### Manual Setup for Local Agent

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations/github-copilot/README.md

Install dependencies, copy the .env example, set your GitHub token, and run the main Python script. The agent will start on http://localhost:8088/.

```bash
pip install -r requirements.txt
cp .env.example .env  # skip if .env already exists
# Edit .env and set GITHUB_TOKEN=github_pat_...
python main.py
```

--------------------------------

### Install Dependencies and Start

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/javascript/foundry-local/audio-transcription-example/README.md

Install project dependencies and start the application using npm.

```bash
npm install
npm start
```

--------------------------------

### Manual Setup and Run Agent

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/notetaking-agent/README.md

Provides steps for manual setup, including copying the environment file, installing dependencies, and starting the agent script.

```bash
# Copy and edit environment file
cp .env.example .env  # skip if .env already exists

# Install dependencies
pip install -r requirements.txt

# Start the agent
python main.py
```

--------------------------------

### Manual Setup for Local Development

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations/toolbox/README.md

This sequence sets up the environment, installs dependencies, and runs the agent manually. Ensure you have Python and pip installed. The agent will start on `http://localhost:8088`.

```bash
cp .env.example .env  # skip if .env already exists
# Edit .env and fill in your values, then:
export $(grep -v '^#' .env | xargs)

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python main.py
```

--------------------------------

### Full Deployment Example

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/livekit-server/README.md

A comprehensive example combining initialization, ACR configuration, environment variable setup, and deployment. Replace all placeholders with your specific Azure and LiveKit details.

```bash
az account set --subscription <subscription-id>

mkdir -p ~/azd-deploys/livekit-server && cd ~/azd-deploys/livekit-server

azd ai agent init \
  -m <path-to-repo>/samples/python/hosted-agents/bring-your-own/invocations_ws/livekit-server/agent.manifest.yaml \
  -p "/subscriptions/<subscription-id>/resourceGroups/<resource-group>/providers/Microsoft.CognitiveServices/accounts/<foundry-account>/projects/<foundry-project>" \
  --no-prompt

azd env set AZURE_CONTAINER_REGISTRY_NAME     <acr-name>
azd env set AZURE_CONTAINER_REGISTRY_ENDPOINT <acr-name>.azurecr.io
azd env set USE_EXISTING_CONTAINER_REGISTRY   true
azd env set CONTAINER_REGISTRY_RESOURCE_GROUP <acr-resource-group>

# ...azd env set the AZURE_/LIVEKIT_ variables above...

azd deploy livekit-server-azd
```

--------------------------------

### Manual Setup and Run

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations/langgraph-chat/README.md

Sets up the environment by copying the example .env file and then runs the Python script for the agent. Edit .env values as needed.

```bash
cp .env.example .env  # then edit values (skip if .env already exists)
pip install -r requirements.txt
python main.py
```

--------------------------------

### Local Quick Start - Linux/macOS

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/langgraph-toolbox/README.md

Installs dependencies and starts the agent locally on Linux/macOS. Requires environment variables to be set in `.env`.

```bash
# 1. Copy and fill in the environment file
cp .env.example .env  # skip if .env already exists
# Edit .env — set FOUNDRY_PROJECT_ENDPOINT, AZURE_AI_MODEL_DEPLOYMENT_NAME,
#              and TOOLBOX_ENDPOINT at minimum

# 2. Install dependencies
pip install -r requirements.txt

# 3. Start the agent
python main.py

# 4. Invoke
curl -X POST http://localhost:8088/responses \
  -H "Content-Type: application/json" \
  -d '{"input": "What tools do you have?"}'
```

--------------------------------

### Install Dependencies and Run Agent Host

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/csharp/hosted-agents/agent-framework/browser-automation/README.md

Installs necessary dependencies and starts the local hosted-agent server. Ensure you are in the correct project directory.

```bash
dotnet restore browser-automation.csproj
npm install -g @playwright/cli@latest
playwright-cli install --skills
dotnet run --project browser-automation.csproj
```

