### Install and Run Gemini ADK Agent (Bash)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/gemini-fullstack/README.md

Installs project dependencies and starts the development servers for the Gemini Fullstack ADK agent. Assumes the user is in the correct project directory.

```bash
make install && make dev
```

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

### Navigate and Run Agent from New Project Directory (Bash)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/gemini-fullstack/README.md

Changes the directory to the newly created agent project and then installs dependencies and starts the development servers.

```bash
cd my-fullstack-agent && make install && make dev
```

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

### Create Virtual Environment and Install Agent Starter Pack (Python/Bash)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/gemini-fullstack/README.md

Creates a Python virtual environment and installs the agent-starter-pack. It then uses the starter pack to create a new fullstack agent project.

```bash
# Create and activate a virtual environment
python -m venv .venv && source .venv/bin/activate # On Windows: .venv\Scripts\activate

# Install the starter pack and create your project
pip install --upgrade agent-starter-pack
agent-starter-pack create my-fullstack-agent -a adk@gemini-fullstack
```

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

### Create Project with Agent Starter Pack (venv)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/blog-writer/README.md

Creates and activates a virtual environment, then installs the agent-starter-pack and creates a new blogger-agent project. This is a standard Python project setup.

```bash
# Create and activate a virtual environment
python -m venv .venv && source .venv/bin/activate
# On Windows: .venv\\Scripts\\activate

# Install the starter pack and create your project
pip install --upgrade agent-starter-pack
agent-starter-pack create my-blogger-agent -a adk@blogger-agent
```

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

### Run Deployment Script

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/brand-search-optimization/README.md

This bash script automates the setup process for the brand search optimization agent. It clones the repository, installs Python dependencies using Poetry, and populates the BigQuery database.

```Bash
# Clone this repository.
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/brand-search-optimization

# Run This script
# 1. Creates and activates a new virtual env
# 2. Installs python packages
# 3. Populates BigQuery data using variables set in `.env` file
sh deployment/run.sh
```

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

### Start PostgreSQL on MacOS

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/software-bug-assistant/README.md

Starts a local PostgreSQL instance using Homebrew services on macOS. This command assumes PostgreSQL has been installed via Homebrew.

```bash
brew services start postgresql
```

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

### Install Dependencies with Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/RAG/README.md

Installs project dependencies using Poetry. It includes a note for Linux users to disable keyring if they encounter errors during installation.

```bash
poetry config keyring.enabled false
poetry install
```

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

### Install Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/customer-service/README.md

Installs the Poetry dependency manager if it's not already installed.

```bash
pip install poetry
```

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

### Run MCP Airbnb Example

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

This snippet shows how to execute the MCP Airbnb example using Python. It requires Node.js and npx to be installed and configured correctly. The command initiates the test from the travel-concierge directory.

```Shell
python -m tests.mcp_abnb
```

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

### Install and Deploy Financial Advisor

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/data-engineering/README.md

Installs project dependencies with deployment configurations and deploys the Financial Advisor agent to Vertex AI Agent Engine. Requires `poetry` for installation.

```bash
poetry install --with deployment
python3 deployment/deploy.py --create
```

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

### MCP Toolbox Server Output

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/software-bug-assistant/README.md

This is an example of the expected output when the MCP Toolbox server starts successfully. It indicates the initialization of sources, authentication services, tools, and toolsets, confirming the server is ready.

```bash
2025-05-30T02:06:57.479344419Z INFO "Initialized 1 sources."
2025-05-30T02:06:57.479696869Z INFO "Initialized 0 authServices."
2025-05-30T02:06:57.479973769Z INFO "Initialized 9 tools."
2025-05-30T02:06:57.480054519Z INFO "Initialized 2 toolsets."
2025-05-30T02:06:57.480739499Z INFO "Server ready to serve!"
```

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

### Install Project Dependencies with uv

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/blog-writer/README.md

Installs project dependencies from requirements.txt using uv. This command also sets up a virtual environment for the project.

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

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

### Install Project Dependencies with Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/customer-service/README.md

Installs all project dependencies defined in the pyproject.toml file using Poetry.

```bash
poetry install
```

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

### Clone Repository and Install Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/academic-research/README.md

Clones the adk-samples repository and installs project dependencies using Poetry. Includes a note for Linux users on disabling the keyring if installation errors occur.

```Bash
# Clone this repository.
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/academic-research
# Install the package and dependencies.
# Note for Linux users: If you get an error related to `keyring` during the installation, you can disable it by running the following command:
# poetry config keyring.enabled false
# This is a one-time setup.
poetry install
```

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

### Install Poetry and Virtual Environment

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/personalized-shopping/README.md

This snippet shows the commands to set up a Python virtual environment, install Poetry for dependency management, and activate the environment.

```Bash
cd ~\npython3 -m venv myenv\nsource ~/myenv/bin/activate\npip install poetry
```

