darkarnium / kako Goto Github PK

In order to ensure that Kako continues to work, and falls in line with best practices, it should work with Python 3.

Fix.

Currently Kako has no test suites. This is awful and needs to be fixed. Appropriate unit and integration tests should be implemented with a Tox file provided for validation.

Perhaps investigate Travis for merge to master testing for candidates to be published to PyPi (master).

In order to track the output and status from Kako simulations, all capture() methods should generate a standard JSON output format. This format should look something like the below:

{
  "source_ip": "1.1.1.1",
  "source_port": "45302",
  "simulation": "mirai_cpeserver_soap",
  "capture": {
    ...DATA...
  }
}

Once generated, this JSON should be submitted into a Kako results queue in AWS SQS for processing / ingestion.

In order to more easily create simulations for common services, it should be possible to define simulations from YAML. This should allow for definition of banners and expected command inputs and outputs.

The definitions could look something like the following:

---
simulation:
  port: 7547
  listener: 'HTTP'
  interpreter: ''
  routes:
    - 'GET':
      - path: '/'
         response_code: 403
         response_body: 'NOPE'
    - 'POST':
      - path: '/UD/act'
         response_code: 200
         response_body: ''

In order to allow for rapid deployment of new simulations, simulations should not live in the Kako project and should be published in an external repository.

This also allows for easier use of private simulations, which may be required in certain environments.

Kako should be published to PyPi to simplify deployment.

In order to remove requirement for AWS, Kako should allow capture data to be written to file on the honeypot in JSON format. This should be configurable, and allow for SNS or JSON file output inside of the configuration.

This can then be consumed by some other process to push into the relevant collection system.

Presently the classification is unknown. It will be better if we tag the vulnerability with CVE id CVE-2017-17215.

"destination_ip": "TODO",
"destination_port": 37215,
"node": "ip-172-31-44-80",
"simulation_name": "Huawei uPnP Simulation",
"simulation_version": "0.1.0",
"source_ip": "59.6.101.97",
"source_port": 4858,
"timestamp": 1515413257,
"vulnerability": "UNKNOWN"
}