agent/agent-framework
1
0
Fork 0
mirror of https://github.com/microsoft/agent-framework.git synced 2026-06-16 21:04:09 +08:00
Code Issues Packages Projects Releases Wiki Activity

Python: Update hosted agent samples with agent manifests (#2240)

Browse Source 
* Add agent manifests

* Correct agent manifest

* Correct agent manifest 2

* use resource substitution

* address comments
This commit is contained in:
Tao Chen
2025-11-14 17:10:57 -08:00
committed by GitHub
Unverified
parent c7e7020c32
commit 580a0c431a
12 changed files with 105 additions and 168 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/samples/demos/hosted_agents/agent_with_hosted_mcp/.env.example
-2
View File
@@ -1,2 +0,0 @@
OPENAI_CHAT_MODEL_ID="gpt-4o-2024-08-06"
OPENAI_API_KEY="your-openai-api-key"
python/samples/demos/hosted_agents/agent_with_hosted_mcp/README.md
-47
View File
@@ -1,47 +0,0 @@
# Hosted Agents with Hosted MCP Demo
This demo showcases an agent that has access to a MCP tool that can talk to the Microsoft Learn documentation platform, hosted as an agent endpoint running locally in a Docker container.
## What the Project Does
This project demonstrates how to:
- Create an agent with a hosted MCP tool using the Agent Framework
- Host the agent as an agent endpoint running in a Docker container
## Prerequisites
- OpenAI API access and credentials
- Required environment variables (see Configuration section)
## Configuration
Follow the `.env.example` file to set up the necessary environment variables for OpenAI.
## Docker Deployment
Build and run using Docker:
```bash
# Build the Docker image
docker build -t hosted-agent-mcp .
# Run the container
docker run -p 8088:8088 hosted-agent-mcp
```
> If you update the environment variables in the `.env` file or change the code or the dockerfile, make sure to rebuild the Docker image to apply the changes.
## Testing the Agent
Once the agent is running, you can test it by sending queries that contain the trigger keywords. For example:
```bash
curl -sS -H "Content-Type: application/json" -X POST http://localhost:8088/responses -d '{"input": "How to create an Azure storage account using az cli?","stream":false}'
```
Expected response:
```bash
{"object":"response","metadata":{},"agent":null,"conversation":{"id":"conv_6Y7osWAQ1ASyUZ7Ze0LL6dgPubmQv52jHb7G9QDqpV5yakc3ay"},"type":"message","role":"assistant","temperature":1.0,"top_p":1.0,"user":"","id":"resp_Vfd6mdmnmTZ2RNirwfldfqldWLhaxD6fO2UkXsVUg1jYJgftL9","created_at":1763075575,"output":[{"id":"msg_6Y7osWAQ1ASyUZ7Ze0PwiK2V4Bb7NOPaaEpQoBvFRZ5h6OfW4u","type":"message","status":"completed","role":"assistant","content":[{"type":"output_text","text":"To create an Azure Storage account using the Azure CLI, you'll need to follow these steps:\n\n1. **Install Azure CLI**: Make sure the Azure CLI is installed on your machine. You can download it from [here](https://docs.microsoft.com/en-us/cli/azure/install-azure-cli).\n\n2. **Log in to Azure**: Open your terminal or command prompt and use the following command to log in to your Azure account:\n\n ```bash\n az login\n ```\n\n This command will open a web browser where you can log in with your Azure account credentials. If you're using a service principal, you would use `az login --service-principal ...` with the appropriate parameters.\n\n3. **Select the Subscription**: If you have multiple Azure subscriptions, set the default subscription that you want to use:\n\n ```bash\n az account set --subscription \"Your Subscription Name\"\n ```\n\n4. **Create a Resource Group**: If you don’t already have a resource group, create one using:\n\n ```bash\n az group create --name myResourceGroup --location eastus\n ```\n\n Replace `myResourceGroup` and `eastus` with your desired resource group name and location.\n\n5. **Create the Storage Account**: Use the following command to create the storage account:\n\n ```bash\n az storage account create --name mystorageaccount --resource-group myResourceGroup --location eastus --sku Standard_LRS\n ```\n\n Replace `mystorageaccount` with a unique name for your storage account. The storage account name must be between 3 and 24 characters in length, and may contain numbers and lowercase letters only. You can also choose other `--sku` options like `Standard_GRS`, `Standard_RAGRS`, `Standard_ZRS`, `Premium_LRS`, based on your redundancy and performance needs.\n\nBy following these steps, you'll create a new Azure Storage account in the specified resource group and location with the specified SKU.","annotations":[],"logprobs":[]}]}],"parallel_tool_calls":true,"status":"completed"}
```
python/samples/demos/hosted_agents/agent_with_hosted_mcp/agent.yaml
+30
View File
@@ -0,0 +1,30 @@
# Unique identifier/name for this agent
name: agent-with-hosted-mcp
# Brief description of what this agent does
description: >
An AI agent that uses Azure OpenAI with a Hosted Model Context Protocol (MCP) server.