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

### Clone Repository and Install Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/fomc-research/README.md

Clones the ADK samples repository and navigates to the FOMC Research agent directory. It then installs Poetry, configures it to disable keyring if necessary, and installs the agent's Python requirements.

```bash
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/fomc-research
pip install poetry
poetry config keyring.enabled false
poetry install
```

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

### Agent Response Example (What to Pack)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/tests/post_booking_sample.md

An example of a response from the 'what_to_pack_agent' detailing suggested items for a trip, such as clothing and accessories.

```json
Response from: "what_to_pack_agent"
response: {
  "items": [
    "light jacket",
    "umbrella",
    "comfortable walking shoes",
    "layers",
    "sunscreen"
  ]
}
```

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

### Install Agent Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/auto-insurance-agent/README.md

Installs the project's dependencies using Poetry.

```bash
poetry install
```

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

### Deployment Dockerfile Example

Source: https://github.com/mbettan/adk-samples/blob/main/java/README.md

A Dockerfile example for packaging the Java agent into a container image. This allows for consistent deployment across different environments.

```Dockerfile
FROM openjdk:17-jdk-slim

WORKDIR /app

COPY target/your-agent.jar /app/agent.jar

CMD ["java", "-jar", "agent.jar"]
```

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

### Install Dependencies with Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

Installs all necessary project dependencies using Poetry. This command should be run after cloning the repository and potentially disabling the keyring.

```bash
poetry install
```

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

### Install uv

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/blog-writer/README.md

Installs the uv dependency manager using pip. uv is used for efficient dependency management in Python projects.

```bash
pip install uv
```

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

### Clone Repository and Install Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/financial-advisor/README.md

Clones the adk-samples repository and installs the project and its dependencies using Poetry. Includes a note for Linux users regarding keyring configuration.

```bash
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/financial_advisor
# Install the package and dependencies.
# Note for Linux users: If you get an error related to `keyring` during the installation, you can disable it by running the following command:
# poetry config keyring.enabled false
# This is a one-time setup.
poetry install
```

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

### Example Agent Interaction

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/fomc-research/README.md

Demonstrates a typical interaction with the FOMC Research agent, starting with a greeting and then providing a specific date for analysis. It shows the expected input and output format.

```bash
$ adk run .
Log setup complete: /tmp/agents_log/agent.20250405_140937.log
To access latest log: tail -F /tmp/agents_log/agent.latest.log
Running agent root_agent, type exit to exit.
user: Hello. What can you do for me?
[root_agent]: I can help you analyze past Fed Open Market Committee (FOMC) meetings and provide you with a thorough analysis report. To start, please provide the date of the meeting you would like to analyze. If you have already provided it, please confirm the date. I need the date in ISO format (YYYY-MM-DD).

user: 2025-01-29
[analysis_agent]: Here is a summary and analysis of the January 29, 2025 FOMC meeting, based on the available information:
...
```

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

### Clone Gemini Fullstack ADK Repository (Bash)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/gemini-fullstack/README.md

Clones the Gemini Fullstack ADK sample repository and navigates into the project directory. This is the first step for setting up the agent using Google AI Studio.

```bash
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/gemini-fullstack
```

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

### Create Project using uv with Agent Starter Pack (Bash)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/gemini-fullstack/README.md

A shortcut command using 'uvx' to create a new fullstack agent project with the ADK template, bypassing the need to pre-install the starter pack into a virtual environment.

```bash
uvx agent-starter-pack create my-fullstack-agent -a adk@gemini-fullstack
```

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

### Install Project Dependencies with Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/personalized-shopping/README.md

This command installs the project's dependencies using Poetry after navigating to the project directory. It includes a note for Linux users on disabling the keyring if necessary.

```Bash
cd adk-samples/python/agents/personalized-shopping\n# Note for Linux users: If you get an error related to `keyring` during the installation, you can disable it by running the following command:\n# poetry config keyring.enabled false\n# This is a one-time setup.\npoetry install
```

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

### Start Local PostgreSQL Instance (macOS)

Source: https://github.com/mbettan/adk-samples/blob/main/java/agents/software-bug-assistant/README.md

Starts a local PostgreSQL instance using Homebrew services on macOS. This is required for the Software Bug Assistant agent to connect to its PostgreSQL bug ticket database.

```Bash
brew services start postgresql
```

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

### Install Development Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/machine-learning-engineering/README.md

Installs the necessary dependencies for development and testing of the machine learning engineering agent.

```bash
poetry install --with dev
```

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

### Install Development Dependencies with Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/llm-auditor/README.md

This command installs the necessary development dependencies for the project using Poetry. The `--with dev` flag ensures that only the dependencies required for development and testing are installed, optimizing the installation process.

```bash
poetry install --with dev
```

