Awesome
Caracara
<!-- ![PyPI - Status](https://img.shields.io/pypi/status/caracara) [![Pylint](https://github.com/CrowdStrike/caracara/actions/workflows/pylint.yml/badge.svg)](https://github.com/CrowdStrike/caracara/actions/workflows/pylint.yml) [![Flake8](https://github.com/CrowdStrike/caracara/actions/workflows/flake8.yml/badge.svg)](https://github.com/CrowdStrike/caracara/actions/workflows/flake8.yml) [![Bandit](https://github.com/CrowdStrike/caracara/actions/workflows/bandit.yml/badge.svg)](https://github.com/CrowdStrike/caracara/actions/workflows/bandit.yml) [![CodeQL](https://github.com/CrowdStrike/caracara/actions/workflows/codeql.yml/badge.svg)](https://github.com/CrowdStrike/caracara/actions/workflows/codeql.yml) -->A friendly wrapper to help you interact with the CrowdStrike Falcon API. Less code, less fuss, better performance, and full interoperability with FalconPy.
Features
A few of the developer experience enhancements provided by the Caracara toolkit include:
Feature | Details |
---|---|
Automatic pagination with concurrency | Caracara will handle all request pagination for you, so you do not have to think about things like batch sizes, batch tokens or parallelisation. Caracara will also multithread batch data retrieval requests where possible, dramatically reducing data retrieval times for large datasets such as host lists. |
Friendly to your IDE (and you!) | Caracara is written with full support for IDE autocomplete in mind. We have tested autocomplete in Visual Studio Code and PyCharm, and will accept issues and patches for more IDE support where needed. Furthermore, all code, where possible, is written with type hints so you can be confident in parameters and return values. |
Logging | Caracara is built with the in-box logging library provided with Python 3. Simply set up your logging handlers in your main code file, and Caracara will forward over debug , info and error logs as they are produced. Note that the debug logs are very verbose, and we recommend writing these outputs to a file as opposed to the console when retrieving large amounts of lightly filtered data. |
Real Time Response (RTR) batch session abstraction | Caracara provides a rich interface to RTR session batching, allowing you to connect to as many hosts as possible. Want to download a specific file from every system in your Falcon tenant? Caracara will even extract it from the .7z container for you. |
Rich and detailed sample code | Every module of Caracara comes bundled with executable, fully configurable code samples that address frequent use cases. All samples are built around a common structure allowing for code reuse and easy reading. Just add your API credentials to config.yml , and all samples will be ready to go. |
Simple filter syntax | Caracara provides an object-orientated Falcon Query Language (FQL) generator. The FalconFilter object lets you specify filters such as Hostname , OS and Role , automatically converting them to valid FQL. Never write a FQL filter yourself again! |
Single authentication point of entry | Authenticate once and have access to every module. |
100% FalconPy compatibility | Caracara is built on FalconPy, and can even be configured with a FalconPy OAuth2 object via the auth_object constructor parameter, allowing you to reuse FalconPy authentication objects across Caracara and FalconPy. Authenticate once with FalconPy, and access every feature of FalconPy and Caracara. |
Installation Instructions
Caracara supports all major Python packaging solutions. Instructions for Poetry and Pip are provided below.
Caracara supports Python versions that are still supported by the Python Software Foundation, i.e., Python 3.8 and up.
<details> <summary><h3>Installing Caracara from PyPI using Poetry (Recommended!)</h3></summary>Poetry: Installation
poetry add caracara
Poetry: Upgrading
poetry update caracara
Poetry: Removal
poetry remove caracara
</details>
<details>
<summary><h3>Installing Caracara from PyPI using Pip</h3></summary>
Pip: Installation
python3 -m pip install caracara
Pip: Upgrading
python3 -m pip install caracara --upgrade
Pip: Removal
python3 -m pip uninstall caracara
</details>
Basic Usage Examples
"""List Windows devices.
This example will use the API credentials provided as keywords to list the
IDs and hostnames of all systems within your Falcon tenant that run Windows.
"""
from caracara import Client
client = Client(
client_id="12345abcde",
client_secret="67890fghij",
)
filters = client.FalconFilter()
filters.create_new_filter("OS", "Windows")
response_data = client.hosts.describe_devices(filters)
print(f"Found {len(response_data)} devices running Windows")
for device_id, device_data in response_data.items():
hostname = device_data.get("hostname", "Unknown Hostname")
print(f"{device_id} - {hostname}")
You can also leverage the built in context manager and environment variables.
"""List stale sensors.
This example will use the API credentials set in the environment to list the
hostnames and IDs of all systems within your Falcon tenant that have not checked
into your CrowdStrike tenant within the past 7 days.
This is determined based on the filter LastSeen less than or equal (LTE) to 7 days ago (-7d).
"""
from caracara import Client
with Client(client_id="${CLIENT_ID_ENV_VARIABLE}", client_secret="${CLIENT_SECRET_ENV_VARIABLE}") as client:
filters = client.FalconFilter()
filters.create_new_filter("LastSeen", "-7d", "LTE")
response_data = client.hosts.describe_devices(filters)
print(f"Found {len(response_data)} stale devices")
for device_id, device_data in response_data.items():
hostname = device_data.get("hostname", "Unknown Hostname")
print(f"{device_id} - {hostname}")
Examples Collection
Each API wrapper is provided alongside example code. Cloning or downloading/extracting this repository allows you to execute examples directly.
Using the examples collection requires that you install our Python packaging tool of choice, Poetry. Please refer to the Poetry project's installation guide if you do not yet have Poetry installed.
Once Poetry is installed, make sure you run poetry install
within the root repository folder to set up the Python virtual environment.
To configure the examples, first copy examples/config.example.yml
to examples/config.yml
. Then, add your API credentials and example-specific settings to examples/config.yml
. Once you have set up profiles for each Falcon tenant you want to test with, execute examples using one of the two options below.
Executing the Examples
There are two ways to use Poetry to execute the examples.
<details> <summary><h4>Executing from a Poetry Shell</h4></summary>The poetry shell
command will enter you into the virtual environment. All future commands will run within the Caracara virtual environment using Python 3, until you run the deactivate
command.
poetry shell
examples/get_devices/list_windows_devices.py
</details>
<details>
<summary><h4>Executing without Activating the Virtual Environment</h4></summary>
If you do not want to enter the Caracara virtual environment (e.g., because you are using your system's installation of Python for other purposes), you can use the poetry run
command to temporarily invoke the virtual environment for one-off commands.
poetry run examples/get_devices/list_windows_devices.py
All examples are also configured in the pyproject.toml
file as scripts, allowing them to be executed simply.
poetry run stale-sensors
</details>To get a complete list of available examples, execute the command
util/list-examples.sh
from the root of the repository folder.
Documentation
Coming soon!
Contributing
Interested in taking part in the development of the Caracara project? Start here.
Why Caracara?
Simple! We like birds at CrowdStrike, so what better bird to name a Python project after one that eats just about anything, including snakes :)