The agent answers questions by searching Microsoft Learn documentation using MCP tools.
metadata:
# Categorization tags for organizing and discovering agents
authors:
- Microsoft Agent Framework Team
tags:
- Azure AI AgentServer
- Microsoft Agent Framework
- Model Context Protocol
- MCP
template:
name: agent-with-hosted-mcp
# The type of agent - "hosted" for HOBO, "container" for COBO
kind: hosted
protocols:
- protocol: responses
environment_variables:
- name: AZURE_OPENAI_ENDPOINT
value: ${AZURE_OPENAI_ENDPOINT}
- name: AZURE_OPENAI_CHAT_DEPLOYMENT_NAME
value: "{{chat}}"
resources:
- kind: model
id: gpt-4o-mini
name: chat
python/samples/demos/hosted_agents/agent_with_hosted_mcp/main.py
+4 -4
View File
@@ -1,14 +1,14 @@
# Copyright (c) Microsoft. All rights reserved.
from agent_framework import HostedMCPTool
from agent_framework.openai import OpenAIChatClient
from agent_framework.azure import AzureOpenAIChatClient
from azure.ai.agentserver.agentframework import from_agent_framework # pyright: ignore[reportUnknownVariableType]
from azure.identity import DefaultAzureCredential
def main():
# Create an Agent using the OpenAI Chat Client with a MCP Tool that connects to Microsoft Learn MCP
agent = OpenAIChatClient().create_agent(
# Create an Agent using the Azure OpenAI Chat Client with a MCP Tool that connects to Microsoft Learn MCP
agent = AzureOpenAIChatClient(credential=DefaultAzureCredential()).create_agent(
name="DocsAgent",
instructions="You are a helpful assistant that can help with microsoft documentation questions.",
tools=HostedMCPTool(
python/samples/demos/hosted_agents/agent_with_text_search_rag/.env.example
-2
View File
@@ -1,2 +0,0 @@
OPENAI_CHAT_MODEL_ID="gpt-4o-2024-08-06"
OPENAI_API_KEY="your-openai-api-key"
python/samples/demos/hosted_agents/agent_with_text_search_rag/README.md
-54
View File
@@ -1,54 +0,0 @@
# Hosted Agents with Text Search RAG Demo
This demo showcases an agent that uses Retrieval-Augmented Generation (RAG) with text search capabilities that will be hosted as an agent endpoint running locally in a Docker container.
## What the Project Does
This project demonstrates how to:
- Build a customer support agent using the Agent Framework
- Implement a custom `TextSearchContextProvider` that simulates document retrieval
- Host the agent as an agent endpoint running in a Docker container
The agent responds to customer inquiries about:
- **Return & Refund Policies** - Triggered by keywords: "return", "refund"
- **Shipping Information** - Triggered by keyword: "shipping"
- **Product Care Instructions** - Triggered by keywords: "tent", "fabric"
## Prerequisites
- OpenAI API access and credentials
- Required environment variables (see Configuration section)
## Configuration
Follow the `.env.example` file to set up the necessary environment variables for OpenAI.
## Docker Deployment
Build and run using Docker:
```bash
# Build the Docker image
docker build -t hosted-agent-rag .