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

### Install Project Dependencies with Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/persona_ad_gen/README.md

Installs all project dependencies using Poetry and creates a virtual environment.

```bash
poetry install
```

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

### Start ADK Web Interface

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/image-scoring/README.md

Launches the ADK web interface, which provides a graphical chatbot interface for interacting with agents like the Image Scoring Agent.

```Bash
adk web
```

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

### Deploy Agent to Google Cloud Development Environment

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/gemini-fullstack/README.md

Commands to set the Google Cloud project ID and build the backend for a development deployment. Ensure you replace 'YOUR_DEV_PROJECT_ID' with your actual project ID.

```bash
# Replace YOUR_DEV_PROJECT_ID with your actual Google Cloud Project ID
gcloud config set project YOUR_DEV_PROJECT_ID
make backend
```

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

### Install Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/machine-learning-engineering/README.md

Installs the Poetry package manager, which is used for dependency management and packaging in Python projects. It reads the pyproject.toml file to install dependencies into a virtual environment.

```bash
pip install poetry
```

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

### Install Deployment Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/marketing-agency/README.md

Installs the dependencies required for deploying the marketing agency to Vertex AI Agent Engine.

```bash
poetry install --with deployment
```

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

### Install Development Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/marketing-agency/README.md

Installs the necessary dependencies for running tests and evaluation for the marketing agency project.

```bash
poetry install --with dev
```

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

### Install Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/auto-insurance-agent/README.md

Installs Poetry, a dependency management and packaging tool for Python, using pip.

```bash
pip install poetry
```

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

### Run ADK Agent with Web UI

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/RAG/README.md

Starts the ADK Web UI, allowing the RAG agent to be selected and run from a dropdown menu.

```bash
adk web
```

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

### Install Development Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/data-engineering/README.md

Installs additional dependencies required for running tests and evaluations for the agent.

```bash
poetry install --with dev
```

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

### Start ADK Web Interface

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/machine-learning-engineering/README.md

Starts a local web server for interacting with the ADK agent, providing a web-based interface for communication.

```bash
adk web
```

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

### Install Dependencies with uv

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/data-science/README.md

Installs project dependencies using uv by reading the pyproject.toml file. This command also creates a virtual environment if one doesn't exist.

```bash
uv sync
```

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

### Configure Research Parameters in Python

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/gemini-fullstack/README.md

Example of how to modify key research parameters, such as Gemini models and research loop iterations, within the 'ResearchConfiguration' dataclass in the Python backend code.

```python
You can edit key parameters (Gemini models, research loop iterations) in the `ResearchConfiguration` dataclass within `app/config.py`.
```

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

### Install Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/financial-advisor/README.md

Installs the Poetry dependency management tool using pip. This is a prerequisite for managing project dependencies.

```bash
pip install poetry
```

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

### Clone Repository and Install Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/camel/README.md

Clones the adk-samples repository and installs the project's Python dependencies using Poetry. Includes a note for Linux users regarding keyring configuration.

```bash
# Clone this repository.
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/camel
# Install the package and dependencies.
# Note for Linux users: If you get an error related to `keyring` during the installation, you can disable it by running the following command:
# poetry config keyring.enabled false
# This is a one-time setup.
poetry install
```

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

### Install Development Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/academic-research/README.md

Installs the necessary dependencies for running tests and evaluation. This command uses 'poetry' to manage project dependencies.

```bash
poetry install --with dev
```

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

### Start ADK Web Interface

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/academic-research/README.md

Starts a local web server using the ADK tool, providing a graphical user interface for interacting with agents. Users can select agents and engage in conversations through a chatbot interface.

```Bash
adk web
```

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

### Run Agent via Web Interface

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/personalized-shopping/README.md

Start a local web server to interact with the agent through a chatbot interface. This involves navigating to the agent's directory and running the 'adk web' command.

```bash
cd personalized-shopping
adk web
```

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

### Clone and Install Marketing Agency Agent

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/marketing-agency/README.md

Clones the adk-samples repository and installs the marketing-agency Python package along with its dependencies using Poetry. Includes a note for Linux users regarding keyring configuration.

```bash
# Clone this repository.
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/marketing-agency
# Install the package and dependencies.
# Note for Linux users: If you get an error related to `keyring` during the installation, you can disable it by running the following command:
# poetry config keyring.enabled false
# This is a one-time setup.
poetry install
```

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

### Set Gemini ADK Environment Variables (Bash)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/gemini-fullstack/README.md

Creates a .env file in the app directory to configure the Gemini ADK. It sets the use of Google AI Studio and specifies the API key.

```bash
echo "GOOGLE_GENAI_USE_VERTEXAI=FALSE" >> app/.env
echo "GOOGLE_API_KEY=YOUR_AI_STUDIO_API_KEY" >> app/.env
```