--------------------------------

### Setup Virtual Environment and Install Dependencies

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/foundry-local/verify-winml/README.md

Installs the WinML SDK variant and its dependencies using pip. Ensure you are in a fresh virtual environment or have cleaned up previous installations.

```bash
python -m venv .venv
.venv\Scripts\Activate.ps1
pip install --upgrade -r requirements.txt
```

```bash
pip install --upgrade -r requirements.txt
```

--------------------------------

### Manual Setup for Hosted Agent

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/env-vars-agent/README.md

Manually set up the hosted agent by installing dependencies, copying the environment file, and exporting necessary environment variables. Then, run the main Python script. The agent will start on `http://localhost:8088/`.

```bash
pip install -r requirements.txt
cp .env.example .env  # then edit values — fill in any test values you like (skip if .env already exists)
export FOUNDRY_PROJECT_ENDPOINT="https://your-project.services.ai.azure.com/api/projects/your-project"
export AZURE_AI_MODEL_DEPLOYMENT_NAME="gpt-5.4-mini"
export SECRET_API_KEY="ab12-fake-test-key"
export TARGET="https://api.example.com"
export SECRET_KEY="p@ssw0rd-test-value"
export NON_SECRET_KEY="westus2"
python main.py
```

--------------------------------

### Manual Setup for Hosted Agent

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/openai-agents-sdk/README.md

Install dependencies, copy the environment file, and set necessary environment variables before running the agent script. Ensure `FOUNDRY_PROJECT_ENDPOINT` and `AZURE_AI_MODEL_DEPLOYMENT_NAME` are configured in the `.env` file. The agent will start on `http://localhost:8088/`.

```bash
pip install -r requirements.txt
cp .env.example .env  # then set FOUNDRY_PROJECT_ENDPOINT and AZURE_AI_MODEL_DEPLOYMENT_NAME (skip if .env already exists)
python main.py
```

--------------------------------

### Start Shared Web Portal

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/pipecat-ws-server/README.md

Navigate to the client directory, set up a virtual environment, install dependencies, configure the local WebSocket URL in .env, and run the web portal script.

```bash
cd samples/python/hosted-agents/bring-your-own/invocations_ws/pipecat-ws-server/chat_client

python -m venv ../.venv          # (or reuse the server's venv)
source ../.venv/bin/activate
pip install -r requirements.txt

# Local mode — point at the server you just started
echo 'PIPECAT_WEBSOCKET_LOCAL_URL=ws://localhost:8088/invocations_ws' >> .env

python web_portal.py             # listens on :9527
```

--------------------------------

### Manual Setup for Hosted Agent

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations/ag-ui/README.md

Install dependencies, copy the environment file, and run the main script to set up the agent manually. The agent listens on `http://localhost:8088/`.

```bash
pip install -r requirements.txt
cp .env.example .env  # then fill in values (skip if .env already exists)
python main.py
```

--------------------------------

### Start Chat Client Portal

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/livekit-server/README.md

Installs dependencies and starts the chat client portal. It includes an option to configure it for a local LiveKit server instead of a hosted Foundry agent.

```bash
cd chat_client
pip install -r requirements.txt
cp .env.example .env
# To use this LOCAL livekit-server instead of the hosted Foundry agent,
uncomment in chat_client/.env:
#   LIVEKIT_LOCAL_URL=ws://localhost:8088/invocations_ws
python web_portal.py
```

--------------------------------

### Local Quick Start - Windows PowerShell

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/langgraph-toolbox/README.md

Installs dependencies and starts the agent locally on Windows using PowerShell. Requires environment variables to be set in `.env`.

```powershell
# 1. Copy and fill in the environment file
Copy-Item .env.example .env  # skip if .env already exists
# Edit .env — set FOUNDRY_PROJECT_ENDPOINT, AZURE_AI_MODEL_DEPLOYMENT_NAME,
#              and TOOLBOX_ENDPOINT at minimum

# 2. Install dependencies
pip install -r requirements.txt

# 3. Start the agent
python main.py

# 4. Invoke
Invoke-RestMethod -Method POST http://localhost:8088/responses `
  -ContentType "application/json" `
  -Body '{"input": "What tools do you have?"}'
```