# Run the container
docker run -p 8088:8088 hosted-agent-rag
```
> If you update the environment variables in the `.env` file or change the code or the dockerfile, make sure to rebuild the Docker image to apply the changes.
## Testing the Agent
Once the agent is running, you can test it by sending queries that contain the trigger keywords. For example:
```bash
curl -sS -H "Content-Type: application/json" -X POST http://localhost:8088/responses -d '{"input": "What is the return policy","stream":false}'
```
Expected response:
```bash
{"object":"response","metadata":{},"agent":null,"conversation":{"id":"conv_2GbSxDpJJ89B6N4FQkKhrHaz78Hjtxy9b30JEPuY9YFjJM0uw3"},"type":"message","role":"assistant","temperature":1.0,"top_p":1.0,"user":"","id":"resp_Bvffxq0iIzlVkx2I8x7hV4fglm9RBPWfMCpNtEpDT6ciV2IG6z","created_at":1763071467,"output":[{"id":"msg_2GbSxDpJJ89B6N4FQknLsnxkwwFS2FULJqRV9jMey2BOXljqUz","type":"message","status":"completed","role":"assistant","content":[{"type":"output_text","text":"As of the most recent update, Contoso Outdoors' return policy allows customers to return products within 30 days of purchase for a full refund or exchange, provided the items are in their original condition and packaging. However, make sure to check your purchase receipt or the company's website for the most updated and specific details, as policies can vary by location and may change over time.","annotations":[],"logprobs":[]}]}],"parallel_tool_calls":true,"status":"completed"}
```
python/samples/demos/hosted_agents/agent_with_text_search_rag/agent.yaml
+33
View File
@@ -0,0 +1,33 @@
# Unique identifier/name for this agent
name: agent-with-text-search-rag
# Brief description of what this agent does
description: >
An AI agent that uses a ContextProvider for retrieval augmented generation (RAG) capabilities.
The agent runs searches against an external knowledge base before each model invocation and
injects the results into the model context. It can answer questions about Contoso Outdoors
policies and products, including return policies, refunds, shipping options, and product care
instructions such as tent maintenance.
metadata:
# Categorization tags for organizing and discovering agents
authors:
- Microsoft Agent Framework Team
tags:
- Azure AI AgentServer
- Microsoft Agent Framework
- Retrieval-Augmented Generation
- RAG
template:
name: agent-with-text-search-rag
# The type of agent - "hosted" for HOBO, "container" for COBO
kind: hosted
protocols:
- protocol: responses
environment_variables:
- name: AZURE_OPENAI_ENDPOINT
value: ${AZURE_OPENAI_ENDPOINT}
- name: AZURE_OPENAI_CHAT_DEPLOYMENT_NAME
value: "{{chat}}"
resources:
- kind: model
id: gpt-4o-mini
name: chat
python/samples/demos/hosted_agents/agent_with_text_search_rag/main.py
+4 -3
View File
@@ -7,8 +7,9 @@ from dataclasses import dataclass
from typing import Any
from agent_framework import ChatMessage, Context, ContextProvider, Role
from agent_framework.openai import OpenAIChatClient
from agent_framework.azure import AzureOpenAIChatClient
from azure.ai.agentserver.agentframework import from_agent_framework # pyright: ignore[reportUnknownVariableType]
from azure.identity import DefaultAzureCredential
if sys.version_info >= (3, 12):
from typing import override
@@ -91,8 +92,8 @@ class TextSearchContextProvider(ContextProvider):
def main():
# Create an Agent using the OpenAI Chat Client
agent = OpenAIChatClient().create_agent(
# Create an Agent using the Azure OpenAI Chat Client
agent = AzureOpenAIChatClient(credential=DefaultAzureCredential()).create_agent(
name="SupportSpecialist",
instructions=(
"You are a helpful support specialist for Contoso Outdoors. "
python/samples/demos/hosted_agents/agents_in_workflow/.env.example
-2
View File
@@ -1,2 +0,0 @@
OPENAI_CHAT_MODEL_ID="gpt-4o-2024-08-06"
OPENAI_API_KEY="your-openai-api-key"
python/samples/demos/hosted_agents/agents_in_workflow/README.md
-49
View File
@@ -1,49 +0,0 @@
# Hosted Workflow Agents Demo
This demo showcases an agent that is backed by a workflow of multiple agents running concurrently, hosted as an agent endpoint in a Docker container.