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

### Install Deployment Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/machine-learning-engineering/README.md

Installs the dependencies required for deploying the machine learning engineering agent to Vertex AI Agent Engine.

```bash
poetry install --with deployment
```

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

### Start ADK Web Interface

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/marketing-agency/README.md

Starts a local web server for the ADK, allowing interaction with agents through a web-based chatbot interface. The marketing_agency can be selected from a dropdown.

```bash
adk web
```

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

### Install Project Dependencies with Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/machine-learning-engineering/README.md

Installs all project dependencies defined in the pyproject.toml file using Poetry. It also provides a workaround for potential 'keyring' errors on Linux by disabling it.

```bash
# Install the Poetry package and dependencies.
# Note for Linux users: If you get an error related to `keyring` during the installation, you can disable it by running the following command:
# poetry config keyring.enabled false
# This is a one-time setup.
poetry install
```

```bash
python -m poetry install
```

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

### Run Agent via ADK Web UI

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/customer-service/README.md

Starts the ADK Web UI, allowing the customer service agent to be selected and run from a dropdown menu.

```bash
adk web
```

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

### Run ADK Java Agent with Maven

Source: https://github.com/mbettan/adk-samples/blob/main/java/agents/software-bug-assistant/README.md

Maven command to compile the ADK Java agent code and start a development web server with specified configurations for server port, agent source directory, and logging levels.

```bash
cd software-bug-assistant/
mvn compile exec:java "-Dexec.args=--server.port=8080 \
     --adk.agents.source-dir=src/ \
     --logging.level.com.google.adk.dev=DEBUG \
     --logging.level.com.google.adk.demo.agents=DEBUG"
```

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

### Run Financial Advisor Agent via Web Interface

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/financial-advisor/README.md

Starts a local web server using the ADK CLI, allowing interaction with the financial advisor agent through a web browser.

```bash
adk web
```

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

### Clone Repository and Install Agent

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/data-engineering/README.md

Clones the ADK samples repository and installs the Data Engineering Agent along with its dependencies using Poetry.

```bash
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/data-engineering-agent
poetry install
```

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

### Install Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/marketing-agency/README.md

Installs the Poetry dependency management tool for Python projects. Poetry is used for packaging and managing project dependencies.

```bash
pip install poetry
```

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

### Install ADK Development Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/financial-advisor/README.md

Installs the necessary development dependencies for the ADK project using Poetry, enabling testing and evaluation functionalities.

```bash
poetry install --with dev
```

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

### Start MCP Inspector

Source: https://github.com/mbettan/adk-samples/blob/main/java/agents/software-bug-assistant/README.md

Command to download and run the web-based MCP Inspector tool to test the MCP Toolbox server.

```bash
npx @modelcontextprotocol/inspector
```

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

### Install Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/data-engineering/README.md

Installs the Poetry package manager, used for dependency management and packaging in Python projects.

```bash
pip install poetry
```

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

### Clone and Install ADK Samples

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/image-scoring/README.md

Clones the ADK samples repository and installs the Image Scoring Agent along with its deployment dependencies using Poetry. It also provides a workaround for potential keyring issues on Linux.

```Bash
# Clone this repository.
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/image-scoring
# Install the package and dependencies.
# Note for Linux users: If you get an error related to `keyring` during the installation, you can disable it by running the following command:
# poetry config keyring.enabled false
# This is a one-time setup.
poetry install --with deployment
```

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

### Disable Keyring for Poetry Installation

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

Disables the keyring feature in Poetry, which can sometimes cause issues during installation on Linux systems. This is a one-time setup command.

```bash
poetry config keyring.enabled false
```

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

### Clone Repository and Install Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/llm-auditor/README.md

Clones the adk-samples repository and installs the LLM Auditor agent's Python package and its dependencies using Poetry. Includes a note for Linux users regarding keyring configuration.

```bash
# Clone this repository.
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/llm-auditor
# Install the package and dependencies.
# Note for Linux users: If you get an error related to `keyring` during the installation, you can disable it by running the following command:
# poetry config keyring.enabled false
# This is a one-time setup.
poetry install
```

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

### Install Testing Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/data-science/README.md

Installs necessary dependencies for running tests and evaluations for the Data Science agent. This command should be executed from the 'agents/data-science' directory.

```bash
uv sync
```

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

### Create Project with Agent Starter Pack (uvx)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/blog-writer/README.md

Creates a new blogger-agent project using uvx and the agent-starter-pack in a single command. This method avoids pre-installing the package into a virtual environment.

```bash
uvx agent-starter-pack create my-blogger-agent -a adk@blogger-agent
```

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

### Install Agent Dependencies (Bash)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/README.md

Installs the Python dependencies required for a specific agent using Poetry. This command is typically executed from the agent's main directory.

```bash
poetry install
```

