home / mcp / aws s3 mcp server
Provides MCP endpoints to manage AWS S3 buckets, objects, and policies from a single server.
Configuration
View docs{
"mcpServers": {
"gangadharrr-aws-s3-mcp": {
"command": "npx",
"args": [
"aws-s3"
],
"env": {
"AWS_REGION": "us-east-1",
"AWS_ACCESS_KEY_ID": "your-access-key-id",
"AWS_SECRET_ACCESS_KEY": "your-secret-access-key"
}
}
}
}This MCP server provides a unified interface to manage AWS S3 resources from a single, programmatic point. You can list and manage buckets, upload and download objects, and get or set bucket policies through clear, explicit endpoints that integrate with your existing MCP workflows.
You interact with the AWS S3 MCP Server through an MCP client by calling the available operations. Use it to list all buckets, create or delete buckets (bucket must be empty to delete), list objects within a bucket (optionally filtered by prefix), upload and download objects, and inspect or update bucket policies. Each operation is exposed as a dedicated tool you can invoke from your MCP client, with parameters tailored to the action you want to perform.
Prerequisites: you need Node.js and npm installed on your machine.
1. Create a configuration file for the MCP server with your AWS credentials and default region.
2. Save the configuration snippet below to a file named mcp_config.json.
3. Run your MCP orchestration tool that loads the MCP configuration. If you are starting the stdio server directly in a development environment, you can invoke the provided CLI as shown.
4. Ensure your AWS credentials are available in the environment or through the provided configuration so the MCP server can authenticate with AWS.
5. Use your MCP client to call the tools described in the available endpoints to manage buckets, objects, or policies.
{
"mcpServers": {
"aws-s3": {
"command": "npx",
"args": ["aws-s3"],
"env": {
"AWS_REGION": "us-east-1",
"AWS_ACCESS_KEY_ID": "your-access-key-id",
"AWS_SECRET_ACCESS_KEY": "your-secret-access-key"
},
"autoApprove": ["list_buckets", "list_objects", "get_bucket_policy"]
}
}
}The server uses the AWS SDK and relies on credentials provided in environment variables or shared credentials. Remember to provide your AWS credentials securely, and prefer environment variables or IAM roles when running in AWS environments.
Lists all S3 buckets in the AWS account.
Creates a new S3 bucket with a given name and optional region.
Deletes an empty S3 bucket specified by name.
Lists objects in a bucket with optional prefix and pagination controls.
Uploads a file or content to a bucket at a specified key.
Downloads an object from a bucket, with optional local output or inline content.
Deletes an object from a bucket by key.
Retrieves the policy attached to a bucket.
Sets or updates the policy for a bucket.