> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/gnosis/prediction-market-agent/llms.txt
> Use this file to discover all available pages before exploring further.

# Installation

> Install Gnosis Prediction Market Agent using Poetry and Python 3.11+

## Prerequisites

Before installing, ensure you have:

* **Python 3.11 or higher** - The project requires Python \~3.11.0
* **pip** - Python package installer
* **Git** - For cloning the repository

<Note>
  The project uses Poetry for dependency management, which provides reproducible builds and better dependency resolution than pip alone.
</Note>

## Installation steps

<Steps>
  <Step title="Clone the repository">
    First, clone the Gnosis Prediction Market Agent repository:

    ```bash theme={null}
    git clone https://github.com/gnosis/prediction-market-agent.git
    cd prediction-market-agent
    ```
  </Step>

  <Step title="Install Poetry">
    Install Poetry using Python 3.11's pip:

    ```bash theme={null}
    python3.11 -m pip install poetry
    ```

    <Note>
      If `python3.11` is not found, you may need to install Python 3.11 first using your system's package manager or from [python.org](https://www.python.org/downloads/).
    </Note>
  </Step>

  <Step title="Install project dependencies">
    Use Poetry to install all project dependencies:

    ```bash theme={null}
    python3.11 -m poetry install
    ```

    This will:

    * Create a virtual environment
    * Install all required dependencies from `pyproject.toml`
    * Set up the project for development

    <Warning>
      The installation may take several minutes as the project has many dependencies including LangChain, OpenAI, web3, and various AI frameworks.
    </Warning>
  </Step>

  <Step title="Activate the Poetry shell">
    Activate the Poetry-managed virtual environment:

    ```bash theme={null}
    python3.11 -m poetry shell
    ```

    You should see your terminal prompt change to indicate you're in the Poetry environment.
  </Step>

  <Step title="Verify installation">
    Verify the installation by checking the available commands:

    ```bash theme={null}
    python prediction_market_agent/run_agent.py --help
    ```

    You should see the help output with available agents and market types.
  </Step>
</Steps>

## Key dependencies

The project includes these major dependencies:

<AccordionGroup>
  <Accordion title="AI and LLM frameworks">
    * **OpenAI** (^1.91.0) - GPT models for agent reasoning
    * **LangChain** (^1.0.0) - Framework for building LLM applications
    * **CrewAI** (^1.0.0) - Multi-agent orchestration
    * **Pydantic AI** (^1.0.0) - Type-safe AI applications
    * **Autogen** (^0.4.0) - Multi-agent conversations
  </Accordion>

  <Accordion title="Prediction market tooling">
    * **prediction-market-agent-tooling** (^0.69.24) - Core APIs for interacting with prediction markets
    * **prediction-prophet** - Forecasting and research tools
    * **web3** (version 6-8) - Ethereum blockchain interaction
  </Accordion>

  <Accordion title="Web scraping and search">
    * **Tavily Python** (^0.5.0) - Web search capabilities
    * **BeautifulSoup4** (version 4.x) - HTML parsing
    * **DuckDuckGo Search** - Privacy-focused search
    * **Google Search Results** - Google search integration
  </Accordion>

  <Accordion title="Development tools">
    * **Streamlit** (1.49.1) - Interactive web apps
    * **Typer** (0.9.x) - CLI framework
    * **pytest** - Testing framework
    * **mypy** (^1.8.0) - Static type checking
    * **black** (^23.12.1) - Code formatting
  </Accordion>
</AccordionGroup>

## Development dependencies

If you're contributing to the project, Poetry will also install development dependencies including:

* **Jupyter** and **JupyterLab** - For notebook-based development
* **ape-foundry** - Local blockchain testing
* **freezegun** - Time mocking for tests
* **matplotlib** and **seaborn** - Data visualization

## Next steps

<CardGroup cols={2}>
  <Card title="Configuration" icon="gear" href="/configuration">
    Set up your environment variables and API keys
  </Card>

  <Card title="Quickstart" icon="rocket" href="/quickstart">
    Run your first agent with the coinflip example
  </Card>
</CardGroup>