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

### Install Deployment Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/academic-research/README.md

Installs dependencies required for deploying the Academic Co-Research agent. This command ensures all necessary packages for deployment are available.

```bash
poetry install --with deployment
```

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

### Install Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/camel/README.md

Installs the Poetry dependency management tool using pip. Poetry is required for managing packages and building the project.

```bash
pip install poetry
```

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

### Run ADK Python Agent Web Interface

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/software-bug-assistant/README.md

This bash command starts a web server for the ADK Python agent, allowing interaction through a graphical user interface. Access the provided URL to use the chatbot.

```bash
uv run adk web
```

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

### Start Travel Concierge API Server

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

Starts a FastAPI development server for the Travel Concierge package, making its API accessible at http://127.0.0.1:8000. API documentation is available at http://127.0.0.1:8000/docs.

```bash
adk api_server travel_concierge
```

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

### Start ADK Web Interface

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/llm-auditor/README.md

Launch a local web server using the ADK CLI to interact with agents through a graphical user interface. The command outputs the URL to access the interface.

```bash
adk web
```

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

### Run Travel Concierge Agent via Web Interface

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

Starts a local web server for the Travel Concierge agent using ADK. This provides a graphical interface for interacting with the agent via a web browser.

```bash
# Under the travel-concierge directory:
adk web
```

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

### Programmatic Example Output

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

Illustrates the output from the programmatic example, showing user input, agent transfers, and structured JSON responses from agents like the place agent. This output is used by the front-end to render information.

```text
[user]: "Inspire me about Maldives"

... 

[root_agent]: transfer_to_agent( {"agent_name": "inspiration_agent"} )

... 

[inspiration_agent]: place_agent responds -> {
  "id": "af-be786618-b60b-45ee-a801-c40fd6811e60",
  "name": "place_agent",
  "response": {
    "places": [
      {
        "name": "Malé",
        "country": "Maldives",
        "image": "https://upload.wikimedia.org/wikipedia/commons/thumb/1/1c/Male%2C_Maldives_panorama_2016.jpg/1280px-Male%2C_Maldives_panorama_2016.jpg",
        "highlights": "The vibrant capital city, offering bustling markets, historic mosques, and a glimpse into local Maldivian life.",
        "rating": "4.2"
      },
      {
        "name": "Baa Atoll",
        "country": "Maldives",
        "image": "https://upload.wikimedia.org/wikipedia/commons/thumb/9/95/Baa_Atoll_Maldives.jpg/1280px-Baa_Atoll_Maldives.jpg",
        "highlights": "A UNESCO Biosphere Reserve, famed for its rich marine biodiversity, including manta rays and whale sharks, perfect for snorkeling and diving.",
        "rating": "4.8"
      },
      {
        "name": "Addu Atoll",
        "country": "Maldives",
        "image": "https://upload.wikimedia.org/wikipedia/commons/thumb/c/c3/Addu_Atoll_Maldives.jpg/1280px-Addu_Atoll_Maldives.jpg",
        "highlights": "The southernmost atoll, known for its unique equatorial vegetation, historic WWII sites, and excellent diving spots with diverse coral formations.",
        "rating": "4.5"
      }
    ]
  }
}

[app]: To render a carousel of destinations

[inspiration_agent]: Maldives is an amazing destination! I see three great options:

1.  **Malé:** The capital city, where you can experience local life, markets, and mosques.
2.  **Baa Atoll:** A UNESCO Biosphere Reserve, perfect for snorkeling and diving, with manta rays and whale sharks.
3.  **Addu Atoll:** The southernmost atoll, offering unique vegetation, WWII history, and diverse coral for diving.

Are any of these destinations sound interesting? I can provide you with some activities you can do in the destination you selected.

[user]: "Suggest some acitivities around Baa Atoll"

... 

```

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

### Install Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/llm-auditor/README.md

Installs the Poetry dependency management tool using pip. Poetry is required for managing project dependencies and packaging.

```bash
pip install poetry
```

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

### Install Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/academic-research/README.md

Installs the Poetry dependency management tool for Python projects. Poetry is used for managing packages and project dependencies.

```Bash
pip install poetry
```

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

### Install Development Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/image-scoring/README.md

Installs the necessary extra dependencies for running evaluation and tests using Poetry. This command ensures that all development-related packages are available for the project.

```bash
poetry install --with dev
```

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

### Install Deployment Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/persona_ad_gen/README.md

This bash command installs the necessary dependencies for deploying the agent, including optional dependencies for deployment.

```Bash
poetry install --with deployment
```

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

### Clone Repository and Navigate

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/RAG/README.md

Clones the adk-samples repository and navigates to the RAG agent directory in Python. This is the initial step for setting up the project.

```bash
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/RAG
```