--------------------------------

### Start Pipecat WebRTC Server

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/pipecat-webrtc/README.md

Sets up a Python virtual environment, installs dependencies, and starts the Pipecat WebRTC server. Ensure the .env file is configured before running.

```bash
cd samples/python/hosted-agents/bring-your-own/invocations_ws/pipecat-webrtc

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

# Make sure .env is filled in (see "Server" table above)
python server.py
```

--------------------------------

### Start Pipecat WebSocket Server

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/pipecat-ws-server/README.md

Navigate to the server directory, set up a virtual environment, install dependencies, and run the server script. Ensure the .env file is configured.

```bash
cd samples/python/hosted-agents/bring-your-own/invocations_ws/pipecat-ws-server

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

# Make sure .env is filled in (see "Server" table above)
python server.py
```

--------------------------------

### Install Sample Dependencies

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/javascript/foundry-local/README.md

Navigate to a specific sample directory and install its Node.js dependencies.

```bash
cd native-chat-completions
npm install
```

--------------------------------

### Run the Transcription Example

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/javascript/foundry-local/live-audio-transcription/README.md

Execute the main application file to start the live audio transcription.

```bash
node app.js
```

--------------------------------

### Example Script Arguments for Travel Guide

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/agent-framework/responses/07-skills/skills/travel-guide/SKILL.md

Demonstrates the format and possible arguments for the `create_travel_guide.py` script. Use this to construct CLI arguments for generating a custom travel guide.

```json
[
  "--city", "Lisbon", "--days", "3", "--interests", "food,viewpoints,neighborhoods", "--tone", "first-time visitor"
]
```

--------------------------------

### Install Sample Dependencies

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/foundry-local/README.md

Navigate to a specific sample directory and install its Python dependencies using pip.

```bash
cd native-chat-completions
pip install -r requirements.txt
```

--------------------------------

### Smaller/Cheaper Deployment Example

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/infrastructure/infrastructure-setup-bicep/16-private-network-standard-agent-apim-setup/extensions/byom-cross-region/preflight/README.md

Use this example to verify a deployment with a single model on the backend and reduced capacity.

```powershell
.
check-region.ps1 -Subscription <sub-id> `
  -ProjectRegion eastus -BackendRegion eastus2 `
  -BackendModels 'gpt-4o@2024-11-20' `
  -BackendModelCapacity 5
```

--------------------------------

### Manual Setup for Agent

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/csharp/hosted-agents/bring-your-own/responses/background-agent/README.md

Set the `FOUNDRY_PROJECT_ENDPOINT` and `AZURE_AI_MODEL_DEPLOYMENT_NAME` environment variables, then run the agent using `dotnet run`. The agent will start on http://localhost:8088/.

```bash
export FOUNDRY_PROJECT_ENDPOINT="https://your-resource.services.ai.azure.com/api/projects/your-project"
export AZURE_AI_MODEL_DEPLOYMENT_NAME="gpt-5.4-mini"
dotnet run
```

--------------------------------

### Install Sample Dependencies

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/foundry-local/web-server-responses/README.md

Install the necessary Python packages for the sample. This includes the Foundry Local SDK and the OpenAI client. Specific SDK versions may be installed based on the operating system.

```bash
pip install -r requirements.txt
```

--------------------------------

### Setup and Run Local Agent

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/hello-world/README.md

Installs dependencies, sets up the environment, and runs the local agent using Uvicorn. Ensure to fill in your Azure Voice Live endpoint and model in the .env file.

```bash
cd samples/python/hosted-agents/bring-your-own/invocations_ws/hello-world

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

cp .env.example .env
# Fill in AZURE_VOICELIVE_ENDPOINT and AZURE_VOICELIVE_MODEL.
export $(grep -v '^#' .env | xargs)

python main.py
# → Uvicorn running on http://0.0.0.0:8088
```

