at the end of the month, smarter trade-offs during design, and more confident conversations about cost across teams. No more guesswork—just answers, when and where you need them.

Ready to dive deeper in the cloud? Explore the following CloudLabs projects:

• Building Multiple Agents Using CrewAI and Bedrock

• Automating Data Processing with AWS Glue DataBrew