AWS Marketplace
Providers Guide

AMI Product Checklist

Before submitting your AMI product request to AWS Marketplace, review this checklist. Validating this information will help to make sure your submission goes through the publication process smoothly.

Product usage:

  • Your AMI must be production-ready.

  • Your AMI cannot restrict product usage by time or any other measurements.

  • Your AMI must be compatible with the 1-Click fulfillment experience.

  • Everything required to use the product is in the software, including client applications.

  • The default user uses a randomized password, or creating the initial user requires verification that the subscriber is authorized to use the instance using a value unique to the instance such as instance ID.

For free or paid products:

  • No additional license is required to use the product.

  • The subscriber doesn't have to provide personally identifiable information (e.g., email address) to use the product.

AMI preparation:

  • Uses HVM virtualization and 64-bit architecture

  • Doesn't contain any known vulnerabilities, malware or viruses

  • Subscribers have OS-level administration access to the AMI

  • Run your AMI through AMI Self-Service Scanning

For Windows AMIs:

  • Uses the most recent version of EC2ConfigService

  • Ec2SetPassword, Ec2WindowsActivate, and Ec2HandleUserData are enabled in your AMI

  • No guest accounts or remote desktop users are present

For Linux AMIs:

  • Root login is locked or disabled

  • No authorized keys, default passwords, or other credentials are included

Product Load Form or Product tab

  • All required fields are completed

  • All values are within specified character limits

  • All URLs load without error

  • The product image is at least 110 pixels wide and between a 1:1 and 2:1 ratio

  • Pricing is specified for all enabled instance types (for hourly, hourly-based monthly pricing, and hourly-based annual pricing models)

  • Monthly pricing is specified (for hourly-based monthly and monthly pricing models)