This article explains how to set up an environment to make the most of the Datadog API and includes how to pull or push events, metrics, and monitors from Datadog’s public API to a local file.

Prerequisite: Python and pip installed on your localhost. Windows users see Installing Python 2 on Windows.

1. Open a terminal.

2. Verify the directory: pwd on macOS, dir on Windows.

3. Create a new folder: mkdir <NAME_OF_THE_FOLDER>.

4. Enter the folder: cd <NAME_OF_THE_FOLDER>.

5. Download the script api_query_data.py to the folder created in step 3 and edit it:

   a. Replace <YOUR_DD_API_KEY> and <YOUR_DD_APP_KEY> with your Datadog API and app keys.

   b. Replace system.cpu.idle with a metric you want to fetch. A list of your metrics is displayed in the Datadog Metric Summary.

   c. Optionally, replace * with a host to filter the data. A list of your hosts is displayed in the Datadog Infrastructure List.

   d. Optionally, change the time period to collect the data. The current setting is 3600 seconds (one hour). Note: If you run this too aggressively, you may reach the Datadog API limits.

   e. Save your file and confirm its location.

Once the above is complete:

1. It is best practice to create a virtual environment to install Python packages into. One virtual environment manager is virtualenv.
2. Create a new virtual environment in the directory you created earlier by running virtualenv venv.
3. Activate the environment by running source venv/bin/activate (Mac/Linux) or > \path\to\env\Scripts\activate (Windows).
4. Run pip install datadog to install the Datadog API package. This enables the Python file to interact with the Datadog API.
5. In the terminal, run the script: python api_query_data.py.

If successful, your data displays in the terminal and a file is created in your folder named out.txt.

See additional examples in the Datadog API documentation.