AWS Data Processing MCP Server

0.1.28 · active · verified Thu Apr 16

The AWS Data Processing MCP server (Model Context Protocol) provides AI code assistants with comprehensive data processing tools and real-time pipeline visibility across AWS Glue, Amazon EMR-EC2, and Athena. It enables AI code assistants to streamline data processing workflows through intelligent guidance, from initial data discovery and cataloging through complex ETL pipeline orchestration and big data analytics optimization. The current version is 0.1.28, and it is actively maintained as part of the AWS Labs MCP server initiative.

Common errors

```
No solution found when resolving tool dependencies: ... Because aws-aws-dataprocessing-mcp-server was not found in the package registry...
```
cause This error typically occurs due to a typo in the package name during client configuration. The correct PyPI package name uses `awslabs.` prefix.

fix Correct the server package name in your MCP client configuration (e.g., `mcp.json`) from `aws-aws-dataprocessing-mcp-server` to `awslabs.aws-dataprocessing-mcp-server@latest`.
```
Not authorized to perform an action in AWS MCP Server (Preview) or Not authorized to perform iam:PassRole
```
cause The IAM role or user running the MCP server lacks the necessary permissions to perform the requested AWS action or to pass a role to an AWS service.

fix Review and update your IAM policies to grant the required permissions to the entity running the MCP server. Ensure that `iam:PassRole` is allowed if the server needs to assume other roles. Consult AWS IAM documentation for specific service permissions.
```
MCP server operations are failing for write actions (e.g., create, update, delete) but read actions work.
```
cause By default, the AWS Data Processing MCP Server runs in a read-only mode for security reasons.

fix To enable write operations, you must explicitly include the `--allow-write` argument in the MCP server's command definition within your client configuration.

Warnings

breaking On May 26th, 2025, Server Sent Events (SSE) support was removed from all MCP servers in their latest major versions to align with the Model Context Protocol specification's backwards compatibility guidelines.
Fix: Clients should prepare for Streamable HTTP support. Check client documentation for updates on streaming capabilities.
gotcha Enabling `--allow-write` or `--allow-sensitive-data-access` flags grants significant privileges to the MCP server. Exercise extreme caution and only enable these when necessary and in trusted environments.
Fix: Always review the IAM permissions granted to the role or user running the MCP server and restrict them to the minimum necessary. Only use write and sensitive data access in controlled setups.
gotcha The DataProcessing MCP Server can only update or delete resources that were originally created through it. Resources created by other means cannot be modified or deleted using this server.
Fix: Plan your resource management strategy accordingly, understanding that resources created outside of the MCP server's scope will require manual intervention or alternative tools for modification/deletion.
gotcha The Model Context Protocol (MCP) implementations, including this server, are still evolving and an AWS Builder Center article from July 2025 indicates they are not yet considered production-ready due to potential security, data sovereignty, authentication, and operational monitoring gaps.
Fix: Organizations should carefully evaluate these challenges and consider MCP for production deployments with caution. Prioritize security reviews, robust monitoring, and compliance frameworks.

Install

pip install awslabs-aws-dataprocessing-mcp-server PyPI installation
uvx awslabs.aws-dataprocessing-mcp-server@latest UVX installation (recommended for MCP clients)

Quickstart

The AWS Data Processing MCP Server is typically run by an MCP client (e.g., Amazon Q Developer CLI, Kiro, Cursor) rather than being directly imported into Python code. To get started, configure your MCP client (e.g., by editing `~/.aws-q-developer-cli/mcp/mcp.json` for Amazon Q Developer CLI or `~/.kiro/settings/mcp.json` for Kiro) with the server definition. This JSON snippet provides an example configuration for launching the server, enabling write operations, and specifying the AWS region and profile. Replace `us-east-1` and `my-mcp-profile` with your desired values. Ensure your AWS credentials are configured for the specified profile.

{
  "mcpServers": {
    "aws.dp-mcp": {
      "autoApprove": [],
      "disabled": false,
      "command": "uvx",
      "args": [
        "awslabs.aws-dataprocessing-mcp-server@latest",
        "--allow-write" 
      ],
      "env": {
        "AWS_REGION": "us-east-1",
        "AWS_PROFILE": "my-mcp-profile" 
      }
    }
  }
}

view raw JSON →