IntentKit Setup: Docker and Local Development

​

Docker (Recommended)

Docker is the fastest and easiest way to set up IntentKit with minimal configuration. It creates a containerized environment that isolates dependencies and configurations.

1. Create a new directory:

   mkdir intentkit && cd intentkit
   

2. Download the required files:

   # Download docker-compose.yml
   curl -O https://raw.githubusercontent.com/crestalnetwork/intentkit/main/docker-compose.yml
   
   # Download example environment file
   curl -O https://raw.githubusercontent.com/crestalnetwork/intentkit/main/example.env
   

3. Set up your environment:

   # Rename example.env to .env
   mv example.env .env
   
   # Edit the .env file to add your configuration
   nano .env
   

   Ensure you add your OPENAI_API_KEY in the .env file. This is required for the AI capabilities of IntentKit. Full guide on how to generate an OpenAI API key can be found here.

4. Start the services:

   docker compose up
   

   This will launch all necessary services, including the API server and autonomous agent scheduler.

5. After installation, you’ll want to create your first agent. Head over to our Agent Creation and Configuration guide to learn how to create and configure agents.

​

Local Development

For developers who want to customize or extend IntentKit, setting up a local development environment provides more control and flexibility.

1. Clone the repository:

   git clone https://github.com/crestalnetwork/intentkit.git
   cd intentkit
   

2. Install dependencies: Ensure you have Python 3.10-3.12 installed (3.12 is recommended). If you don’t have poetry installed, do so first:

   pip install poetry
   

   Then, install the project dependencies:

   poetry install --with dev
   poetry shell
   

3. Set up your environment:

   cp example.env .env
   nano .env
   

   Configure the .env file with your settings, including the required OPENAI_API_KEY. Full guide on how to generate an OpenAI API key can be found here.

4. Run the application:

   • Start the API server:

     uvicorn app.entrypoints.api:app --reload
     

   • Start the autonomous agent scheduler:

     python -m app.entrypoints.autonomous
     

   Both services are now running locally and ready for use.

5. Creating Your First Agent

After installation, you’ll want to create your first agent. Head over to our Agent Creation and Configuration guide to learn how to create and configure agents.