From f55625abaa1e79eb96ba11e83adb26dbbcfa6c57 Mon Sep 17 00:00:00 2001 From: n0vabyte Date: Tue, 21 Jul 2026 13:26:18 -0400 Subject: [PATCH] Haystack and Microsoft Agent Framework docs to QDA --- .../marketplace-docs/guides/haystack/index.md | 91 ++++++++++++++++++ .../guides/microsoft-agent-framework/index.md | 95 +++++++++++++++++++ 2 files changed, 186 insertions(+) create mode 100644 docs/marketplace-docs/guides/haystack/index.md create mode 100644 docs/marketplace-docs/guides/microsoft-agent-framework/index.md diff --git a/docs/marketplace-docs/guides/haystack/index.md b/docs/marketplace-docs/guides/haystack/index.md new file mode 100644 index 00000000000..4f174ef1ff4 --- /dev/null +++ b/docs/marketplace-docs/guides/haystack/index.md @@ -0,0 +1,91 @@ +--- +title: "Deploy Haystack" +description: "This tutorial will show you how to deploy Haystack as a Quick Deploy App." +published: 2026-07-21 +keywords: ['AI Framework', 'AI'] +tags: ["quick deploy apps", "linode platform", "cloud manager"] +external_resources: +- '[Haystack Documentation](https://docs.haystack.deepset.ai/docs/get-started)' +aliases: ['/products/tools/marketplace/guides/haystack/', '/guides/haystack/'] +authors: ["Akamai"] +contributors: ["Akamai"] +license: '[CC BY-ND 4.0](https://creativecommons.org/licenses/by-nd/4.0)' +marketplace_app_id: 2165521 +marketplace_app_name: "Haystack" +--- + +Haystack is an open-source framework for building applications that use large language models (LLMs). Instead of requiring developers to write custom orchestration logic for every use case, it provides a structured way to assemble AI functionality through reusable components connected in pipelines. This approach makes it well suited for creating retrieval-augmented generation (RAG) systems, AI agents, search experiences, chat applications, and other LLM-driven solutions. + +## Deploying a Quick Deploy App + +{{% content "deploy-marketplace-apps-shortguide" %}} + +{{% content "marketplace-verify-standard-shortguide" %}} + +{{< note >}} +**Estimated deployment time:** Haystack should be fully installed within 5 minutes after the Compute Instance has finished provisioning. +{{< /note >}} + +## Configuration Options + +- **Supported distributions:** Ubuntu 24.04 LTS +- **Recommended plan:** All plan types and sizes can be used. + +## Haystack Options + +{{% content "marketplace-required-limited-user-fields-shortguide" %}} + +{{% content "marketplace-special-character-limitations-shortguide" %}} + +## Getting Started after Deployment + +### Testing Agent + +Once the deployment is complete, the `haystack-ai` library should already be installed on your instance. This will allow you to import the library into your software. + +1. To get started, create an example directory called `science`. + + ```command + mkdir science + ``` + +2. To get started, create a test Python file called `agent.py` that will allow us to use our AI model. + +``` +vim agent.py +``` + +3. Enter the following content into the `agent.py` Python file. + +```python +from haystack.components.agents import Agent +from haystack.components.generators.chat import OpenAIChatGenerator +from haystack.dataclasses import ChatMessage +from haystack.tools import ComponentTool +from haystack.utils import Secret + +agent = Agent( + chat_generator=OpenAIChatGenerator( + api_base_url="http://localhost:8000/v1", + api_key=Secret.from_token("EMPTY"), + model="Qwen/Qwen3-14B-AWQ", + ), + system_prompt="You are a helpful assistant that can search the web for information.", +) + +result = agent.run( + messages=[ChatMessage.from_user("What is Haystack AI?")] +) + +print(result["last_message"].text) +``` + +4. Once you save the file you can execute it with the following command. + +```command +python3 sky.py +``` + +This example uses a self-hosted model exposed via vLLM's API. If you want to use a provider model you can refer to Haystack documentation. + +{{% content "marketplace-update-note-shortguide" %}} diff --git a/docs/marketplace-docs/guides/microsoft-agent-framework/index.md b/docs/marketplace-docs/guides/microsoft-agent-framework/index.md new file mode 100644 index 00000000000..09b0168baa2 --- /dev/null +++ b/docs/marketplace-docs/guides/microsoft-agent-framework/index.md @@ -0,0 +1,95 @@ +--- +title: "Deploy Microsoft Agent Framework" +description: "This tutorial will show you how to deploy Microsoft Agent Framework as a Quick Deploy App." +published: 2026-07-21 +keywords: ['AI Framework', 'AI'] +tags: ["quick deploy apps", "linode platform", "cloud manager"] +external_resources: +- '[Microsoft Agent Framework Documentation](https://learn.microsoft.com/en-us/agent-framework/)' +aliases: ['/products/tools/marketplace/guides/microsoft-agent-framework/', '/guides/microsoft-agent-framework/'] +authors: ["Akamai"] +contributors: ["Akamai"] +license: '[CC BY-ND 4.0](https://creativecommons.org/licenses/by-nd/4.0)' +marketplace_app_id: 2165612 +marketplace_app_name: "Microsoft Agent Framework" +--- + +Microsoft Agent Framework is an open-source framework for building, orchestrating, and deploying AI agents and multi-agent applications. It provides developers with the tools to create intelligent agents that can reason, use tools, collaborate, and automate complex workflows. + +## Deploying a Quick Deploy App + +{{% content "deploy-marketplace-apps-shortguide" %}} + +{{% content "marketplace-verify-standard-shortguide" %}} + +{{< note >}} +**Estimated deployment time:** Microsoft Agent Framework should be fully installed within 5 minutes after the Compute Instance has finished provisioning. +{{< /note >}} + +## Configuration Options + +- **Supported distributions:** Ubuntu 24.04 LTS +- **Recommended plan:** All plan types and sizes can be used. + +## CrewAI Options + +{{% content "marketplace-required-limited-user-fields-shortguide" %}} + +{{% content "marketplace-special-character-limitations-shortguide" %}} + +## Getting Started after Deployment + +### Testing Python's SDK + +Once the deployment is complete, the `agent-framework` library should already be installed on your instance. This will allow you to import the library into your software. + +1. To get started, create an example directory called `science`. + + ```command + mkdir science + ``` + +2. Create a test Python file called `sky.py` that will allow us to use our AI model. + +``` +cd science +vim sky.py +``` + +3. Enter the following content into the `sky.py` Python file. + +```python +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()) +``` + +4. Once you save the file you can execute it with the following command. + +```command +python3 sky.py +``` + +This example uses a self-hosted model exposed via vLLM's API. If you want to use a provider model you can refer to Microsoft Agent Framework documentation. + +{{% content "marketplace-update-note-shortguide" %}}