Cortex, a free, open source software allows security analysts and threat hunters to analyze and enrich observables (IP addresses, hashes, domains, …) collected in the course of an investigation or received from third parties, for example through MISP, the de facto standard for threat sharing.
On March 29, 2018, we released Cortex 2, a major improvement over the previous version which brought, among other cool features, authentication, caching, multi-tenancy (RBAC) and rate limiting. Instead of deploying several Cortex 1 instances behind reverse proxies which would implement authentification, administrators can deploy a single Cortex 2, create multiple organizations and serve the needs of various information security populations while enjoying extra features.
On May 31, 2018, we published a brand new API guide so that developers can take advantage of the powerful REST API of the product. Sadly, Cortex4py, the FOSS Python library we provide to interact with the API was not compatible with Cortex 2. Until today.
Thanks to the hard work of our dear Nabil Adouani, we are happy to announce the immediate availability of Cortex4py 2.0.0, a complete rewrite of the library in Python 3. Cortex4py 2.0.0 is fully compatible with Cortex 2. However, it doesn’t work with Cortex 1.
While TheHive, the highly popular free and open source Security Incident Response Platform (SIRP) we develop has native support for many Cortex 2 instances, Python developers can leverage Cortex4py to interact with Cortex 2, manage organizations, users, analyzer configurations and analyze observables at scale from alternative SIRPs, SIEMs or custom scripts thanks to the 83 analyzers Cortex 2 has as of June 18, 2018.
To install Cortex4py, use PIP3:
$ sudo -H pip3 install cortex4py
If you are using Python on a Windows operating system, please forgo the sudo command.
Cortex4py 2 comes with a usage guide which includes many examples. For example, if you want to fetch the last 10 successful jobs that have been executed against domain names and display the result summaries of those 10 jobs you could write something like:
Migrating from Cortex4py 1
If you have already written scripts using Cortex4py 1.x (for Cortex 1), we tried to keep the already available methods. However, we recommend you adapt your code to leverage the new Cortex4py 2 classes and methods as soon as feasible. Moreover, the existing scripts must be updated to support authentication if you intend to use them with Cortex 2. Please read the Cortex4py 2 usage guide for more information.
Cortex 2.0.0 is brand new software. As such, it might contain bugs and limitations. If you find any or encounter problems, please ask on our user forum, contact us on Gitter, or send us an email at email@example.com. We are here to help.
On June 6, 2018, we released Cortex-Analyzers 1.10, which contained 11 new analyzers, bringing the total to 83 programs. You read that correctly: 83 ways to assess and gain insight on observables collected during the course of an investigation or while performing threat intelligence thanks to Cortex, our free & open source analysis engine. One day after, we published version 1.10.1 which fixed the name of the Anomali STAXX reports for TheHive. Since then we uncovered a few additional issues which version 1.10.2 corrects:
TheHive Project Chefs are pleased to announce the immediate availability of Cortex-Analyzers 1.10.1. To install this new release and benefit from 11 new analyzers and some fixes:
$ cd /path/to/Cortex-Analyzers
$ sudo git pull
$ for I in analyzers/*/requirements.txt; do sudo -H pip2 install -r $I; done && \
for I in analyzers/*/requirements.txt; do sudo -H pip3 install -r $I || true; done
After running these commands, read the Analyzer Requirements Guide, log into the Cortex 2 Web UI as an orgAdmin, click on the Refresh Analyzers button in the Cortex Web UI, configure the new analyzers and enjoy!
This analyzer comes in only one flavor. You need a valid FireEye iSIGHT Threat Intelligence subscription to use the analyzer. Retrieve the API key associated with your account and provide it as a value to the key parameter. Obtain the password associated with the API key and provide it as a value to the pwd parameter.
Determine whether an IP has known scanning activity using GreyNoise.
This analyzer comes in only one flavor. It can be used out of the box without configuration. However, if you make many requests, you need to obtain an API key. Please contact GreyNoise to ask for one. Once you get the API key, provide it as the value of the key parameter.
Query domains, IPs, hashes and URLs against IBM X-Force Threat Intelligence sharing platform.
This analyzer comes in only one flavor. Access to IBM X-Force Threat Exchange requires an IBM ID. Once you have access to the service, supply the URL of the service as value for the url parameter, the API key associated with your account as value for the key parameter and the associated password as the value of the pwd parameter.
Query Malwares.com and get reports on files, hashes, domain names and IP addresses.