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

### Configure Environment Variables

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/fomc-research/README.md

Sets up environment variables by copying an example .env file and then exporting its contents into the current shell session. This is necessary for the agent to access project-specific configurations.

```bash
set -o allexport
. .env
set +o allexport
```

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

### Install ADK Evaluation Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/persona_ad_gen/README.md

Installs optional evaluation dependencies for the ADK, including pandas, tabulate, and rouge-score.

```bash
pip install "google-adk[eval]"

# If using pipx for ADK installation
pipx inject google-adk pandas tabulate rouge-score
```

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

### Install Eval Dependencies

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/persona_ad_gen/README.md

Installs necessary dependencies for the evaluation module, including pandas, tabulate, and rouge-score, using pipx.

```bash
pipx inject google-adk pandas tabulate rouge-score
```

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

### Run All Tests with Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

Installs development dependencies using Poetry and then executes all tests using pytest.

```bash
poetry install --with dev
pytest
```

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

### Install Poetry

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/image-scoring/README.md

Installs the Poetry dependency management tool, which is a prerequisite for setting up the Image Scoring Agent. Poetry is used for managing packages and dependencies in Python projects.

```Bash
pip install poetry
```

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

### Clone Repository and Navigate

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/customer-service/README.md

Clones the adk-samples repository from GitHub and navigates into the customer-service agent directory.

```bash
git clone https://github.com/google/adk-samples.git
cd adk-samples/python/agents/customer-service
```

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

### Initial Product Listing Data

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/personalized-shopping/tests/example_interactions/image_search_denim_skirt.session.md

Displays the initial product listing data, including size and color options, product name, price, and rating. This data is presented before specific sections like 'Features' or 'Reviews' are accessed.

```JSON
{
  "result": "Back to Search [SEP] < Prev [SEP] size [SEP] 12 plus [SEP] 12 plus petite [SEP] 14 plus [SEP] 14 plus petite [SEP] 16 plus [SEP] 16 plus petite [SEP] 18 plus [SEP] 18 plus petite [SEP] 20 plus [SEP] 20 plus petite [SEP] 22 plus [SEP] 22 plus petite [SEP] 24 plus [SEP] 24 plus petite [SEP] 26 plus [SEP] 26 plus petite [SEP] 28 plus [SEP] 28 plus petite [SEP] 30 plus [SEP] 30 plus petite [SEP] 32 plus [SEP] 32 plus petite [SEP] 34 plus [SEP] 34 plus petite [SEP] 36 plus [SEP] 36 plus petite [SEP] 38 plus [SEP] 38 plus petite [SEP] color [SEP] indigo [SEP] natural khaki [SEP] black [SEP] light wash [SEP] stonewash [SEP] white [SEP] Woman Within Women's Plus Size Petite Flared Denim Skirt [SEP] Price: $24.41 to $29.95 [SEP] Rating: N.A. [SEP] Description [SEP] Features [SEP] Reviews [SEP] Buy Now"
}
```

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

### Process Product Page Information

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/personalized-shopping/tests/example_interactions/text_search_floral_dress.session.md

This snippet shows the agent's response after clicking on a product. It includes available options like color and size, along with product details and navigation elements.

```json
Response from: "click"
response: {
  "result": "Back to Search [SEP] < Prev [SEP] color [SEP] black [SEP] blacka [SEP] blackb [SEP] blue [SEP] bluea [SEP] blueb [SEP] coffee [SEP] gray [SEP] green [SEP] hot pink [SEP] navy [SEP] pink [SEP] pinka [SEP] purple [SEP] purplea [SEP] sky blue [SEP] sky bluea [SEP] white [SEP] whitea [SEP] yellow [SEP] yellowa [SEP] size [SEP] small [SEP] medium [SEP] large [SEP] x-large [SEP] xx-large [SEP] Women Summer Cold Shoulder Dress,Beach Vintage Bohemian Print Sundress Party Prom Swing Loose A-Line Dress with Pocket [SEP] Price: $16.29 to $17.59 [SEP] Rating: N.A. [SEP] Description [SEP] Features [SEP] Reviews [SEP] Buy Now"
}
```

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

### Java Agent Main Entry Point (Main.java)

Source: https://github.com/mbettan/adk-samples/blob/main/java/README.md

The Main.java file serves as the entry point for running the Java agent, especially when executed as a standalone application. It typically initializes the agent and starts its execution.

```Java
package com.google.adk.samples.agent;

public class Main {
    public static void main(String[] args) {
        // Initialize and run the agent
        Agent agent = new Agent();
        // ... setup and execution logic ...
        System.out.println("Agent started.");
    }
}
```

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

### Run Blogger Agent Evaluation Tests

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/blog-writer/eval/README.md

Executes the evaluation tests for the blogger-agent using pytest. Ensure all dependencies are installed first.

```bash
uv run pytest eval/test_eval.py
```