## What the Project Does
This project demonstrates how to:
- Build a workflow of agents using the Agent Framework
- Host the workflow agent as an agent endpoint running in a Docker container
The agent responds to product launch strategy inquiries by concurrently leveraging insights from three specialized agents:
- **Researcher Agent** - Provides market research insights
- **Marketer Agent** - Crafts marketing value propositions and messaging
- **Legal Agent** - Reviews for compliance and legal considerations
## Prerequisites
- OpenAI API access and credentials
- Required environment variables (see Configuration section)
## Configuration
Follow the `.env.example` file to set up the necessary environment variables for OpenAI.
## Docker Deployment
Build and run using Docker:
```bash
# Build the Docker image
docker build -t hosted-agent-workflow .
# Run the container
docker run -p 8088:8088 hosted-agent-workflow
```
> If you update the environment variables in the `.env` file or change the code or the dockerfile, make sure to rebuild the Docker image to apply the changes.
## Testing the Agent
Once the agent is running, you can test it by sending queries that contain the trigger keywords. For example:
```bash
curl -sS -H "Content-Type: application/json" -X POST http://localhost:8088/responses -d '{"input": "We are launching a new budget-friendly electric bike for urban commuters.","stream":false}'
```
> Expected response is not shown here for brevity. The response will include insights from the researcher, marketer, and legal agents based on the input prompt.
python/samples/demos/hosted_agents/agents_in_workflow/agent.yaml
+28
View File
@@ -0,0 +1,28 @@
# Unique identifier/name for this agent
name: agents-in-workflow
# Brief description of what this agent does
description: >
A workflow agent that responds to product launch strategy inquiries by concurrently leveraging insights from three specialized agents.
metadata:
# Categorization tags for organizing and discovering agents
authors:
- Microsoft Agent Framework Team
tags:
- Azure AI AgentServer
- Microsoft Agent Framework
- Workflows
template:
name: agents-in-workflow
# The type of agent - "hosted" for HOBO, "container" for COBO
kind: hosted
protocols:
- protocol: responses
environment_variables:
- name: AZURE_OPENAI_ENDPOINT
value: ${AZURE_OPENAI_ENDPOINT}
- name: AZURE_OPENAI_CHAT_DEPLOYMENT_NAME
value: "{{chat}}"
resources:
- kind: model
id: gpt-4o-mini
name: chat
python/samples/demos/hosted_agents/agents_in_workflow/main.py
+6 -5
View File
@@ -1,27 +1,28 @@
# Copyright (c) Microsoft. All rights reserved.
from agent_framework import ConcurrentBuilder
from agent_framework.openai import OpenAIChatClient
from azure.ai.agentserver.agentframework import from_agent_framework # pyright: ignore[reportUnknownVariableType]
from agent_framework.azure import AzureOpenAIChatClient
from azure.ai.agentserver.agentframework import from_agent_framework
from azure.identity import DefaultAzureCredential # pyright: ignore[reportUnknownVariableType]
def main():
# Create agents
researcher = OpenAIChatClient().create_agent(
researcher = AzureOpenAIChatClient(credential=DefaultAzureCredential()).create_agent(
instructions=(
"You're an expert market and product researcher. "
"Given a prompt, provide concise, factual insights, opportunities, and risks."
),
name="researcher",
)
marketer = OpenAIChatClient().create_agent(
marketer = AzureOpenAIChatClient(credential=DefaultAzureCredential()).create_agent(
instructions=(
"You're a creative marketing strategist. "
"Craft compelling value propositions and target messaging aligned to the prompt."
),
name="marketer",
)
legal = OpenAIChatClient().create_agent(
legal = AzureOpenAIChatClient(credential=DefaultAzureCredential()).create_agent(
instructions=(
"You're a cautious legal/compliance reviewer. "
"Highlight constraints, disclaimers, and policy concerns based on the prompt."