The analyzer comes in two flavors:
– Malwares_pDNS_GetReport: get the latest Malwares report for a file,
hash, domain or an IP address.
– Malwares_pDNS_Scan: scan a file or URL.
You need to sign up for a Malwares.com account. An API key to use the service’s API should be associated with your account. Supply it as the value of the key parameter.
Query IP addresses and domain names against Mnemonic Passive DNS service.
This analyzer comes in two flavors:
Mnemonic_pDNS_Public: query Mnemonic’s public service.
When using the public service, the analyzer can be used out of the box with no further configuration. When using the closed service, you need to contact Mnemonic to get an API key which you’ll need to supply as the value of the key parameter.
This analyzer comes in only one flavor. No configuration is needed. It can be used out of the box.
Follow redirects of shortened URLs to reveal the real ones.
This analyzer comes in only one flavor.
No configuration is required. It can be used out of the box.
Warning: using this analyzer without extra caution might lead to unexpected consequences. For example, if the URL you are seeking to unshorten is an attacker-controlled one, you may end up leaving undesired traces in the threat actor’s infrastructure logs. The TLP values Cortex allows you to configure to prevent the use of an analyzer if the TLP associated with an observable is above the authorized level won’t be of much help since Unshortenlink have to access the shortened URL. Please do not activate this analyzer unless you (and your fellow analysts) know what they are doing.
YARA analyzer had a bug which was fixed in version 1.9.7 of the Cortex-analyzers repository. If you install 1.10, you’ll obviously benefit from the fix 😉
A permission problem that prevented using the Cuckoo Sandbox analyzer was corrected (thanks Felix Bauer!)
Correction: June 6, 2018 An earlier version of this post mentioned that GreyNoise needs an API key. That’s only necessary if a certain level of requests are made. Also, to configure analyzers, you need to be orgAdmin.
After announcing Cortex 2.0.0 and TheHive 3.0.7, the first version of your favorite SIRP that is (supposedly) compatible with the brand-new version of Cortex, last week, we thought it was time to relax and enjoy the upcoming, long Easter weekend, the sunny sky of Paris (if you can pierce the veil of the Forever Grey Cloud™ that is hanging over the city of lights), and great jazz music. Heck, I even tweeted about it … only to be proven wrong by Life (and Murphy).
We literally field tested Cortex 2 for 3 weeks, we squashed bugs here and there, until almost the very last minute before the release. And yet, our QA needs to be improved by leaps and bounds as we had to release Cortex 2.0.1 one day after unveiling 2.0.0 to correct some additional bugs. And then some members of the core team and of our growing user community took it for a spin. And all hell broke lose. Well, almost 🙂
Session collisions (when TheHive and Cortex 2 are used on the same machine), analyzer malfunctions, connectivity problems … issues that were not identified during the testing phase, even in a production environment, where everything worked as expected. And we call this ‘Computer Science’. Right, right…
So we worked hard, took out our Code Hammer (it’s like Thor’s but cyber) and blasted away all the bugs that we found out or that were reported to us (arigato gozaimasu!) and we are happy to announce the immediate availability of Cortex 2.0.2, TheHive 3.0.8, Cortexutils 1.2.3 and Cortex-Analyzers 1.9.2.
TL;DR Install or upgrade Cortex 2.0.2, update Cortexutils, git pull the Cortex-analyzers repo to get the latest version of the repository, upgrade to TheHive 3.0.8, follow the Quick Start Guide and have a drink.
If you have time (which is admittedly quite scarce nowadays), please read on the changelogs:
As stated in the previous post, we will release a new version of Cortex4py in order to make it compatible with Cortex 2, continue the work we started with our MISP Project friends to support MISP attribute enrichment through Cortex 2 (MISP currently only supports enrichment using Cortex 1), and perform a long-overdue overhaul of our documentation. We will also release a brand new version of TheHive4py.
Last but not least, we’ll take a hard look at ourselves and our QA. You expect us from us high quality and we hold ourselves to high standards. And we will deliver.
Something does not work as expected? You have troubles installing or upgrading? Spotted new bugs? No worries, please open issues on GitHub or comment on existing ones, join our user forum, contact us on Gitter, or send us an email at firstname.lastname@example.org. We are here to help.
Update: Cortex 2.0.1 was released since this post went live. It corrects a few bugs we uncovered in 2.0.0 as described in the changelog. Please install Cortex 2.0.1 instead of 2.0.0.
