Re: [cti] Common repository for STIX CTI objects

[Andrew Storms <storms@newcontext.com>]

Hi Rich.

I have a question of clarification.Â

2 potential workflows:

1. As an analyst, would they go to this repository, find what they need.ÂThen copy and paste into their piece of threat intel they are creating in STIX. And perhaps edit things slightly for their specific case? Essentially a repo of templates for use.

2. Or instead would they find what they want, grab the URL to the object in the repo. Then instead of actually putting the content in their threat intel, they would then only paste the URL to the object in the repo?

Hopefully my questions make sense.

--A

On Mon, Jul 6, 2020 at 7:59 AM Rich Piazza <rpiazza@mitre.org> wrote:

Many entities in cyber threat intelligence are common and having many duplicate STIX objects to represent the same concept has always been seen as wasteful and problematic. Several decisions made when writing the STIX specification tried to take this into account. This includes: specification defined instances of TLP data markings, kill chain phases referred to by their names and deterministic identifiers for STIX cyber objects (SCOs). However, having an easily available repository of common CTI objects has always been on the âwish listâ of members of the CTI-TC. ÂDHS has tasked MITRE with investigating creating such a repository.ÂÂ

Â

MITRE has already started a similar repository of STIX objects to represent both the ATT&CK and CAPEC frameworks. It is available atÂhttps://github.com/mitre/cti. It certainly is the case that other organizations might want to do something similar with their cyber threat intellectual property. However, there are other STIX objects that are general enough to be hosted in a common repository, defined once and re-used by the broader STIX community. Such a repository would foster consistency across STIX threat sharing efforts. In creating and using this repository, the amount of data transmitted over the wire could be reduced because only identifier references would need to be shared.

ÂÂ

Some of the objects types that immediately come to mind to include in this repository are locations (e.g., countries) and identities (e.g., industry sectors). Others types, like software (e.g., Microsoft Word â version x) or tools (e.g., RDP) might be useful. Objects like IP addresses, which already can be considered unique if using deterministic identifiers, could be âstoredâ in this repository, so they need not be shared. Vulnerability objects representing each CVE could be housed there also. Iâm sure there are other objects that could be included.

Â

There are many issues to be discussed as part of setting up such a repository:

Â

• Where would it be hosted? It is envisioned that it could be available on the GitHub oasis-open web site, however this is just an initial suggestion.
• How is the content stored? Is it âfrontedâ by a TAXII server?
• Who maintains it? MITRE, DHS, OASISâ?
• Who decides what should be in the repository? The maintainers, a CTI subcommittee?
• Would the STIX community actually use this repository?

Â

If you would like to be involved in making this happen, or just have some ideas, please get in touch.

We can have a kickoff discussion at a future working call.

Â

ÂÂÂÂÂÂÂÂÂÂÂ Rich P.

--Â

Rich Piazza

Lead Cyber Security Engineer

The MITRE Corporation

781-271-3760

Â

Â

Â

Â

--

Andrew Storms VP of Security Services PÂÂ707-477-4335 ÂÂ