--------------------------------

### Client Configuration Example

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/pipecat-webrtc/README.md

Example environment variables for the chat client, showing how to configure it for local server connection or a Foundry-hosted agent.

```bash
PROJECT_ENDPOINT=https://{account}.services.ai.azure.com/api/projects/{project}
PIPECAT_WEBRTC_AGENT_NAME=pipecat-ws-webrtc
# PIPECAT_WEBRTC_LOCAL_URL=ws://localhost:8089/invocations_ws
```

--------------------------------

### Start Pipecat WebRTC Chat Client

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/pipecat-webrtc/README.md

Sets up the Python virtual environment, installs dependencies, and starts the web portal client. It configures the client to connect to the locally running server by setting the PIPECAT_WEBRTC_LOCAL_URL environment variable.

```bash
cd samples/python/hosted-agents/bring-your-own/invocations_ws/pipecat-webrtc/chat_client

python -m venv ../.venv          # (or reuse the server's venv)
source ../.venv/bin/activate
pip install -r requirements.txt

# Local mode — point at the server you just started
echo 'PIPECAT_WEBRTC_LOCAL_URL=ws://localhost:8089/invocations_ws' >> .env

python web_portal.py             # listens on :9528 (open http://localhost:9528)
```

--------------------------------

### Python Agent Hosting and Server Setup (main.py)

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/agent-framework/responses/09-declarative-customer-support/README.md

Sets up and hosts the customer support workflow as an agent using Python. It initializes agents, registers them with a factory, loads the workflow, and starts a REST API endpoint compatible with the OpenAI Responses protocol.

```python
import os

from dotenv import load_dotenv

from agent_framework.foundry.chat_client import FoundryChatClient
from agent_framework.foundry.responses_host import ResponsesHostServer
from agent_framework.foundry.workflow_factory import WorkflowFactory

from .agents import TriageAgent, TechSupportAgent, BillingAgent
from .responses import TriageResponse


def main():
    load_dotenv()

    # Initialize the Foundry chat client
    client = FoundryChatClient(
        api_key=os.environ.get("FOUNDRY_API_KEY")
    )

    # Initialize agents
    triage_agent = TriageAgent(client=client, response_format=TriageResponse)
    tech_support_agent = TechSupportAgent(client=client)
    billing_agent = BillingAgent(client=client)

    # Create a workflow factory and register agents
    factory = WorkflowFactory()
    factory.register("TriageAgent", triage_agent)
    factory.register("TechSupportAgent", tech_support_agent)
    factory.register("BillingAgent", billing_agent)

    # Load the workflow from YAML
    workflow_agent = factory.load_workflow("workflow.yaml").as_agent()

    # Initialize and run the Responses host server
    server = ResponsesHostServer(agent=workflow_agent)
    server.run()


if __name__ == "__main__":
    main()
```

--------------------------------

### Install Dependencies

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/foundry-models/model-router/README.md

Install project dependencies by running this command after creating a virtual environment.

```bash
pip install -r requirements.txt
```

--------------------------------

### Install Dependencies and Run Application

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/foundry-local/embeddings/README.md

Installs project dependencies using pip and then runs the main application script.

```bash
pip install -r requirements.txt
python src/app.py
```

--------------------------------

### Install Python Dependencies

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/livekit-server/chat_client/README.md

Navigate to the chat client directory, create a virtual environment, activate it, and install the required Python packages.

```bash
cd samples/python/hosted-agents/bring-your-own/invocations_ws/livekit-server/chat_client

python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
```

--------------------------------

### Install Dependencies

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/foundry-local/live-audio-transcription/README.md

Install the required Python packages for the project. `pyaudio` is optional and provides microphone capture.

```bash
pip install -r requirements.txt
```

```bash
pip install pyaudio
```

--------------------------------

### Run a Sample

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/javascript/foundry-local/README.md

Execute a sample application after installing its dependencies.

```bash
npm start
```

--------------------------------

### Set up .env file with azd

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations/github-copilot/README.md

