Microsoft Agent Framework

Deploy Microsoft Agent Framework, an open source framework for building, orchestrating, and deploying AI agents and multi-agent applications to automate complex workflows.

1. Deploy Microsoft Agent Framework

Estimated deployment time: 5 minutes

Follow the deployment instructions from the Get started section to configure the app, deploy it, and verify software installation.

For information on the app-specific configurations, see the Configuration options section.

Base distribution and plan

Supported distribution: Ubuntu 24.04 LTS
Recommended plans: All plan types and sizes can be used.

Configuration options

Configure the required options to deploy your instance. For additional customization, add advanced options.

The table maps the Cloud Manager UI fields to their corresponding API/CLI keys (stackscript_data) required for automated deployments.

StackScript ID: 2165612

UI field	API/CLI key	Description
Required options
Limited sudo user	`user_name`	Your preferred username for the limited sudo user, entered without any capital letters, spaces, or special characters. When adding a limited sudo user, the user is created with a strong generated password for your new Linode instance, and the account is assigned to the `sudo` group, which provides elevated permissions when running commands with the `sudo` prefix. Note: For easier and more secure access with the sudo user, add an account SSH key for the Cloud Manager user during deployment and select that user as an `authorized_user`. Their SSH pubkey will be assigned to both the root and limited sudo users.
Advanced options
Disable root access over SSH	`disable_root`	Applies to a limited sudo user. To block the root user from logging in over SSH, select Yes. Defaults to No. Note: When you disable the root user from logging in over SSH and don't provide a valid Account SSH Key assigned to the `authorized_user`, you can still switch to the root user. To do that, log in as root via the Lish console and run `cat /home/$USERNAME/.credentials` to view the generated password for the limited sudo user.

Note: After the app deployment completes, the password for your limited sudo user is generated and stored in the .credentials file in the home directory, along with application-specific passwords. Log in to your instance as root through the Lish console or SSH, then run cat /home/$USERNAME/.credentials to view its contents.

Use API, CLI, or Terraform

In addition to deploying the app to a new Linode instance via Cloud Manager, you can also use the Linode API, CLI, or Terraform. When running the operation, you need to provide the StackScript ID, supported Linux distribution, and app-specific fields along with the standard Linode deployment configurations.

Note: Generate a personal access token to authenticate your API, CLI, or Terraform requests.

curl --location 'https://api.linode.com/v4/linode/instances' \
    --header 'Content-Type: application/json' \
    --header 'Accept: application/json' \
    --header 'Authorization: Bearer abc123def456hij789klm' \
    --data-raw '{
        "region": "us-east",
        "type": "g6-standard-2",
        "image": "linode/ubuntu24.04",
        "label": "my-microsoft-agent-framework-one-click-app",
        "root_pass": "@C0mpl3x#P@ssw0rd",
        "stackscript_id": 2165612,
        "stackscript_data": {
            "user_name":"jsmith"
        }
    }'

linode-cli linodes create \
    --region us-east \
    --type g6-standard-2 \
    --label my-microsoft-agent-framework-one-click-app \
    --image linode/ubuntu24.04 \
    --root_pass @C0mpl3x#P@ssw0rd \
    --stackscript_id 2165612 \
    --stackscript_data '{"user_name":"jsmith"}'

resource "linode_instance" "my-linode" {
    region         = "us-east"
    type           = "g6-standard-2"
    label          = "my-microsoft-agent-framework-one-click-app"
    image          = "linode/ubuntu24.04"
    root_pass      = "@C0mpl3x#P@ssw0rd"
    stackscript_id = 2165612
    stackscript_data = {
        "user_name" = "jsmith"
    }
}

2. Test the Python SDK

Once the deployment is complete, the agent-framework library gets installed on your instance. This allows you to import the library into your software. To get started:

Create an example directory named science.
```
mkdir science
```
Create a test Python file, agent.py. It allows you to use our AI model.
```
cd science
vim agent.py
```

Enter this code into the agent.py Python file.

import asyncio
from openai import AsyncOpenAI
from agent_framework import Agent
from agent_framework.openai import OpenAIChatClient

async def main():
    client = OpenAIChatClient(
        model="Qwen/Qwen3-14B-AWQ",
        base_url="http://localhost:8000/v1",
        api_key="dummy",
    )

    agent = client.as_agent(
        name="Assistant",
        instructions="You are a helpful assistant.",
    )

    result = await agent.run("Why is the sky blue?")

    print(result)

if __name__ == "__main__":
    asyncio.run(main())

Once you save the file, execute it with the python3 agent.py command.

Additional resources

This example uses a self-hosted model exposed via the LLM’s API. To use a provider model, refer to the Microsoft Agent Framework documentation.

📘
Note that we can't vouch for the accuracy or timeliness of externally hosted resources.