Cortex 2 supports all the authentication methods that TheHive supports: LDAP, Active Directory, local accounts, API Keys, and X.509 SSO.
To connect your favorite Security Incident Response Platform with Cortex 2, you will need to update TheHive to Cerana 0.7 (TheHive 3.0.7) which was released today as well. This version fixes a regression pertaining to case templates introduced by Cerana 0.6 and is the first version to fully support Cortex 2’s API changes and authentication.
To make TheHive 3.0.7 analyze observables at scale through Cortex 2, you have to create an account on Cortex 2 with the read and analyze roles (see the next section) and generate the associated API Key. Next, feed the key in TheHive’s /etc/thehive/application.confas described in the documentationet voilà !
TheHive 3.0.7 remains compatible with Cortex 1 and you can connect it to a mixed set of Cortex 1 and/or Cortex 2 instances with no issues.
Organizations, Analyzers and Rate Limiting
Cortex 2 introduces multi-tenancy through organizations and each organization can have its own set of users, with different roles, its own set of analyzers and, if necessary, rate limits that will prevent analysts from burning quotas.
Multi-tenancy has several interesting use cases. For instance, if you are the CSIRT or CERT of a large multinational organization with several regional teams, you can create an organization for each region within your constituency and enable the analyzers that they may need to use. Let’s assume that you bought a VirusTotal subscription that limits you to 5000 requests per month. You can configure the corresponding analyzers to give each region a fair share of that quota and keeping some requests for your own use.
In case you are a commercial CSIRT or an MSSP, you could do the same for your customers by installing only one Cortex 2 instance and creating an organization for each customer.
By default, Cortex 2 is shipped with the default cortex organization which sole purpose is to create other ones and manage the users within each organization and their associated powers. The cortex organization hosts all users with the superAdmin role and it cannot be used to configure or run analyzers.
As described in the new Quick Start Guide, after installing Cortex 2, updating its database and creating the first user who will have super admin powers, you’ll have to create your first organization and at least one user within that organization with orgAdmin rights.
You can then log out and log in using the orgAdmin account to create further users within that organization, enable and configure analyzers etc. Please note that no analyzer is enabled by default and you need at least v 1.9.0 of the cortex-analyzers repository. To update your set of analyzers to 1.9.0, please run git pull.
Besides the superAdmin and orgAdmin roles, Cortex 2 introduces the read role which allows users to access analyzer reports and read them but not execute analyzers. For that, users need the analyze role (which implies the read role). orgAdmin users can also run analyzers. superAdmin users are limited to the default cortex organization. While they can create organizations and manage users within them, they cannot access analyzer configurations such as confidential API keys or job reports.
Report Persistence and Caching
Cortex 2 relies on Elasticsearch 5.x to store many configuration items but also all the analyzer reports that have been generated. Unlike its predecessor, you won’t lose your existing reports should you need to restart the service or the host it is running on.
Cortex 2 also introduces report caching. By default the cache.job parameter is set to 10 minutes in /etc/cortex/application.conf. That means that if an analysis on a given observable with a defined TLP is requested and that a report has been previously generated in the last 10 minutes, Cortex 2 will serve that report instead of running a new analysis. This feature can help prevent soliciting analyzers, particularly those which require a subscription or have quotas, when there is no need to do so. Please note that this parameter is global to all the analyzers and all the organizations that are configured in the Cortex 2 instance. We do have plans to make it more granular in future versions.
Migrating from Cortex 1
If you are migrating from Cortex 1.x, we recommend that you:
Save the configuration of your analyzers (which ones are enabled and what their configuration items are, such as users/passwords or API keys).
Follow the remaining steps of the Quick Start Guide to enable the analyzers you need and reinject their configuration.
In the upcoming weeks, we will release a new version of Cortex4py in order to make it compatible with Cortex 2, continue the work we started with our MISP Project friends to support MISP attribute enrichment through Cortex 2 (MISP currently only supports enrichment using Cortex 1), and perform a long-overdue overhaul of our documentation.
Feeling Generous? Donate!
As you know, we are a FOSS project and donations are always welcome to make our products even better for the community.
All donations go to Creative Source, the non-profit organization we have created, and we will use them to improve TheHive, Cortex & Hippocampe but also to develop (even better) integrations with other FOSS solutions such as MISP.
Creative Source can also provide so-called professional, entreprise-grade support, help integrating the products, train your analysts before they drain or assist you in specific areas such as developing in-house analyzers for Cortex.