Copy the example .env file and set your GitHub token. This is automatically loaded when running locally with `azd`.

```bash
cp .env.example .env  # skip if .env already exists
# Edit .env and set GITHUB_TOKEN=github_pat_...
```

--------------------------------

### Manual Agent Setup and Run

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/csharp/hosted-agents/bring-your-own/invocations/notetaking-agent/README.md

Build the project, copy and edit the environment file, set necessary environment variables for project endpoint and model deployment, and then run the application.

```bash
dotnet build
cp .env.example .env  # then edit values (skip if .env already exists)
export FOUNDRY_PROJECT_ENDPOINT="https://your-project.services.ai.azure.com/api/projects/your-project"
export AZURE_AI_MODEL_DEPLOYMENT_NAME="gpt-5.4-mini"
dotnet run
```

--------------------------------

### Deploy Dynamic Discovery APIM Connection using Azure CLI

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/infrastructure/infrastructure-setup-bicep/01-connections/apim/README.md

Deploys an APIM connection with dynamic model discovery. This requires prior configuration of list and get endpoints in APIM as per the setup guide. Edit the parameters file with your resource IDs before deployment. The API key is automatically retrieved from the APIM service.

```bash
az deployment group create \
  --resource-group <your-resource-group> \
  --template-file connection-apim.bicep \
  --parameters @samples/parameters-dynamic-discovery.json
```

--------------------------------

### Start LangGraph Agent (Linux/macOS)

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/langgraph-toolbox-user-identity/README.md

Starts the LangGraph agent on Linux or macOS. Ensure environment variables in .env are set and dependencies are installed.

```bash
# 1. Copy and fill in the environment file
cp .env.example .env  # skip if .env already exists
# Edit .env — set FOUNDRY_PROJECT_ENDPOINT, MODEL_DEPLOYMENT_NAME,
#              and TOOLBOX_ENDPOINT at minimum

# 2. Install dependencies
pip install -r requirements.txt

# 3. Start the agent
python main.py

# 4. Invoke
curl -X POST http://localhost:8088/responses \
  -H "Content-Type: application/json" \
  -d '{"input": "What tools do you have?"}'
```

--------------------------------

### Start LangGraph Agent (Windows PowerShell)

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/langgraph-toolbox-user-identity/README.md

Starts the LangGraph agent on Windows using PowerShell. Ensure environment variables in .env are set and dependencies are installed.

```powershell
# 1. Copy and fill in the environment file
Copy-Item .env.example .env  # skip if .env already exists
# Edit .env — set FOUNDRY_PROJECT_ENDPOINT, MODEL_DEPLOYMENT_NAME,
#              and TOOLBOX_ENDPOINT at minimum

# 2. Install dependencies
pip install -r requirements.txt

# 3. Start the agent
python main.py

# 4. Invoke
Invoke-RestMethod -Method POST http://localhost:8088/responses `
  -ContentType "application/json" `
  -Body '{"input": "What tools do you have?"}'
```

--------------------------------

### Run a Sample Application

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/foundry-local/README.md

Execute a Python sample application after installing its dependencies.

```bash
python src/app.py
```

--------------------------------

### Start Web Portal

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/pipecat-ws-server/README.md

Navigate to the chat_client directory and run the web portal script to connect to the agent.

```bash
cd chat_client && python web_portal.py
```

--------------------------------

### Start Azure Function Locally

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/infrastructure/infrastructure-setup-bicep/19-private-network-agent-tools/azure-function-server/README.md

Navigates to the function server directory, installs Python dependencies, and starts the Azure Functions host locally. This allows for local testing and debugging.

```bash
cd azure-function-server
pip install -r requirements.txt
func start
```

--------------------------------

### Manual Setup for Claude Agent

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations/claude-agent-sdk/README.md

Install dependencies, copy the environment file, and run the main Python script to set up the Claude agent manually. Ensure you have Python 3.10+ installed.

```bash
pip install -r requirements.txt
cp .env.example .env  # skip if .env already exists
python main.py

