cumulus

Cumulus is a Python script for executing cryptocurrency purchases through the Coinbase Advanced Trade API. Users specify their desired crypto allocations in a YAML configuration file, and the script executes the purchases according to these preferences. When combined with a task scheduler (like cron), this script can be used to implement a dollar-cost averaging (DCA) strategy.

Why Use Coinbase's Advanced Trade?

Trading through Advanced Trade instead of the traditional Coinbase UI results in significant fee savings.

Assume a $10 weekly investment over a year:

Coinbase UI: $51.48 in fees/year (9.9% per transaction)

Advanced Trade: $6.24 in fees/year (1.2% per transaction)

While the Coinbase UI provides convenience with built-in DCA features, using this script with Advanced Trade can reduce fees by approximately 87.9%. Over a year, this savings on a $10 weekly investment is equivalent to getting 5 extra weeks of investing for free!

Notes

1. All fees and charges are subject to change, and it is recommended to review the most up-to-date fee information on the Coinbase platform.
2. The previous calculations assume you don't have a Coinbase One subscription.

Getting started

Create an API Key

To use the Advanced Trade APIs, you must create an API key. Note that Coinbase Pro keys are no longer supported. Follow the instructions in the Getting Started with Advanced Trade APIs guide.

Ensure that you have the "Trade" permission checked for your API key.

1. After setting the permissions, click the Create & download button to generate your API key. This will download a file named cdp_api_key.json.

2. Place the cdp_api_key.json file in the secrets directory of your project.

Setting Up Python and Virtual Environment

1. Navigate to your project directory (e.g., /home/<user>/cumulus).

2. Create a virtual environment by running:

   python -m venv venv

   This will create a folder named venv, which is ignored by git.

3. Activate the virtual environment:

   • On macOS/Linux:

     source venv/bin/activate

   • On Windows:

     venv\Scripts\activate

4. Install the required dependencies within the virtual environment:

   pip install -r requirements.txt

Edit the Trading Plan

1. Create a trading plan file named trading_plan.yml in your project directory. You can use the provided template trading_plan.yml.template as a starting point.

2. Specify the amount in USD to spend on each cryptocurrency in the trades section of the trading_plan.yml file. For example:

   trades:
     ETH: $10.00
     BTC: $10.00
     SOL: $10.00 

Note: YAML is picky about spacing, so ensure your indentation stays consistent.

For assets available on Coinbase Exchange, please go here.

Running Cumulus

Cumulus is now installed, configured, and ready to run. By default, the script runs in sandbox mode. To execute it in production mode, use the -p or --production flag. You can execute the script manually by entering:

./cumulus.py

or

python cumulus.py

To run in production mode:

./cumulus.py -p

or

python cumulus.py --production

Setting Up a Cron Job

To automate the execution of Cumulus, set up a cron job using crontab -e to run the script at your desired interval. Remember that the script runs in sandbox mode by default, so if you want it to run in production mode, include the -p or --production flag in your cron job command. You can use this online cron expression generator for assistance.

Examples

• To run the script every Friday at 6:00 AM server time in production mode:

  0 6 * * 5 /path/to/cumulus/venv/bin/python /path/to/cumulus/cumulus.py -p

• To run the script every minute in sandbox mode:

  * * * * * /path/to/cumulus/venv/bin/python /path/to/cumulus/cumulus.py

• To run the script every minute in production mode:

  * * * * * /path/to/cumulus/venv/bin/python /path/to/cumulus/cumulus.py -p

Profit!

Special Thanks

Thank you to Frank Pelosi for shedding light on Coinbase's high fees, the savings possible with Advanced Trade, and the idea of using the API for automation.