Ansible Collection for Amazon AI and ML Resources

The amazon.ai Ansible Collection provides automation modules for AWS AI and ML services. Currently, the collection includes modules for: - Amazon DevOps Guru: configure monitoring, manage insights, and integrate notification channels. - Amazon Bedrock: interact with foundation models for AI/ML applications.

The collection is designed to be extensible and will grow to support additional services such as Amazon SageMaker, Rekognition, Comprehend, Translate, and Textract.

As a Red Hat Ansible Certified Content, this collection is entitled to support through Ansible Automation Platform (AAP) through the Red Hat Ansible team.

Our mission

At the amazon.ai collection, our mission is to empower users to deploy and manage AWS AI/ML services using simple, declarative automation workflows with Ansible.

We aim to lower the barrier to entry for machine learning operations (MLOps) on AWS, enabling developers, data scientists, and DevOps engineers to build intelligent applications faster and more reliably.

Code of Conduct

We follow the Ansible Code of Conduct in all our interactions within this project.

If you encounter abusive behavior, please refer to the policy violations section of the Code for information on how to raise a complaint.

Communication

• Join the Ansible forum:
• Get Help: get help or help others. Please add appropriate tags if you start new discussions.
• Social Spaces: gather and interact with fellow enthusiasts.
• News & Announcements: track project-wide announcements including social events. The Bullhorn newsletter, which is used to announce releases and important changes, can also be found here.

For more information about communication, see the Ansible communication guide.

Ansible version compatibility

This collection has been tested against the following Ansible versions: >=2.17.0.

Plugins and modules within a collection may be tested with only specific Ansible versions. A collection may contain metadata that identifies these versions. PEP440 is the schema used to describe the versions of Ansible.

Python version compatibility

This collection depends on the AWS SDK for Python (Boto3 and Botocore). Due to the AWS SDK Python Support Policy this collection requires Python 3.9 or greater.

AWS SDK version compatibility

Version 1.0.0 of this collection supports boto3 >= 1.35.0 and botocore >= 1.35.0.

Included content

Modules

Using this collection

Installing the Collection from Ansible Galaxy

To consume this Collection from Automation Hub, please ensure that you add the following lines to your ansible.cfg file.

[galaxy]
server_list = automation_hub

[galaxy_server.automation_hub]
url=https://cloud.redhat.com/api/automation-hub/
auth_url=https://sso.redhat.com/auth/realms/redhat-external/protocol/openid-connect/token
token=<SecretToken>

The token can be obtained from the Automation Hub Web UI.

Once the above steps are done, you can run the following command to install the collection.

ansible-galaxy collection install amazon.ai

You can also include it in a requirements.yml file and install it with ansible-galaxy collection install -r requirements.yml, using the format:

---
collections:
  - name: amazon.ai

You can also install a specific version of the collection, for example, if you need to downgrade when something is broken in the latest version (please report an issue in this repository). Use the following syntax to install version 1.0.0:

ansible-galaxy collection install amazon.ai:==1.0.0

See using Ansible collections for more details.

The Python module dependencies are not installed by ansible-galaxy. They can be manually installed using pip:

pip install -r requirements.txt

or:

pip install boto3 botocore

Refer to the following resources for more details:

• Amazon Web Services Guide
• using Ansible collections for more details.

Using this collection

You can either call modules, rulebooks and playbooks by their Fully Qualified Collection Name (FQCN), such as amazon.ai.bedrock_agent, or you can call modules by their short name if you list the amazon.ai collection in the playbook's collections keyword:

---
- name: Create Bedrock Agent
  amazon.ai.bedrock_agent:
    state: present
    agent_name: "{{ agent_name }}"
    foundation_model: "{{ foundation_model }}"
    instruction: "{{ instruction }}"
    agent_resource_role_arn: "{{ agent_role_arn }}"

- name: Create Bedrock Agent Alias
  amazon.ai.bedrock_agent_alias:
    state: present
    agent_name: "{{ agent_name }}"
    alias_name: "{{ alias_name }}"

Testing

The project uses ansible-lint and black. Assuming this repository is checked out in the proper structure, e.g. collections_root/ansible_collections/amazon/ai/, run:

tox -e ansible-lint

or

tox -e black

Sanity and unit tests are run as normal:

ansible-test sanity

If you want to run cloud integration tests, ensure you log in to the cloud:

# using the "default" profile on AWS
aws configure set aws_access_key_id     my-access-key
aws configure set aws_secret_access_key my-secret-key
aws configure set region                eu-north-1

ansible-test integration [target]

This collection is tested using GitHub Actions.

Contributing to this collection

The content of this collection is made by people like you, a community of individuals collaborating on making the world better through developing automation software.

We are actively accepting new contributors and all types of contributions are very welcome.

Don't know how to start? Refer to the Ansible community guide!

More information

• Ansible user guide
• Ansible developer guide
• Ansible collections requirements
• Ansible community Code of Conduct
• The Bullhorn (the Ansible contributor newsletter)
• Important announcements for maintainers

Support

As Red Hat Ansible Certified Content, this collection is entitled to support through the Ansible Automation Platform (AAP) using the Create issue button on the top right corner. If a support case cannot be opened with Red Hat and the collection has been obtained either from Galaxy or GitHub, there may community help available on the Ansible Forum.

You can also join us on:

• Libera.Chat IRC - the #ansible-aws irc.libera.chat channel

Release notes

See the changelog.

Licensing

GNU General Public License v3.0 or later.

See LICENSE to see the full text.