```

--------------------------------

### Setup Local Diagnostic Agent

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations/diagnostic-agent/README.md

Set up a Python virtual environment, install dependencies, and run the agent locally. The agent will listen on http://localhost:8088/.

```bash
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python main.py
```

--------------------------------

### Run the Sample Application

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/foundry-local/web-server-responses/README.md

Execute the main Python script to start the local web service, send API requests, and process the responses. This command assumes you are in the sample's root directory.

```bash
python -m venv .venv
.
.venv\Scripts\activate
pip install -r requirements.txt
python src\app.py
```

--------------------------------

### Run Main Sample (Maven)

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/java/enterprise-agent-tutorial/1-idea-to-prototype/README.md

Executes the main Modern Workplace Assistant application using Maven. This starts the agent.

```bash
mvn exec:java -Dexec.mainClass="com.microsoft.azure.samples.ModernWorkplaceAssistant"
```

--------------------------------

### Set up local environment variables

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/voicelive/foreground-background-agents-responses-voicelive/README.md

Copy the example environment file and edit it with your specific values. Then, source the file to load the variables into your current shell session.

```bash
cp .env.example .env
# Edit .env with your values
source .env
```

--------------------------------

### Audio Playback Setup

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/csharp/hosted-agents/bring-your-own/invocations_ws/HelloWorld/chat_client/index.html

Initializes or reinitializes the AudioContext for playback, ensuring the correct sample rate. Schedules audio playback to start at a specific time to manage timing.

```javascript
function ensurePlayCtx(sampleRate) {
  if (!playCtx || playCtx.sampleRate !== sampleRate) {
    if (playCtx) try { playCtx.close(); } catch {}
    playCtx = new (window.AudioContext || window.webkitAudioContext)({ sampleRate });
    playTime = playCtx.currentTime;
  }
}

function playPcm(uint8, sampleRate, channels) {
  ensurePlayCtx(sampleRate);
  const i16 = new Int16Array(uint8.buffer, uint8.byteOffset, uint8.byteLength / 2);
  const frames = i16.length / channels;
  const buf = playCtx.createBuffer(channels, frames, sampleRate);
  for (let c = 0; c < channels; c++) {
    const ch = buf.getChannelData(c);
    for (let i = 0; i < frames; i++) ch[i] = i16[i * channels + c] / 32768;
  }
  const src = playCtx.createBufferSource();
  src.buffer = buf;
  src.connect(playCtx.destination);
  const now = playCtx.currentTime;
  if (playTime < now) playTime = now;
  src.start(playTime);
  playTime += buf.duration;
  playSources.push(src);
  src.onended = () => {
    const i = playSources.indexOf(src);
    if (i >= 0) playSources.splice(i, 1);
  };
}

// Stop and drop all queued playback (e.g. when user interrupts the bot).
function cancelPlayback() {
  for (const s of playSources) {
    try { s.stop(); } catch {}
  }
  playSources = [];
  if (playCtx) playTime = playCtx.currentTime;
}
```

--------------------------------

### Set up Environment Variables Locally

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/hello-world/README.md

Copy the example environment file and edit it with your specific values. Source the file to load the variables into your current shell session. This is for local development when not using `azd ai agent run`.

```bash
cp .env.example .env  # skip if .env already exists
# Edit .env with your values
source .env
```

--------------------------------

### Test Hosted Agent Responses with curl (Streaming)

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/langgraph-chat/README.md

Send a POST request to the agent's endpoint to get a response. This example demonstrates streaming responses and chaining conversations using `previous_response_id`.

```bash
# Turn 1 — ask for the time (triggers tool call)
curl -N -X POST http://localhost:8088/responses \
    -H "Content-Type: application/json" \
    -d '{"model": "chat", "input": "What time is it right now?", "stream": true}'

# Turn 2 — chain via previous_response_id
curl -N -X POST http://localhost:8088/responses \
    -H "Content-Type: application/json" \
    -d '{"model": "chat", "input": "What is 42 * 17?", "previous_response_id": "<ID>", "stream": true}'

# Turn 3 — context recall
curl -N -X POST http://localhost:8088/responses \
    -H "Content-Type: application/json" \
    -d '{"model": "chat", "input": "Add 100 to that result", "previous_response_id": "<ID>", "stream": true}'
```