Two months ago, TheHive Chefs announced that Cortex passed the 30 analyzers mark as they added HybridAnalysis, EmergingThreats and Shodan, all three contributed by our continuously growing user community.
It’s 2018 already and to wish you a very happy new DFIR year, Nils and Jérôme got out of their way and reviewed many outstanding pull requests for new analyzers and fixed several bugs. Kudos bees!
The latest release of Cortex-Analyzers, v 1.8.0, contains not one, not two, not even three but ten new analyzers! Isn’t that good omen for a fresh new year fighting cybercrime?
The ten new analyzers, described below, are:
Bluecoat: contributed by our longtime friends from CERT La Poste.
MISP WarningLists: Nils strikes again (watch out Jérôme! the youngster is gonna leave you way behind ;).
Onyphe: contributed by Pierre Baudry and Adrien Barchapt. It comes in five different flavors.
PayloadSecurity: submitted by Emmanuel Torquato. The analyzer comes in two flavors.
Robtex: added by… Nils again! It has three flavors.
SinkDB: guess who developed that one? Wow, impressive! How did you figure it out? Yes, Nils!
Tor Blutmagie: contributed by Marc-André Doll.
Tor Project: also contributed by Marc-André Doll.
We would like to wholeheartedly thank all the individuals and teams listed above for their invaluable contributions. So a big merci for your work!
The Bluecoat analyzer queries the Symantec – previously known as Bluecoat – WebPulse site review API for the currently assigned site category of URLs or domains. The analyzer needs no further configuration. When executed through TheHive, the analyzer produces short and long reports as shown below:
The C1fApp analyzer queries the C1fApp service, an Open Source threat feed aggregation application, using the API for IP addresses, domains and URL.
Before using the analyzer, you need to create an account on the C1fApp website and get the associated API key which you’ll need to provide as a value for the key parameter of the analyzer config section of /etc/cortex/application.conf as shown below. Once you’ve done so, you’ll need to restart Cortex.
key="<insert API key here>"
When launched using TheHive, the analyzer produces short and long reports such as the following:
Censys.io continually monitors every reachable server and device on the Internet, so you can search for them and analyze them in real time. Using the corresponding analyzer, information about a website certificate can be obtained using the associated IP, domain or certificate hash.
In order to use this analyzer, an account at censys.io has to be registered and the API ID and secret need to be added to the Cortex configuration file:
uid="<Your ID here>"
key="<Your secret here>"
Once done, you’ll have to restart Cortex. When ran from TheHive, the analyzer produces short and long reports such as the following:
Details about the ports can be obtained with a click on the specific button.
In order to detect false positives soon enough in the analysis process, our good friends at the MISP Project published their so called warning lists which contain lists of well-known services or indicators.
This analyzer queries observables against the MISP warning lists. Observables can be an IP address, a hash, a domain, a FQDN or a URL.
To iterate through all the warning lists, the repository itself must be available on the Cortex instance:
We highly recommend you create a cron entry or use a similar mechanism to keep the lists fresh. While the default path for the lists is the misp-warninglists subdirectory it can be adjusted in the configuration file:
When called from TheHive, the analyzer produces short and long reports as shown below:
As you can see, The MISP WarningLists analyzer checks if the repository is up-to-date 😉
The Onyphe analyzer leverages Onyphe’s API to query the service, which provides data about the IP address space and the publicly available information in a single, handy location.
The service comes in five flavors:
Onyphe_Forward: retrieves forward DNS lookup information we have for the given IPv4/IPv6 address with history of changes.
Onyphe_Geolocate: retrieves geolocation information for the given IPv4/IPv6 address.
Onyphe_Ports: retrieves synscan information we have for the given IPv4/IPv6 address with history of changes.
Onyphe_Reverse: retrieves reverse DNS lookup information we have for the given IPv4/IPv6 address with history of changes.
Onyphe_Threats: retrieves Onyphe threats information on anIPv4/IPv6 address with associated history.
To use the analyzer, you need to create an account on the Onyphe website. Provide the API key associated with your account as a value for the key parameter and add the lines below to the config section of /etc/cortex/application.conf then restart the cortex service.
key = "<insert API key here>"
When ran from TheHive, the analyzer produces short and long reports such as the following:
The PayloadSecurity analyzer let you submit observables to a on-premises PayloadSecurity instance. To use it, you need to create an account on the PayloadSecurity service. Provide the API/secret pair as values for the key and secretparameters, collect the URL and environmentid of the service, and add the lines below to the config section of /etc/cortex/application.conf. Then restart the cortex service.