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

### Request Product Description

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/personalized-shopping/tests/example_interactions/text_search_floral_dress.session.md

This snippet shows the agent initiating the retrieval of the product's description by clicking the 'Description' button. It's a function call with a 'click' operation.

```json
function call: "click"
args: {
  "button_name": "Description"
}
```

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

### Initiate Product Purchase

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/personalized-shopping/tests/example_interactions/text_search_floral_dress.session.md

This snippet shows the agent's action to initiate the purchase of the selected product. It uses a 'click' function with the 'button_name' argument set to 'Buy Now'.

```javascript
function call: "click"
args: {
  "button_name": "Buy Now"
}
```

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

### Install and Deploy Agent to Vertex AI Agent Engine

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/auto-insurance-agent/README.md

Installs project dependencies with the deployment extra and runs the deployment script to deploy the agent to Vertex AI Agent Engine. The deployment script will output the resource ID and add AGENT_ENGINE_ID to the .env file.

```bash
poetry install --with deployment
python3 deployment/deploy.py
```

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

### Run Agent via ADK Web UI

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/auto-insurance-agent/README.md

Starts the ADK web interface, allowing the agent to be run from a graphical dropdown menu.

```bash
adk web
```

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

### Configure Environment Variables

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/RAG/README.md

Sets up necessary Google Cloud project and location details in a .env file for the script to use.

```bash
GOOGLE_CLOUD_PROJECT=your-project-id
GOOGLE_CLOUD_LOCATION=your-location  # e.g., us-central1
```

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

### Retrieve Product Information

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/personalized-shopping/tests/example_interactions/text_search_floral_dress.session.md

This snippet shows how to retrieve detailed product information after a user interaction. The response includes various attributes such as available colors, sizes, price range, and a textual description.

```JSON
{
  "result": "Back to Search [SEP] < Prev [SEP] color [SEP] black [SEP] blacka [SEP] blackb [SEP] blue [SEP] bluea [SEP] blueb [SEP] coffee [SEP] gray [SEP] green [SEP] hot pink [SEP] navy [SEP] pink [SEP] pinka [SEP] purple [SEP] purplea [SEP] sky blue [SEP] sky bluea [SEP] white [SEP] whitea [SEP] yellow [SEP] yellowa [SEP] size [SEP] small [SEP] medium [SEP] large [SEP] x-large [SEP] xx-large [SEP] Women Summer Cold Shoulder Dress,Beach Vintage Bohemian Print Sundress Party Prom Swing Loose A-Line Dress with Pocket [SEP] Price: $16.29 to $17.59 [SEP] Rating: N.A. [SEP] Description [SEP] Features [SEP] Reviews [SEP] Buy Now"
}
```

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

### Deploy Financial Advisor Agent

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/financial-advisor/README.md

Installs deployment dependencies and deploys the Financial Advisor agent to Vertex AI Agent Engine. It requires Python 3 and Poetry.

```bash
poetry install --with deployment
python3 deployment/deploy.py --create
```

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

### Create and Deploy Agent

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/academic-research/README.md

Initiates the deployment process for the Academic Co-Research agent to Vertex AI Agent Engine. The --create flag starts the deployment, and the output will include the agent's resource ID.

```bash
python3 deployment/deploy.py --create
```

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

### Build Agent with Maven

Source: https://github.com/mbettan/adk-samples/blob/main/java/README.md

Builds a Java agent project using Apache Maven. This command cleans the project and installs dependencies.

```bash
mvn clean install
```

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

### Deploy Agent to Vertex AI

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

Installs deployment dependencies and initiates the agent deployment process to Vertex AI Agent Engine. It requires the resource ID for subsequent operations.

```bash
poetry install --with deployment
python deployment/deploy.py --create
```

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

### Run MCP Toolbox Server

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/software-bug-assistant/README.md

This bash command executes the MCP Toolbox server using a specified tools configuration file. It's used to start the server that enables agent function-calling for databases.

```bash
cd deployment/mcp-toolbox/
./toolbox --tools-file="tools.yaml"
```

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

### Run Programmatic Example

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

Execute the Python script to interact with ADK agents programmatically. This demonstrates the flow of user input to agent responses and how these responses are structured.

```bash
python tests/programmatic_example.py
```

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

### Access Product Features

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/personalized-shopping/tests/example_interactions/text_search_floral_dress.session.md

This snippet demonstrates how to access the 'Features' section of a product. The response provides detailed information about the product's material, occasion suitability, design, and care instructions.

```JSON
{
  "function call": "click",
  "args": {
    "button_name": "Features"
  }
}
```

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

### Memorize start date

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/tests/pre_booking_sample.md