--------------------------------

### Run Live Audio Transcription Example

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/rust/foundry-local/live-audio-transcription/README.md

Navigate to the example directory and run the transcription. Use Ctrl+C to stop the live process. Alternatively, run with the --synth flag to generate synthetic audio if no microphone is available.

```bash
cd samples/rust/live-audio-transcription-example

# Live microphone (press Ctrl+C to stop)
cargo run

# Synthetic 440Hz sine wave (no microphone needed)
cargo run -- --synth
```

--------------------------------

### Setup and Run Duplex Live Agent Locally

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations_ws/duplex-live-agent/README.md

Installs dependencies, sets up environment variables, and runs the agent locally. Ensure to fill in your Azure Voice Live endpoint and model in the .env file.

```bash
cd duplex-live-agent

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt websockets

cp .env.example .env
# Fill in AZURE_VOICELIVE_ENDPOINT and AZURE_VOICELIVE_MODEL.
export $(grep -v '^#' .env | xargs)

python main.py
# → Uvicorn running on http://0.0.0.0:8088
```

--------------------------------

### Manual Agent Setup

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/invocations/human-in-the-loop/README.md

Manually set up and run the agent by installing dependencies, configuring environment variables, and executing the main Python script. Ensure you edit the .env file with your specific Azure OpenAI details.

```bash
pip install -r requirements.txt
cp .env.example .env  # then edit values (skip if .env already exists)
export FOUNDRY_PROJECT_ENDPOINT="https://your-project.services.ai.azure.com/api/projects/your-project"
export AZURE_AI_MODEL_DEPLOYMENT_NAME="gpt-5.4-mini"
python main.py
```

--------------------------------

### Install Dependencies

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/typescript/enterprise-agent-tutorial/1-idea-to-prototype/README.md

Installs project dependencies using npm. Ensure Node.js 20.0.0 or later is installed.

```bash
npm install
```

--------------------------------

### Install Python Dependencies

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/infrastructure/infrastructure-setup-bicep/19-private-network-agent-tools/tests/TESTING-GUIDE.md

Installs the required Python packages for running agent tests. Ensure you have Python and pip installed.

```bash
pip install azure-ai-projects azure-identity openai
```

--------------------------------

### Navigate to Sample Directory

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/csharp/hosted-agents/agent-framework/README.md

Navigate to the specific sample directory you want to explore.

```bash
cd hello-world
```

--------------------------------

### Install Foundry CLI Extension Bundle

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/agent-framework/responses/04-foundry-toolbox/README.md

Installs the unified Foundry CLI extension bundle. Uninstall individual extensions if previously installed.

```bash
# If you previously installed individual extensions, uninstall them first:
#   azd ext uninstall azure.ai.agents
#   azd ext uninstall azure.ai.toolboxes
azd ext install microsoft.foundry
```

--------------------------------

### Install Dependencies

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/javascript/foundry-local/live-audio-transcription/README.md

Install the necessary Node.js packages for the project.

```bash
npm install
```

--------------------------------

### Install Unified Foundry CLI Extension Bundle

Source: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/bring-your-own-toolbox/README.md

Install the unified Foundry CLI extension bundle using `azd ext install`. This bundle includes `azd ai agent`, `connection`, `inspector`, `project`, `routine`, `skill`, and `toolbox`. Uninstall individual extensions first if they were previously installed.

```bash
# If you previously installed individual extensions, uninstall them first:
#   azd ext uninstall azure.ai.agents
#   azd ext uninstall azure.ai.toolboxes

# Install the unified bundle (provides azd ai agent, connection, inspector,
# project, routine, skill, and toolbox).
azd ext install microsoft.foundry
```