When launched through TheHive, the analyzer produces short and long reports such as the following:
When collecting data about IPs, domains and FQDNs, Robtex can be a good source of information. According to their statistics, they logged over 20 billion DNS resource records. The corresponding analyzer comes in three flavors:
Robtex_Forward_PDNS_Query: checks domains/FQDNs using the Robtex Passive DNS API
Robtex_IP_Query: checks IPs using the Robtex IP API
Robtex_Reverse_PDNS_Query: checks IPs using the Robtex reverse Passive DNS API
The analyzer uses the free Robtex API which needs no subsequent configuration. However, the free API limits the rate and amount of returned data.
When executed using TheHive, the analyzer produces short and long reports such as the following:
SinkDB is a private service provided by abuse.ch which collects sinkholed IPs. Access to the service is allowed to trusted partners only. If you think you qualify, you can request an access using the form available on the SinkDB website. This is most likely only granted to certain CSIRTs and CERTs and not to individuals.
Provide the API key associated with your account as a value for the key parameter and add the lines below to the config section of /etc/cortex/application.conf then restart the cortex service.
key="<insert API key here>"
When ran from TheHive, the analyzer produces short and long reports such as the following:
Tor Blutmagie analyzer extracts data from torstatus.blutmagie.de and checks if an observable is linked to a Tor node. The observable can be an IP address, a FQDN or a domain.
In order to check if an IP, domain or FQDN is a Tor exit node, this analyzer queries the Tor status service at Blutmagie.de. The analyzer uses a caching mechanism in order to save some time when doing multiple queries, so the configuration includes parameters for the cache directory and the caching duration.
Provide the lines below to the config section of /etc/cortex/application.conf then restart the cortex service.
When ran from TheHive, the analyzer produces short and long reports such as the following:
Tor Project analyzer has also been contributed by Marc-André Doll. As the above analyzer, this one checks if an observable is a Tor exit node. This time, however, the source of information is the official Tor network status which can be queried for IP addresses only.
The accepts another parameter, ttl, which is the threshold in seconds for exit nodes before they get discarded. Provide the lines below to the config section of /etc/cortex/application.conf then restart the cortex service.
George Abitbol* doesn’t feel well. Christmas is approaching at a fast pace and the gift he ordered days ago for his girlfriend didn’t find its way to his mailbox yet. He checked it out three times today and save for some spam catalogues on how to take care of his handsome silhouette, nothing resembling a gift showed up.
He tried to call the French parcel service to know the whereabouts of the luxurious, limited version of the organic sweet potato chips his lovely Jacqueline* likes so much, which sells for four times the regular price (to bear the cost of the enhanced packaging, certainly), but he couldn’t get hold of a living soul all day long. When he placed the order, the delivery was supposed to be lightning fast. It turned out to be a false promise.
With a sinking heart, he climbs back the stairs leading to his apartment, fetches his laptop and sits on his club chair. With his headphones on, immersed in the wonderful jazz of Christian Scott, he wanders randomly through online shopping sites trying to make out his mind on what other presents he could get for his dear Jacqueline, in time for Christmas.
In the middle of the track called Encryption, featuring the uncanny Elena Pinderhugues on flute, a Twitter notification resonates in his ears. He checks it out and learn that TheHive Chefs, as true and elegant gentlemen, have published a new training VM for Cerana 0.3 (a.k.a. TheHive 3.0.3), including Cortex 1.1.4 and the latest set of Cortex-Analyzers.
George loves bees in all shapes and forms, including digital ones so he swiftly downloads the new VM and as the cautious person he is, he verifies the file’s SHA256 hash: 86a87b70627e8db672c57cb57821461f2564ae9b8087cc22fdd1e7a599c16aed. Wonderful! Everything checks out beautifully. He then imports the file in his VM software, starts the virtual machine and logs in as thehive then types in thehive1234 when asked for the password.
He thoroughly reads the documentation to configure various analyzers and integrate his favourite Security Incident Response Platform with MISP. A few minutes later, his VM is ready for prime time and he starts playing with the new multi-source dashboards and interacting with fellow analysts on Gitter.
And he totally forgets about Jacqueline’s gift.
(*) Any resemblance to real and actual names is purely coincidental.
Correction: Dec 23, 2017 An earlier version of this post was referring to a previous training VM that included Cerana 0.2, a version affected by a privilege escalation vulnerability which was corrected in Cerana 0.3. Some typos were corrected as well.