A Repository of curated datasets from various attacks to:

• Easily develop detections without having to build an environment from scratch or simulate an attack.
• Test detections, specifically Splunks Security Content
• Replay/inject into streaming pipelines for validating your detections in your production SIEM

datasets are defined by a common yml structure the structure has the following fields:

For example

author: Patrick Bareiss
date: '2020-10-08'
description: 'Atomic Test Results: Successful Execution of test T1003.002-1 Registry
  dump of SAM, creds, and secrets Return value unclear for test T1003.002-2 Registry
  parse with pypykatz Successful Execution of test T1003.002-3 esentutl.exe SAM copy '
environment: attack_range
technique:
- T1003.002
dataset:
- https://attack-range-attack-data.s3-us-west-2.amazonaws.com/T1003.002/attack-range-windows-domain-controller.json
- https://attack-range-attack-data.s3-us-west-2.amazonaws.com/T1003.002/windows-powershell.log
- https://attack-range-attack-data.s3-us-west-2.amazonaws.com/T1003.002/windows-security.log
- https://attack-range-attack-data.s3-us-west-2.amazonaws.com/T1003.002/windows-security_ssa.log
- https://attack-range-attack-data.s3-us-west-2.amazonaws.com/T1003.002/windows-sysmon.log
- https://attack-range-attack-data.s3-us-west-2.amazonaws.com/T1003.002/windows-system.log
references: 
  - https://attack.mitre.org/techniques/T1003/002/
  - https://github.com/redcanaryco/atomic-red-team/blob/master/atomics/T1003.002/T1003.002.md
  - https://github.com/splunk/security-content/blob/develop/tests/T1003_002.yml
sourcetypes: 
  - XmlWinEventLog:Microsoft-Windows-Sysmon/Operational
  - WinEventLog:Microsoft-Windows-PowerShell/Operational
  - WinEventLog:System
  - WinEventLog:Security

typically datasets generated by the attack_range dump features will start with a folder name of the MITRE ATT&CK technique, and seen by the default filename: dataset.yml otherwise a custom folder and file name can be used.

Environments are a description of where the dataset was collected. At this moment there are no specific restrictions, although we do have a simple template a user can start with here. The most common environment for most datasets will be the attack_range since this is the tool that used to generate attack data sets automatically.

Most datasets generated will be of type JSON. There are two main simple ways to ingest it.

0. Download dataset
1. In splunk enterprise , add data -> Files & Directories -> select dataset
2. Set the sourcetype to JSON
3. Set SHOULD_LINEMERGE to false
4. Explore your data

See a quick demo 📺 of this process here.

To send datasets into DSP the simplest way is to use the scloud command-line-tool as a requirement.

1. Download the dataset
2. Ingest the dataset into DSP via scloud command cat attack_data.json | scloud ingest post-events --format json
3. Build a pipeline that reads from firehose and you should seeing the events.

1. Generate a dataset
2. Upload dataset to s3 bucket or similar hosting service
3. Make PR with dataset .yml file under the corresponding MITRE ATT&CK technique folder.

Note the simplest way to generate a dataset to contribute is to launch your simulations in the attack_range, or manually attack the machines and when done dump the data using the dump function.

See a quick demo 📺 of the process to dump a dataset here.

To contribute a dataset simple create a PR on this repository, for general instructions on create a PR see this guide.

This project takes advantage of automation to generate datasets using the attack_range. You can see details about this service on this sub-project folder attack_data_service.

Copyright 2020 Splunk Inc.

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.