This snippet illustrates the planning_agent using the 'memorize' function to store the trip start date as '2025-05-04'.

```json
{
  "function call": "memorize",
  "args": {
    "key": "start_date",
    "value": "2025-05-04"
  }
}
```

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

### Run Travel Concierge Agent via CLI

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

Starts the Travel Concierge agent locally using the ADK CLI. This allows for direct interaction with the agent through the command line.

```bash
# Under the travel-concierge directory:
adk run travel_concierge
```

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

### Model Configuration Example

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/persona_ad_gen/README.md

This Python code snippet shows how to configure the main agent model. It allows users to change the default model, such as switching to Gemini 2.5 Pro.

```Python
MODEL = "gemini-2.5-flash"  # Change this to use a different model (e.g., Gemini 2.5 Pro)
```

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

### Deploy Image Scoring Agent

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/image-scoring/README.md

Deploys the Image Scoring Agent to Vertex AI Agent Engine. This command installs dependencies and runs the deployment script to create the agent.

```Bash
poetry install --with deployment
poetry run python3 deployment/deploy.py --create
```

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

### Run CaMeLAgent via ADK Web Interface

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/camel/README.md

Starts the CaMeLAgent with a web interface using the ADK command-line tool, providing an alternative way to interact with the agent.

```bash
poetry run adk web
```

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

### Initialize Interactive Blogger Agent

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/blog-writer/agent_explainer.md

Initializes the `interactive_blogger_agent` with its name, model, description, and a detailed instruction set for technical blogging. It includes a workflow for analyzing code, planning, refining, handling visuals, writing, editing, social media promotion, and exporting the blog post. It also specifies sub-agents and tools like `save_blog_post_to_file` and `analyze_codebase`.

```Python
interactive_blogger_agent = Agent(
    name="interactive_blogger_agent",
    model=config.worker_model,
    description="The primary technical blogging assistant. It collaborates with the user to create a blog post.",
    instruction=f"""
    You are a technical blogging assistant. Your primary function is to help users create technical blog posts.

    Your workflow is as follows:
    1.  **Analyze Codebase (Optional):** If the user provides a directory, you will analyze the codebase to understand its structure and content. To do this, use the `analyze_codebase` tool.
    2.  **Plan:** You will generate a blog post outline and present it to the user. To do this, use the `robust_blog_planner` tool.
    3.  **Refine:** The user can provide feedback to refine the outline. You will continue to refine the outline until it is approved by the user.
    4.  **Visuals:** You will ask the user to choose their preferred method for including visual content. You have two options for including visual content in your blog post:

    1.  **Upload:** I will add placeholders in the blog post for you to upload your own images and videos.
    2.  **None:** I will not include any images or videos in the blog post.

    Please respond with "1" or "2" to indicate your choice.
    5.  **Write:** Once the user approves the outline, you will write the blog post. To do this, use the `robust_blog_writer` tool. Be then open for feedback.
    6.  **Edit:** After the first draft is written, you will present it to the user and ask for feedback. You will then revise the blog post based on the feedback. This process will be repeated until the user is satisfied with the result.
    7.  **Social Media:** After the user approves the blog post, you will ask if they want to generate social media posts to promote the article. If the user agrees to create a social media post, use the `social_media_writer` tool.
    8.  **Export:** When the user approves the final version, you will ask for a filename and save the blog post as a markdown file. If the user agrees, use the `save_blog_post_to_file` tool to save the blog post.

    Current date: {datetime.datetime.now().strftime("%Y-%m-%d")} 
    """,
    sub_agents=[
        robust_blog_writer,
        robust_blog_planner,
        blog_editor,
        social_media_writer,
    ],
    tools=[
        FunctionTool(save_blog_post_to_file),
        FunctionTool(analyze_codebase),
    ],
    output_key="blog_outline",
)
```

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

### Test Deployed Agent

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/RAG/README.md

This script connects to the deployed agent on Vertex AI Agent Engine, sends test queries, and displays the agent's responses. Example queries related to Alphabet's 10-K report are included.

```Python
python deployment/run.py
```

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

### Execute Programmatic Example Client

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/travel-concierge/README.md

Runs a Python script that demonstrates programmatic interaction with the Travel Concierge agent's API server. This script simulates two turns of conversation with the agent.

```bash
python tests/programmatic_example.py
```

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

### Run Agent Locally via ADK Dev UI (Bash)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/README.md

Starts the ADK Development UI for an agent, allowing interaction through a web interface. This command is run from the agent's main directory.

```bash
# Example (check agent's README for exact path)
cd agents/fomc-research/
adk web
```

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

### Configure Poetry Keyring (Linux)

Source: https://github.com/mbettan/adk-samples/blob/main/python/agents/auto-insurance-agent/README.md

Disables the keyring integration for Poetry on Linux systems if an error occurs during installation.

```bash
poetry config keyring.enabled false
```